Exhaust Air to Warm water Heat Pump - malt kilning

[MannyCalavera]

Hi folks,

Investigating the feasibility of installing a heat pump at a maltings (malt whisky production). Very new to heat pumps, to get an overview of if its possible what are the key things to calculate?

Exhaust air is 100%RH, 35 degC. Large volume.
Want to heat warm water (45 degC) to as high as possible.

What are the key things i need to calculate before knowing whether this is economically viable? Any help appreciated.

Drew

[MannyCalavera]

Perhaps i should elaborate - i want to take as much heat out of the humid air exhaust as possible, say roughly 1 MW.

I am thinking of butane as a refrigerant since its boiling point is higher than others. So if i were to compress it to 9bar, it would boil at 75 degC. After compression it would be around 90degC, so the warm water i want to heat would have a bit of sensible and alot of latent energy to recover.

Unsure on compressor types / limitations / butane as a refrigerant.

[mad fridgie]

What temperature of water do want, do want a large voulme of water at lower temps or smaller volumes at higher temps.
What is your water entering temperature.
Your refrigeration requirements are the last thing you look at you Must study the application and process.
I can design and supply in conjuction with my UK partners.

[MannyCalavera]

Fridgie,

I have identified another heat source: A stream of warm water at 40-45 degC (30kg/s). Cooling this is also of benefit to the process.

I want to heat a loop of glycol (previously i said warm water, but its glycol) from around 40degC to as high as possible. The flowrate of glycol is roughly 12kg/s.

This glycol acts to pre-heat ambient air for kilning, and so the hotter the glycol through the radiators the better.

So i think heat source = exhaust air, or warm 40-45deg water

Heat sink = glycol loop already at around 40 degC.

Any thoughts on feasibility?

Regards.

[pwned]

jp.sanyo dot com/comp-unit/english/co2/ecocute.html

Not exactly designed for your purpose, but says up to 90C water temp.

It uses dual stage compressor with CO2 :o

[mad fridgie]

Fridgie,

I have identified another heat source: A stream of warm water at 40-45 degC (30kg/s). Cooling this is also of benefit to the process.

I want to heat a loop of glycol (previously i said warm water, but its glycol) from around 40degC to as high as possible. The flowrate of glycol is roughly 12kg/s.

This glycol acts to pre-heat ambient air for kilning, and so the hotter the glycol through the radiators the better.

So i think heat source = exhaust air, or warm 40-45deg water

Heat sink = glycol loop already at around 40 degC.

Any thoughts on feasibility?

Regards.
If we look at your water source, we have lets say for easy a good 2 Mw, that then would heat your glycol to above 80C (actual temp depends upon concentration)
Should be no problem having a COP of over 4. (COP could be over 5 process specific) for heating only, may increase if you have the benefit of cooling

[MannyCalavera]

Thanks for link pwned.

Fridgie, cooling the warm water stream is also a big benefit (currently cooling towers bring it down to ~25 deg C). Two birds with one stone really.

I appreciate there is a large amount of energy in that warm water - in order to estimate the economics I need to work out the cost of compression.

Clearly I will want to compress my refrigerant to a pressure such that it's boiling point is above that of the warm water, and so i can extract the latent heat. Butane is looking like a good one (for the sake of rough calculations). Finding rules of thumb / design procedure for this is proving tricky. I have not done much study on refrigerant compression and am currently using equations given in engineering textbooks to calculate work of compression.

Industrially speaking, a COP of 4 is quite high?

I appreciate the replies - thanks.

[mad fridgie]

Thanks for link pwned.

Fridgie, cooling the warm water stream is also a big benefit (currently cooling towers bring it down to ~25 deg C). Two birds with one stone really.

I appreciate there is a large amount of energy in that warm water - in order to estimate the economics I need to work out the cost of compression.

Clearly I will want to compress my refrigerant to a pressure such that it's boiling point is above that of the warm water, and so i can extract the latent heat. Butane is looking like a good one (for the sake of rough calculations). Finding rules of thumb / design procedure for this is proving tricky. I have not done much study on refrigerant compression and am currently using equations given in engineering textbooks to calculate work of compression.

Industrially speaking, a COP of 4 is quite high?

I appreciate the replies - thanks.
I have just completed some further calculations, so 5 COP is very possible glycol heated to 83C, water cooled from 45 to 27C.
I am not using flammable refrigerants (a big issue in NZ at the moment)
Cost of a system, that is some what more difficult, as there are so many factors that need to be considered. But may be a s a pure budget 150 quid/ KW (300,000 pound) for this application

[MannyCalavera]

I have just completed some further calculations, so 5 COP is very possible glycol heated to 83C, water cooled from 45 to 27C.
I am not using flammable refrigerants (a big issue in NZ at the moment)
Cost of a system, that is some what more difficult, as there are so many factors that need to be considered. But may be a s a pure budget 150 quid/ KW (300,000 pound) for this application


Fridgie, would you mind explaining how you went about your calcs?

Obviously

Q = m.Cp.dT = (m.Hvap + m.Cp.dT)

in terms of heating glycol from latent and sensible heat of superheated -to-condensed-refrigerant.

Q = m.Cp.dT = m.Hvap

in terms of cooling the warm water and evaporating the refrigerant.

My lack of experience is in these areas:

- choosing a compression pressure
- whether multi or single stage compressor/s
- which refrigerants are most efficient
- optimising the trade off between Work In to compressor, to heat imparted to refrigerant

Fridgie, any help / explanation on any of the above? I am not sure what Scotland's stance on flammable refrigerants is.

Regards.

[mad fridgie]

Hi Manny, Firstly I have estimated a number of the process variables, so figures are just that estimated. (plus it is friday night here) But factual on my estimations.
How I actually achieve the very high COPs, is moving into the commercially sensative area. (I hope you understand)
Is this a practical application or just a theoretical model. The two should and are to be approached from different angles. (Theory should give you much higher COPs, but has no relevence to the practical aspects of refrigeration and very little to do with cost)
If it is a practical application, your very question to the client is was is the limits of the process (min/max), or is it steady state, what are the consequences for changes that may occur.
What is your role in this application.
If you are being used as a consultant,
Leave the specifics to the manufacture, but ensure that you stipulate A start point and a end point, requiring a performance guarantee.

[MannyCalavera]

Hi Manny, Firstly I have estimated a number of the process variables, so figures are just that estimated. (plus it is friday night here) But factual on my estimations.
How I actually achieve the very high COPs, is moving into the commercially sensative area. (I hope you understand)
Is this a practical application or just a theoretical model. The two should and are to be approached from different angles. (Theory should give you much higher COPs, but has no relevence to the practical aspects of refrigeration and very little to do with cost)
If it is a practical application, your very question to the client is was is the limits of the process (min/max), or is it steady state, what are the consequences for changes that may occur.
What is your role in this application.
If you are being used as a consultant,
Leave the specifics to the manufacture, but ensure that you stipulate A start point and a end point, requiring a performance guarantee.


Hi Fridgie, cheers for getting back. And i understand your want for not giving away calculations etc!

I am a near-graduated chemical engineer doing a half year placement in malt distilling, scotland. This would very much be a practical application - there is a huge drive currently on energy minimisation and reduced emissions. Producing malt is an extremely energy intensive process, with millions spent annually on heavy fuel oil.

My role is to investigate the feasibility of installing this heat pump. From what I have worked out, and correspondance with yourself and a few others it seems, theoretically, like a good idea. Economically? That's where I don't have experience.

It's only this week I have been looking into the heat pump possiblity, but already from a theoretical POV it seems sound. Again, no point in this design unless running costs of the loop are < money saved from reduced heavy fuel oil burning.

There is limited technical help on refrigeration / practical implications here.

The warm water - this is excess from a distillery - and normally will vary only slightly in flowrate and temperature (we are talking +/- a couple of deg C). Warm water is 24/7 (thats the aim) and the requirement for heating glycol likewise.

I presume my next step would be to consult companies that install / manufacture such systems and ask for quotes & performances, etc once i have fairly concrete "I have this much energy available and I want this as high as possible" figures.

Any further feedback or thoughts appreciated. A COP of 4-5 certainly sounds worthwhile.

Thanks for your replies - enjoy your weekend.

[mad fridgie]

Hi Fridgie, cheers for getting back. And i understand your want for not giving away calculations etc!

I am a near-graduated chemical engineer doing a half year placement in malt distilling, scotland. This would very much be a practical application - there is a huge drive currently on energy minimisation and reduced emissions. Producing malt is an extremely energy intensive process, with millions spent annually on heavy fuel oil.

My role is to investigate the feasibility of installing this heat pump. From what I have worked out, and correspondance with yourself and a few others it seems, theoretically, like a good idea. Economically? That's where I don't have experience.

It's only this week I have been looking into the heat pump possiblity, but already from a theoretical POV it seems sound. Again, no point in this design unless running costs of the loop are < money saved from reduced heavy fuel oil burning.

There is limited technical help on refrigeration / practical implications here.

The warm water - this is excess from a distillery - and normally will vary only slightly in flowrate and temperature (we are talking +/- a couple of deg C). Warm water is 24/7 (thats the aim) and the requirement for heating glycol likewise.

I presume my next step would be to consult companies that install / manufacture such systems and ask for quotes & performances, etc once i have fairly concrete "I have this much energy available and I want this as high as possible" figures.

Any further feedback or thoughts appreciated. A COP of 4-5 certainly sounds worthwhile.

Thanks for your replies - enjoy your weekend.
Firstly congratulations on choosing chemical engineering, many of my freinds choose this path and are now very successful.
I am aware for the drive for efficiency in the UK, I will be over there at the end of the month trialing my technology (already proven in NZ) 80C water with COPs close to 10.
What you have to be careful of is limiting the commercial feasabilty, with the need for technical excellence.
Going down the application, system then is fairly steady, which is good. (keeps the price down)
Then you need to look at serviceability, are you better to have say the load spread across 3 independent machines (allows for break downs, service requirements etc.)
How much is your heavy fuel oil, per "usable" KWhr.
I have asked it is this way so that it can be directly compared to electrical energy (heat pump)
Hoping to sample some of your scottish products this weekend
cheers
Mad

[MannyCalavera]

I will have a look into those details and work through them on Monday - 3 seperatate units - that is the kind of thing I would not have thought of (you see my lack of experience). All this is good food for thought for me, cheers.

I am slowly getting aquainted with the single malts of Scotland - i find it incredible the miniscule tweaks in process conditions that are used to keep the character on target. I am planning on having a few drams of single malt also - good call!


Many thanks.

[mad fridgie]

Just a quick question, You stated that you had air at 35C and 100% and approx 1MW, is this the air exausted from the malting drying process. (If so I presume it has already gone through an air to air heat exchanger, or similar device)
If so then your load is going to be very similar (plus losses).
If you were to have more energy available (from the water loop), would you look at changing your malting process (increased temp, greater fresh air makeup)
What i suggest you do is an energy mass balance on the glycol heat circuit.

[MannyCalavera]

Just a quick question, You stated that you had air at 35C and 100% and approx 1MW, is this the air exausted from the malting drying process. (If so I presume it has already gone through an air to air heat exchanger, or similar device)
If so then your load is going to be very similar (plus losses).
If you were to have more energy available (from the water loop), would you look at changing your malting process (increased temp, greater fresh air makeup)
What i suggest you do is an energy mass balance on the glycol heat circuit.

Yes the air is exhaust air from drying the 45% moisture malt. My understanding is that you have pairs of kilns. One will be operating pre-break point (air-off is close to 100% RH and around 30degC) I don't think they use an air-to-air heat exchanger with this exhaust - it is just taken to atmosphere. After ~ 12 hours RH begins to drop below 100%, and it is recycled for use in its neighbour paired kiln as it still has useful heat. 1 MW was a total estimate - i don't know how much energy you can extract from a high-enthalpy stream of air.

In terms of the malting sequence, takes around 18 hours for a batch (typically 30 tonnes). The temperatures and fan speeds are not open to change - they are set for periods based on the character of the malt they desire. So it's not just a case of hotter air the better. I think it starts at 65 deg C, stays on 75 deg C for a long while, then gets ramped up to 85 degC towards the end of the kilning (something to that affect, anyway).

Next week the maltings is operation again (been down for maintenance for 3 weeks) I am going to collect data so an accurate as possible energy balance can be done - and I agree this will be a good starting point to evaluate the energy available.

Some maltings DO pre (pre)heat incoming air via air-to-air glass tube exhaust exchanger. My maltings of interest doesn't - I will have to ask why.

Cheers

[MannyCalavera]

Sorry - "break point" is the point at which the air-off the malt bed stops being fully saturated.

Typically 11-12 hours from the start of a kilning of a batch.

[mad fridgie]

So would I be correct in saying that the air passing through the malt is single pass, entering at desired process temp, what leaves the process is down to the specific properties.
If we presume, that the actual drying takes 12hours, then you will need roughly 760Kw for vapourization (12 hour not 18, as the drying process is not linear the reason for the higher temps at the end) You also need to including heating of the room, the product, the entering air and any thermal losses (your 1Mw makes sense)
But, you will have large variables in what is entering and existing the heat pump. (This is all to with how systems are presently controlled, valving variable air volumes and the like)
From what you have indicated I would use the dist. water as a primary heat source (you can use the heat pump to start the process)
Again focus on the process(s), once you have this, you can then look at the heat pump requirements, min and max loads, at varing min and max temps.
It is the unaccounted variables that F***S Up refrigeration plants (as design engineers, if we are not given these, then we do not design for them to happen)
Your actual process (s) will be well documented, and are likely to be with a PLC program, look for the extremes

[MannyCalavera]

I am going to find the PFDs on Monday so I can understand the flow of air through the kilns - when it is fresh single pass, or recycled, or a combination of both. The sequence is quite complex. I will pass the info on to yourself when I have it on monday.

I had one question - assuming the warm water was used as the heat source for the heat pump, would there be any configuration that could make use of the saturated exhaust air also?

One advantage of the warm water is it's fairly steady properties - as you say, the exhaust air will be changing more considerably.

Regards

[Peter_1]

Just read very fast this interesting thread. It was a little bit too long reading it thoroughly before going to sleep and especially all the calculations Mad Fridgie already made. I will re-read it tomorrow.

I will give you my view on this and perhaps Mad Fridgie which I respect very much gave you already similar answers or solutions.

Going for a flammable gas in such a big quantities is extremely dangerous. There are better solutions than butane.

If you want high COP's, high end temperatures and certainly in these big ranges of kW's, then NH3 is for all these reasons the gas you have to choose.

You mentioned somewhere that you want to condense higher than the water temperature (correct me if I'm wrong). You will condense always higher than the water temperature, otherwise you can't condense.

But you forget something extremely important, your discharge temperatures (end compression temperatures I call it) - especially with NH3 - will be extremely higher than you condensing temperature. This is very useful for your application.

You could use a PHE to use the compression superheat as a boost arm-up (sorry for my bad English) and finally a second PHE to condense completely at lower temperatures to preheat the water.

Or you can circulate water through different coils where you retract the heat with a central PHE on NH3/water.
Makes it more safer in case of a leak in the kiln.
It' will also easier to control everything.

Giving the act that you have waste heat with a RH of 100% makes this extremely interesting for a reclaim project.
Strange nobody didn't thought sooner on reclaiming this waste-heat.

Giving your rather young age, I have to admit that you go really far in analyzing all this. In French they say chapeau or in English my hat goes off for you:cool:

I think this Whisky company can perhaps benefit from funds the Scottish State will give for such a big reclaim project.

30 years ago, just after leaving school, I worked some years as a mechanic in a company that made special machines for malt kilns (they made one for an African country but I forgot which one), the drying ovens (+/- 20 x 5 x 2 m), the big machines running over and through the malt, rotators which ploughed (correct expression) each day continuously through the malt to bring the malt (grain) from the bottom to the top and vice-versa. They where just like big auger drills and we called those wenders The air was blown from the bottom (perforated grilles) with huge turbines. Stinked extremely there.

[MannyCalavera]

Pete,

Many thanks for taking the time to read the thread and give your thoughts - very much appreciate this.

Really helpful points you have raised there.

"You will condense always higher than the water temperature, otherwise you can't condense."

I completely understand this - from my (shakey) calculations, is it ever an issue whereby your heat of compression (sensible) will heat the SINK temperature to a value higher than at which it will condense?

E.g. glycol at 45 degC. Say heat of vapour after compression = 100 deg C and saturation temp is 65 degC. Is there the possibility this 35 degC sensible heat could raise the water temperature to above 65 degc? (if you know what i mean...).

I am not familiar with the lingo PHE, could you explain?

I agree, it is odd no one has looked into this heat pump idea before, it seems to make a lot of sense, especially given the benefit of cooling the warm water.

I will properly read and go through what yourself and Fridgie have so kindly discussed on Monday.

This is all great guidance. The kilns you describe in your earlier job sounds pretty damn similar to the setup at the maltings I am at - the smells arn't too unpleasant where I am at however! Not as sweet as the distilleries mind.

It is saturday night and I am due a beer - this week I will hopefully be able to work with what you guys have suggested.

Again, many thanks indeed.

[MannyCalavera]

Pre Heat Exchanger? (I blame the university...)

[mad fridgie]

If I was your age, I would have already been on the Beer and chasing the fairer sex. (now that is one complex process "woman")
The actual refrigerant choice, is very much dependent upon, properties, availablity (components), service expertise, longevity and cost.
With any heat recovery, you must ensure that you have a sufficent use for the what is recovered at that temperature.
There are massive amounts of low to medium grade energy streams, that could be Boosted by heat pump technology, but the output temps are still generally to low for many practical processes (most need higher temp steam) There are a number of theorectrical ways to achieve such results, it is the practical and economic restraints that is slowing down research.
One point i will disagree with it that the heated output water temperature, need no below the refrigerant condensing temperature. Quite the opposite. (Peter touched on this area) How ever the inlet has to be lower than the condensing temp.
I would suggest the very first thing you do on Monday, is to calculate the "usable" KWhr cost of the heavy desiel.
"PHE" Plate Heat Exchanger.

[mad fridgie]

One other thing i would do, is go to your companies descion makers (CEO, MD) ask them what investment and return would ensure that such a process, would be basically guaranteed to be undertaken.
This then becomes a "goal"

[mad fridgie]

It is Sunday afternoon, it is cold and miserable outside, nothing worth watching on the box, so I thought i would have a quick play, with all the info you have given and lots and lots of presumptions.
First you glycol flow and figures must relate to both kilns, as the drying load is some what close to half your glycol duty. (remember drying is never steady state when cell structure are being dried)
If we control the refrigeration compression ratios to match the heating requirement (unlike a boiler where the output (temp) is normally fixed). This way we can optomise overall process efficiency.
So down to some number,
Total energy for the twin process.
33.2MWhrs on heavy diesel, (used at the process) DUTY
If the duty was then undertaken by the heat pump with a floating set point ( I have presumed that the glycol to air heat exchangers are able to heat the air to within 5C of approach glycol temp possible but unlikely)
Then the heat pumps would consume
7.5MWhrs of electricity.
If we then attribute some costs to these energy sources
heavy diesel= 3 pence per usable Kwhr
electricity= 7 pence per Kwhr
then a saving of 470 pounds per day or what looks better is an annual saving 170,000 pound. (excludes benefits of cooling)
I believe you could undertake an installed process for under 400,000 pound (not the absolute best, as this could trend towards a million)

[Peter_1]

E.g. glycol at 45 degC. Say heat of vapour after compression = 100 deg C and saturation temp is 65 degC. Is there the possibility this 35 degC sensible heat could raise the water temperature to above 65 degc? (if you know what i mean...).
Sure, it can go almost that high as your end compression temperatures, 100°C in your example.
We have systems running on R404a on a condensing temperature of 40 to 45°C where the water is heated to 75°C.
But the amount of heat water with superheated refrigerant is rather small compared to the heat available to condense the saturated refrigerant.



I am not familiar with the lingo PHE, could you explain? Plate Heat Exchanger.
Lingo..;) new word for me.


The kilns you describe in your earlier job sounds pretty damn similar to the setup at the maltings I am at - the smells arn't too unpleasant where I am at however! Not as sweet as the distilleries mind.
I also remember the very big humidity and the electric shocks I got when I inserted a welding rod in my torch.

Download once the free software Coolpack from the Danish University and play a little with the log p/h and notice end compression temperatures with different gases. Do certainly once a calculation wit NH3.

Did a quick simulation with Coolpack for you, NH3, TE = 10°C, TC = 45°C, SH = 5K gives you +/- 115°C discharge temperature and a COP of 4.6.
Increasing TE to 20°C raises COP to 6.9 (!), end compression temperatures to almost 100°C (with 8K SH)

Only R134a will coming in the same region of COP but with much lower end compression temperatures (+/- 60°C for the last example)

An old invoice from the malt company I worked for. Those old days!!! Where the factory was is now a big office and apartment complex. http://blogimages.seniorennet.be/bruggeherbekeken/61-45715693d9bc2f452c6cf28fe5ad2e00.jpg

[mad fridgie]

I should of but down the glycol supply temps. (entering at 40C, low level glycol concentration)
70C full flow (12L/S), Duty 1510Kw
80C full flow, Duty 2010Kw
90C 66% of full flow, Duty 1680Kw (less actual drying is undertaken)

[MannyCalavera]

Good morning! Hope your weather is better than the Scottish summer.

Regarding NH3 - not so popular here - mainly just due to its toxicity. Any substitutes in terms of performance?

If I could summarise the key points so far briefly: Forgive me for going over it again, it's just to make sure it's easy to follow:-

2 kilns paired. Each has a steam radiator heater, and a (shared) glycol pre-heater (40:60 glycol:water). Stream raised by HFO burner (~ 85,000 ltrs/week).

1. Fairly constant exhaust of saturated air all the time (Site is operational later in week, I can get raw data then) at high flowrate. No recovery on this at all.

2. Fairly constant flowrate & temperature warm water stream at high flowrate.

Currently the glycol is heated with the warm water (which takes it from 52 - > 45 deg C).

My idea would be to use the heat loop and PHE to heat this to much higher: 60-80 degC. This would mean that for a kiln in early stages of drying, there may be no requirement for steam radiators. And for one in the final stages, you could "top up" the air temperature with the steam radiators.

I.e. - have the glycol very hot all the time as there is constant kilning and need for hot air.

Other option would be to have the very high temp vapour through a radiator and condense, however I am not sure if air has the heat capacity to suck out that much energy at such large volumetric flowrates.

It is clear that taking warm water from ~40 deg C to ~25 degC with 100m3/hr has much energy (around 1.5MW). Cooling this may remove the need for (current) cooling towers and (proposed) sea-water cooling. This would save alot of kWh through pumping costs.

The saturated air also has a fair enthalpy (~ 100kJ/kg). While it's great to "recover" this heat, there are no associated cost savings.

I can't see how both could be used (at least not for the same purpose similtaneously).

I am looking up the CarbonTrust to see what insentives ($$$) are available for such recovery systems.

A 3-year payback is considered worth-while, but it depends on each project. However the want to become "carbon neutral" could be a fairly big factor in decisions that effect HF oil consumption.

ELEC: £0.08 per kWh
HFO : £0.04 per kWh equivalent.

I gather elec. is fairly constant, HFO has risen slightly in recent times (more insentive).

I am going to do an energy balance to calculate the effect of, say, 1 degC glycol extra heat, and how much this will save in steam, and then work it back and calculate the cost from saving X litres of HFO.

Pete - TE, TC, SH? Is that evaporation temp, critical temp and ... ?

Fridgie those numbers:
70C full flow (12L/S), Duty 1510Kw
80C full flow, Duty 2010Kw
seem promising - it's that degree of heating that is going to be very useful here.


Any thoughts / comments please let me know.

Kind Regards.

[mad fridgie]

Is the boiler and HFO solely used for this process (one is hoping not) 980,000kwhrs a week (I have converted this into standard numbers for others who may be reading)

[MannyCalavera]

Fridgie,

Two HFO boilers, one used all the time, the other used rarely when one isn't enough.

The steam raised is solely for use in the kiln radiators - a small amount is used to keep the HFO warm.

980,000kWh - is this your calculated savings?

[MannyCalavera]

NB: chlorinated hydrocarbons are not favoured - ammonia would be acceptable

[mad fridgie]

I better show my calcs
85000L HFO, density 949Kg/m3 (furnace oil)
80665Kg at (*) 43.8 MJ/Kg
3553127MJ * 1000 to Kj
355312000/3600 to kwhrs
981424Kwhrs
Total energy used per week by the malting process.
We then need to know the efficiency of the boiler and what steam pressure/temp (wet or dry)
If we assumed that one pair of processes happen per day,
Then 140,202Kwhrs/day
Lets say 80% boiler efficiency
112,161Kwhrs for the process per day
basic process load
60,000Kg at 45% water
So we want to vapourise (lets say all)
27000kgs of water (lets say for ease latent heat 2430KJ/Kg)
Duty6561000KJ/ 3600
18,225Kwhrs.

Are you sure about your usage HFO usage; is it 8500 Litres per week?, if you are correct then you have some serious losses, this being the case then sorting this out should be the first call first

[mad fridgie]

NB: chlorinated hydrocarbons are not favoured - ammonia would be acceptable
I am looking at R134a, not a chlorinated hydrocarbon.

[MannyCalavera]

I will go through those calcs this afternoon so's i follow them. I have found more data:

Weekly malt produced ~ 1770 te
HFO per week ~ 80000 l
This is for 9 kilns (i see why you thought there were losses). 8 indirect heated (steam) 1 direct fired.

Works out at around 45litres oil / te malt (5%wt water).

Cost per year HFO ~ £2m

[Peter_1]

I think NH3 is as popular in your country as in other countries. It's only toxic when you breath it and that's not why a refrigeration system is built.

Don't forget that NH3 has a GWP = 0 (!!) and in the very, very near future, all the GWP refrigerants with a GWP bigger than 2000 (this will become probably the number) will be banned or highly taxed. The environment is becoming extremely important with the existing and the new coming EU regulations.

Every refrigerant wit chlorine in it is no longer allowed in the EU.

Forget about the toxicity, R404a is perhaps more harmful than NH3 because you don't smell it when you breath it. Most large refrigeration systems runs all on NH3.

And the NH3 doesn't have to run in the plant itself, you can design int in such a way that it stays only in the machine-room.

And I'm not a NH3 installer, we just service some older NH3 plants.:p

You have to choose a refrigerant for your application which gives you the highest COP and the highest discharge temperatures, whatever that refrigerant may be. You then have to look if it's allowed in your country and/or application.
NH3 will give you for this large capacities the biggest advantages seen COP and high end compression temperatures.
Or 134a.

Compare also once the price for NH3 and R134a refrigerant. You will need a lot of refrigerant to run this system. Limit therefore the refrigerant to the machine-room and if you're afraid about toxicity, install NH3 sniffers. But you have the best sniffer with you: you nose.
Forget the danger about toxicity of NH3. This belongs more to the world of tales and is said by those who haven't enough experience with this refrigerant and are therefore afraid of it.

Play a little with Coolpack and you will be convinced yourself.

TE = temp evaporation
TC= temp condensation
SH = superheat
SC = subcool

Can't you preheat first the water not only with a watercoil/watercoil setup? Haven't read very accurate all the posts of this thread about real/actual temperatures.

[Peter_1]

I...
Are you sure about your usage HFO usage; is it 8500 Litres per week?, if you are correct then you have some serious losses, this being the case then sorting this out should be the first call first
Mad Fridgie, haven't you forgotten the performance ratio (effectiveness ??) of the burner?
And is the burner not used to heat other sources?

This is a very interesting thread. I will ask some NH3 experienced guys joins this one. They perhaps can help.

[mad fridgie]

I will go through those calcs this afternoon so's i follow them. I have found more data:

Weekly malt produced ~ 1770 te
HFO per week ~ 80000 l
This is for 9 kilns (i see why you thought there were losses). 8 indirect heated (steam) 1 direct fired.

Works out at around 45litres oil / te malt (5%wt water).

Cost per year HFO ~ £2m
That makes a bit more sense, what is "te"?

[MannyCalavera]

Thanks for the info on NH3 Pete, I contacted our environmental dept. and NH3 is not un-favoured.

As you say given the strong emphasis on low GWP and its high performance, it seems like the most logical choice. Far better than hydrocarbons at least.

"Can't you preheat first the water not only with a watercoil/watercoil setup? " Could you explain the two streams you were thinking of here?

Mad, 45 litres/te is quite efficient as maltings go I believe (it'll be less when this heat pump gets going ;))

Te = tonne.

[mad fridgie]

Mad Fridgie, haven't you forgotten the performance ratio (effectiveness ??) of the burner?
And is the burner not used to heat other sources?

This is a very interesting thread. I will ask some NH3 experienced guys joins this one. They perhaps can help.
Hi Peter, i allowed for 80% efficiency for the boiler (it should be a lot higher than this), the HFO heats 9 kilns, not 2 as thought.
This is my "field" before we even contemplate what type of system to use, we must get a handle on the process and its variables.
what we now need to know is the diversity of the 9 kilns, and how are they interconnected.
If we look at the total daily heat requirement, and what is available, form the water and air stream, then we can look at how best to apply it.

[mad fridgie]

Thanks for the info on NH3 Pete, I contacted our environmental dept. and NH3 is not un-favoured.

As you say given the strong emphasis on low GWP and its high performance, it seems like the most logical choice. Far better than hydrocarbons at least.

"Can't you preheat first the water not only with a watercoil/watercoil setup? " Could you explain the two streams you were thinking of here?

Mad, 45 litres/te is quite efficient as maltings go I believe (it'll be less when this heat pump gets going ;))

Te = tonne.
approx 50% efficient not to bad for drying!!!!
Thanks for that

[MannyCalavera]

I am going to have to give this some thought Mad - l'm ignoring kiln 9 for now.

3&4 = a pair = 60 te / batch
5&6, 7&8 = a pair = 30 te / batch

Mad I am speaking total dross - there are seven kilns - (one and two have been redundant for years). So it is 3-9.

So there is quite a load - I firstly need to get the data on WHEN the saturated air is available, how constant it is across the 7 kilns, and also the demand. As I have said, I am assuming un-varying warm water supply.

As I say, later in this week I get to see the sequences of the kilning.

83% efficiency i believe for the boiler. 90% efficiency in the steam radiators.

On a related note, what kind of compressors are usually used for refrigeration?

Cheers guys.

[mad fridgie]

I am going to have to give this some thought Mad - l'm ignoring kiln 9 for now.

3&4 = a pair = 60 te / batch
5&6, 7&8 = a pair = 30 te / batch

Mad I am speaking total dross - there are seven kilns - (one and two have been redundant for years). So it is 3-9.

So there is quite a load - I firstly need to get the data on WHEN the saturated air is available, how constant it is across the 7 kilns, and also the demand. As I have said, I am assuming un-varying warm water supply.

As I say, later in this week I get to see the sequences of the kilning.

83% efficiency i believe for the boiler. 90% efficiency in the steam radiators.

On a related note, what kind of compressors are usually used for refrigeration?

Cheers guys.
As you can gather, there are many options open on the refrigeration side, what refrigerant and what type of compressor and the number of stages, really depends upon evaporating and condensing pressures, which directly relate where your energy stream is coming from (suction pressures) and what temperatures you will require or what is the best use of the absorbed energy and what levels of turn down and limits on extremes.
For this reason, is why total understanding of the process becomes paramount. I have seen many systems that have been well designed for steady state conditions, but become a total balls up when changes happen in the process variables.

[MannyCalavera]

Say a system does change, say the warm water flowrate / temperature decreases, what are the knock-on effects on the refrigeration system?

Could you just list a few things that would go tits up? I.e. would the compressors die if there was liquid in there that hadnt evaporated?

My next step now is to do some quantifying of availablilty of heat sources, as you suggest. Hopefully within a week I can give you some concrete flowrates, etc, and I myself can do some calculations on the energy saved with hypothetical glycol temperature increase.

Regards

[mad fridgie]

Say a system does change, say the warm water flowrate / temperature decreases, what are the knock-on effects on the refrigeration system?

Could you just list a few things that would go tits up? I.e. would the compressors die if there was liquid in there that hadnt evaporated?

My next step now is to do some quantifying of availablilty of heat sources, as you suggest. Hopefully within a week I can give you some concrete flowrates, etc, and I myself can do some calculations on the energy saved with hypothetical glycol temperature increase.

Regards
There are no real problems, if you know the extremes, you design around potential problems, but this does have a factor on the capital cost.
Basically you control the load by many different methods, but when unloaded care has to be taken with other factors, it just a question of balance, protection and retaining efficiency. many of the threads on this forum relate to poor equilbrium.
Its time to go for kip.
look forword to future info

[MannyCalavera]

Alright Fridgie cheers for your help thus far.

I will keep checking the forums incase there are further posts, but later in the week I will get in touch with some concrete numbers.

I feel I am slowly becoming more aware of the wider picture of heat pumps which is most excellent.

[MannyCalavera]

"This is a very interesting thread. I will ask some NH3 experienced guys joins this one. They perhaps can help."

Just saw this Pete - would be very much appreciated.

[mad fridgie]

I thought I would help with the correct thought process.

1 What is the problem (excessive fuel use)
2 Analize the problem (In the case the malting process, this needs to be indepth)
3 Set the criteria for the selection of the possible solutions. This is as important as the solution its self!

[mad fridgie]

just some theoretical interesting facts (not a great deal to do with the practical application and installation)
it is possible to reject more heat out of the heat pumps than the process requires, which is typical for drying processes. (heat absorbed from both the dist. water and the kiln air, Direct)
If the process could be completed with 75C glycol, then COP would be approx 4, which indeed would reduce the running cost by 50%, or a million quid or year.
If we look at any drying process there are 4 main loads
1;Thermal losses
2;Heating of mass (product and building)
3;Water vapourizing
4;Heating of make up air. (this is required to carry the moisture away, warmer air has a higher water vapour carry capacity)

When a product is drying, the product and the "air passing over" drops in temp, whilst pickup moisture, increaseing the RH by the two changes, how ever the energy in the air remains the same (exclude any losses) This is expelled
So it would seem that we should just reclaim all the energy from the process, and introduce it into the in coming air (looks like the simplest of solutions) BUT dry air does not have good energy carry abilities, therefore the air would need to heated to a very high temperature to reach equalbium (balance), this then becomes the achillies heel of a refrigeration heat pump, to achieve these high temps the condensing temperature would also need to be very high. Which means our COPs drop dramicatically (never mind the other technical issues), Would you then buy one?
You should now be think about increasing the fresh air make up flow, to reduce those condensing temperture, More air in means more air out, The air that is expelled, is likely to have a less favouravble properities (low dew point), which then causes the evaporation temperature to fall, again reducing the COPs. This also can have a detremental effect on the product, some products skin if dryed to quickly, reducing the flow of entrapped moisture in the product, which normally has to be overcome with a higher temperature (again not so good for a heat pump)
Why not just but the heating coil in the main air stream, always sounds like a good idea, but in many drying processes the main air stream is all ready heated, so similar sets of problems occur, one of the main limitations to just increasing the air flow is the increase velocity of the air, which if to high can cause the product to leave its container (blows in the wind).
When dealing with higher temperture drying with a heat pump you tend to undersize the heat pump, which normally give excellent COPs and reliabilty and add an addional high temp heat source, which you use for control.
With low temp drying systems you normally oversize your heat pump, and reject excess heat external of the process

So when it comes to drying and heat recoevery with a new system/method it is 80% design and 20% Fudge, more info you have the better you manage the fudge. And is very product specific

[Peter_1]

What you must do once to enlighten this a little bit and this will help you also in the future: make once a simple hand-drawn schematic of the whole system and insert on different copies - for every process phase one (germination after water immersion or spraying I suppose, then the heating) - the right numbers , temperatures ,mass flows of water, their associated temperatures, mass flow of the malt/kiln, time for every production process, mass flow and temperatures of air in's and out's,......

It's a continuous process I suppose where a kiln or several kilns are busy with germinating while others are drying and others are emptied and/or filled up.

So the better we understand clearly the production process and the production flow of it the better we can help you with our experience.

I think MadFridgie will agree with me: this project has very great potentials and can offer your client big savings but it must be very well calculated and documented first. The more precise your numbers are, the more precise you can predict the savings.

[MannyCalavera]

Thanks for advice guys - I have started compiling info since it's up and running again. Hope to really understand the extent of "steady state" that it is actually operating at. Once I have all the parameters of the process fully understood and all the variables, I will see what energy is available and when, and then hopefully you might be able to advise on COPs etc. This within a week or two max I hope.

I had a slightly related question (maybe obvious). Can you control the extent of heating? E.g., if a system was designed to raise a fluid to a temp of 75degC, but for 2 hours you only required 65degC, can heat pump systems cater for this (if source parameters stay constant)?

Pete that invoice looks extremely up-to-date. That factory looks far more asthetic than the maltings I work on!

Regards

[mad fridgie]

Thanks for advice guys - I have started compiling info since it's up and running again. Hope to really understand the extent of "steady state" that it is actually operating at. Once I have all the parameters of the process fully understood and all the variables, I will see what energy is available and when, and then hopefully you might be able to advise on COPs etc. This within a week or two max I hope.

I had a slightly related question (maybe obvious). Can you control the extent of heating? E.g., if a system was designed to raise a fluid to a temp of 75degC, but for 2 hours you only required 65degC, can heat pump systems cater for this (if source parameters stay constant)?

Pete that invoice looks extremely up-to-date. That factory looks far more asthetic than the maltings I work on!

Regards
The simple answer is yes, you have decide is just temperature change, with a similar amouny of produced energy, (reduced glycol flow) or increased temp change and increased energy produced (same glycol flow)
You are now thinking in the correct manor.
Well done
I agree with Pete, its all about understanding the process. I suspect dollar for dollar, that the greatest savings will be made by optomising the diversity of the 6-7 Kilns

[olddog]

Great thread !

Manny, any chance your chief bean counter would release some cask strength 30yo to go DHL to NZ ?

;)

[MannyCalavera]

Haha olddog, not even we get to sample that stuff. F**k healthcare, I want free samples!

You want to go for a Talisker 16yo or a Clyenlish 12yo - they must have that good stuff kicking around NZ?

I don't know what the distribution of single malt whisky is, all i know is around 90%+ gets blended and shipped off to anywhere but Scotland!

Fridge/Pete - got some data coming in now, hopefully within a week I can give you guys concrete numbers (/range of numbers) and you could advise on possible COPs. (if you havn't lost interest already :rolleyes:)

[Peter_1]

Whats yo in 16 yo ?:o
That invoice I found on the net was from 1932. At the end - must be around 1980, it was then already a very old building, ready to demolish it. But it's situated along a nice canal in Bruges...if you know Bruges of course.

I haven't lost interest yet, far to interesting.

Don't forget to search if funds in Scotland are give for such a projects.

[MannyCalavera]

Good morning folks, hope your week is going excellently.

Data is coming in: soon i will know most of what i need to i think, and also i will try and upload a rough schematic of the situation.

Getting specs for the current HX and radiators at the moment. I will keep you posted.

Manny

[mad fridgie]

Looking forword to it.

[MannyCalavera]

Fridgy/Pete,

Thats a rough scetch of the setup attached. The dotted unit is what i reckon might be the most controllable option.

Splitting the hot water in, and using one fractions heat to bolster the remaining flow, which is then thermally better off and that heats the glycol. If the scetch isnt clear let me know.

Numbers to come.

Manny

[mad fridgie]

Fridgy/Pete,

Thats a rough scetch of the setup attached. The dotted unit is what i reckon might be the most controllable option.

Splitting the hot water in, and using one fractions heat to bolster the remaining flow, which is then thermally better off and that heats the glycol. If the scetch isnt clear let me know.

Numbers to come.

Manny
Do you want to put some numbers on the lines
temps, flows (include the glycol)

cheers

mad

[MannyCalavera]

Absolutely. This week I am going to collect the observed data and will ping that onto the diagram.

I will include the range of flows and temperatures of streams that are relevant.


Ta

[mad fridgie]

Hi manny check your calculations.
A couple of points to assist.
The bigger the compression ratio on a refrigeration plant the less efficient.
The larger the number of stages of heat exchangers, the greater the original driving force has to be.
As a "VERY" basic rule, for every 1C you drop your refrig evap temp (SST) you disadvantage the system by 3.5%, for every 1C you increase the refrig cond temp (SCT) you disadvantage the system by 3%.

[MannyCalavera]

Fridgie, I was away to ask;

If i did use ammonia, would i not have to compress it to 23bar for it to condense at 55 deg C? See link.

http://webbook.nist.gov/cgi/fluid.cgi?Action=Load&ID=C7664417&Type=SatP&Digits=5&THigh=120&TLow=-50&TInc=1&RefState=DEF&TUnit=C&PUnit=bar&DUnit=mol%2Fl&HUnit=kJ%2Fkg&WUnit=m%2Fs&VisUnit=uPa*s&STUnit=N%2Fm

I have been taught that roughly a compression ratio of around 3 is good. I.e. to get to 9 bar, want 2 compressions. 1-3, 3-9. etc. 23+ bar? That seems high to me, but i am not experienced with this. Could you have say 1-3, 3-9, 9-27, i.e. 3 stage?

Could you elaborate on the dropping of cond/evap temperatures? I thought the evaporating temperature would be the fluids boiling point at 1atm, or can you regulate the expansion valve to control this?

Much obliged.

[mad fridgie]

Fridgie, I was away to ask;

If i did use ammonia, would i not have to compress it to 23bar for it to condense at 55 deg C? See link.

http://webbook.nist.gov/cgi/fluid.cgi?Action=Load&ID=C7664417&Type=SatP&Digits=5&THigh=120&TLow=-50&TInc=1&RefState=DEF&TUnit=C&PUnit=bar&DUnit=mol%2Fl&HUnit=kJ%2Fkg&WUnit=m%2Fs&VisUnit=uPa*s&STUnit=N%2Fm

I have been taught that roughly a compression ratio of around 3 is good. I.e. to get to 9 bar, want 2 compressions. 1-3, 3-9. etc. 23+ bar? That seems high to me, but i am not experienced with this. Could you have say 1-3, 3-9, 9-27, i.e. 3 stage?

Could you elaborate on the dropping of cond/evap temperatures? I thought the evaporating temperature would be the fluids boiling point at 1atm, or can you regulate the expansion valve to control this?

Much obliged.
You pick your compression ratio to suit the process application.
Where did 55C come from?
Focus on the process not the heat pump
1 What is the fresh air flow through each preheater on each kiln. (range if controlled) primary load
2 What is the air entering temps(s) range to the kilns. primary load
3 What is the glycol flow (total and to each pre heating coil)
4 What is the glycol return temp (range)
5 What would be the ideal glycol supply temp to the kiln coils (this is likely to be design variable)
6 What is your distillary water flow third stage
7 What is you dist. water flow temp (from)
8 water back to dist temp, will be a function of the above energy mass balance
9 You then need to undertake a load profile, to cover the diversity.
10 and only after the other 9 steps have been completed do you then design/concept the heat pump (for performance, duty and control method)
11 look into the practical aspects of the installation.
12 re-look at 10, to see how this is effected by 11.

[Peter_1]

This sketch hasn't the needed figures to make a correct estimate.
We need numbers all over the drawing.
Gathering the basic information is your first task now. Th better you do it, the better you can calculate everything afterwards.
Regarding your NH3 compression.
I haven't read teh article about NH3 compression but far more important is the COP you get.
You again forget your superheat. If you want 55°C, then you don't need to compress at 55°C, more 35°C (discharge of 100°C) or 40°C (113°C)
Like MadFridgie explained you some posts ago - I think - can you use a lot of water at a more lower temperature or do you need higher water temperatures at a lower flow? You can't have both (high flow and temperature) But what is high flow and temperature?

Therefore, what exactly do you need, where and how much at what temperatures.
Again, figures are very important and I don't understand the process like you drawn it.
And I worked some time in malt plants like you know, years ago ;)

Another point, selecting the refrigeration system, let it now NH3, R744 or any other refrigerant will be not that easy. You only can make rough estimate to what this will cost.

[MannyCalavera]

Ok thanks folks. Some more data to get today - tomorrow i will fully annotate that diagram and upload. Hopefully with all the values and ranges of flows and variables in place you might be able to advise on step 11 as you pointed out fridge.

I have most of 1-8 will provide this tomorrow.

I will re-draw the setup so it doesn't look like stevie wonder drew it.

Cheers

[mad fridgie]

Ok thanks folks. Some more data to get today - tomorrow i will fully annotate that diagram and upload. Hopefully with all the values and ranges of flows and variables in place you might be able to advise on step 11 as you pointed out fridge.

I have most of 1-8 will provide this tomorrow.

I will re-draw the setup so it doesn't look like stevie wonder drew it.

Cheers
You can not look at 11 until you have completed 10,
remember on your drawing, to include all existing heat exchangers (how the glycol is heated now and to what temps).
I thought a snail had crawled across the page? lol.

[MannyCalavera]

I am a heat pump novice fridgie but that hurts.

Hopefully get a scan in tomorrow for you guys to peruse if you got a spare minute with all info i currently have.

Manny

[MannyCalavera]

Right sirs,

Find an annotated and hopefully clearer description of what i'm dealing with.

Hard to know exact flow of air - 90,000kg/hr is a good approximation. 5-8 at any one time have 2 batches between them, 3&4 are twice capacity and therefor have 2 between them at any time also.

Air requirement at all times, roughly 360,000kg/hr (it doesn't get much more accurate than this i am afraid). Clearly air temp is going to change alot, fuel used in winter will be much more than in summer, thats the nature of the process.

Glycol properties given. An interesting point, what is desired air (and therefore glycol) temperature. Kilning temps range from 65 - 85 over the kiln period. This is a question of how easy it is to "regulate" the amount of heating the pump supplies. For the sake of a target, i would say preheating the glycol to 65-70 degC would be sound. (i assume this is what you meant by kiln diversity, i.e. profile).

Hot water - yes, large fluctiation there. Nature of the batch process. However, work is being done currently to smooth this out. Average is around 250m3/hr, and hopefully with tuning the variation will drop.

Kiln 3/4, glycol loop is just twice the capacity of glycol and PHE is twice the area.

I guess from a control point of view, the glycol return temperature is one of the biggest variables, depending on ambient air conditions. I.e. in summer when it returns alot warmer, will it have the driving force to perform the condensation of refrigerant? That kind of thing? Glycol can generally be heated to within 2degC of the hot water temp, and returns as indicated on diagram.

Thoughts gentlemen? Some friday afternoon reading ;)

Regards

[mad fridgie]

your present heat reocvery, is saving aprrox 2500Kws (suspect air flow is wrong or wrong TD) Number do not quite add up but close enough.
The figures indicate, 3 out of the 6 kilns are operating in the same mode, at all times.
Total airflow 800K to 1200KM3/hr
Total Glycol 180M3/hr
You should be able to get approx another 5000Kw heating with heat pump power and pumps 1100kw,
COP of 4.5
or
6000Kw heating with heat pump and pumps 1875kw
COP 3.2.
To achieve these results it is a process/system. not just a heat pump

[mad fridgie]

The next question before any stage of the feasability, is do you have the electrical infrastruture, for this level of increased of electrical load.

[Peter_1]

Good point MF about needed power to run this, you will need +/- 600 to 700 kW electrical power to recover 2500 kW.
Giving this, you have to make calculations/decissions if you will use NH3 and spill the absorbed power from the net (open compressors) or use semi-hermetic screws and recover also the consumed power of the net.

[mad fridgie]

Good point MF about needed power to run this, you will need +/- 600 to 700 kW electrical power to recover 2500 kW.
Giving this, you have to make calculations/decissions if you will use NH3 and spill the absorbed power from the net (open compressors) or use semi-hermetic screws and recover also the consumed power of the net.
Already allocated some losses into the calculation, 5000Kw addtional process energy, additional power draw for upgraded process 1100KW. Equipment selection, is not even worth looking into at this stage.

[MannyCalavera]

Good point Fridgie,

That is info I can get on monday. This is an important feasibility study, so even if it turns out to be uneconomical I still need to assess everything. Will reply with info on Monday.

[mad fridgie]

If your figures are correct, then according to my calculations, then this is very viable!
But there is a lot more work to be done.
I am in the UK (first 2 weeks in September) I offer my services to assist you in developing a required process. This would come at cost!
It would 100% need a site visit.

[desA]

A lovely thread, lads. Apologies for my late entry - been moving countries, of late.

I'll go back & re-read the earlier part of the thread to check - but, what kind of process uptime is to be expected from the Process Radiators (24/7?). If not 24/7, this may be something to factor into the overall scheme i.e. a modular system approach may assist.

[MannyCalavera]

Folks,

More numbers coming. Fridgie, unlikely funding would be provided for this "pie in the sky" I, a naive b*stard student, am working on with your guys help. Still, i shall ask ;).

Greetings desA. 24/7, yep. (aside from a few weeks every summer for annual maintance of site).

There is recicrulation of kiln air (as is convention) - i will have to suss out in my head how this effects the quantity of 'fresh' air used during this time.

I have a few more months working here, my aim by the end is to compile a fairly sound report on the feasibility of said heat pump - and I would like it to be fairly worst-case-scenario and as realstic as possible, so that the economic feasibility will not be over stated.

Time for me to call it a day - communix shortly.

Manny.

[desA]

Greetings desA. 24/7, yep. (aside from a few weeks every summer for annual maintance of site).

Ok. Good. Makes things a bit easier.


There is recicrulation of kiln air (as is convention) - i will have to suss out in my head how this effects the quantity of 'fresh' air used during this time.

You may want to estimate fouling factor on the air-side (kiln), as this will impact the radiator design area.

[MannyCalavera]

"You may want to estimate fouling factor on the air-side (kiln), as this will impact the radiator design area."

The radiator in question is a pre-heater, before main steam radiators. Fouling? Don't think this is an issue - fresh air (no sulpherous components, etc) in, glycol solution through tubes.

It is finned - the only fouling i can think of is due to birds sh*tting all over it. (that what you meant?)

I am in the process of finding the specs for these radiators. Currently I am not thinking about replacement of these, more focusing on boosting the glycol temperature which is the working fluid within them. So from that point of view, yes I would have to estimate their effectiveness with different operating conditions to what they were designed for.

[desA]

"You may want to estimate fouling factor on the air-side (kiln), as this will impact the radiator design area."

The radiator in question is a pre-heater, before main steam radiators. Fouling? Don't think this is an issue - fresh air (no sulpherous components, etc) in, glycol solution through tubes.

It is finned - the only fouling i can think of is due to birds sh*tting all over it. (that what you meant?)


If your air is fresh feed (not re-circulated), then you should at least consider dust & general clogging of the fins - unless a pre-filter is used.

This crud will reduce heat-transfer a fair bit & should be allowed for in your calculations/specifications.

You may also want to consider the radiator fin material - copper, or aluminium (coated) etc.

[mad fridgie]

Welcome DesA, I was wondering when you were going to join in.
To gain the efficiencies stated, there are many items that need to be studied, not just the heat pump.
An Annual load profile needs to completed with the diversity of the kilns and the distilling process included (changes in required inputs and out puts and how they correspond to each other)
Manny, if your figures are correct, and they seem to be, then please do not think of this as a "Pie in the Sky" project. There may well be limits on, performance, capital cost. It is only a question of finding the balance.
You are in a position to save as much energy as upgrading a 1000 homes, and likely at a fraction of the cost.
I would be that confident on making substantional savings, that i would install for free, for 60% of the annual savings.

[MannyCalavera]

desA, you are right - probably worth checking more frequently the condition of the radiators, as maximising heat transfer across them is crucial.

"I would be that confident on making substantional savings, that i would install for free, for 60% of the annual savings."

Is this a hypothetical scenario whereby you would do work and you are using this as an example of how much potential there is?
To reiterate as a student it is my role to investigate feasibility and as far as any capital being spent, not a bean until I manage to persuade a few higher up people that this is something that can be economical and operable.

Fridgie regarding available electric, i believe there is somewhere in the region of a MW. Awaiting conformation.

Regards.

[mad fridgie]

Hi Manny, no was not a hypothetical scenario, I would sign a contract tommorrow. (of course due diligance required, i would need to check your numbers "Sorry") I would design, supply and install all equipment required for free to your client, then I would then take 60% of the annuall savings each year (based upon agreed load profile, or in other words your malting production constant).
As you get older you start to look for passive incomes, because young wipper snappers like your self, should and hopefully become the masters of these types of design leaving us old farts (present masters) dealing with past out dated designs and thought patterns.
One benefit of being a student, is that you have all to gain and nothing to loose.
If you remember earlier in the post we talked about commercial reality (payback period), so when calculating these types of situation, it becomes sort of second nature, to inbuild fiscal and engineering principles when calculating (unfortunately this only comes with experience)
PS you would have a job as well!!!!

[MannyCalavera]

Ah, well your confidence in the system I have described can only be a good thing.

Hoping to get elec. avaiable today at the maltings.

May I ask what setup you are currently basing your calcs on fridgie?

Spliting hot water, recovering heat from one fraction to input into the remaining stream to heat the glycol?

Or recovering heat from the warm return stream to put back into the full flow incoming hot stream?

For sake of I will from now on call 50-55deg C water hot, and 40-45 deg C water warm.

Merci

[desA]

@ Manny:

MadFridgie knows what he is talking about. If your company does become serious about your proposal, you would do well to have him come over & work through the finer details with you.

Energy recovery on a large scale requires skillful system designers, with a lot of experience in the game. On this thread you have been privileged to tap into the brains of some of the best practitioners in the field. Peter is top class.

Saving so much energy translates to an extraordinary amount of potential savings - especially if HFO (heavy furnace oil) is being used as a fuel source. In today's world, HFO should be fast moving towards a three-letter swearword. :)

[mad fridgie]

Ah, well your confidence in the system I have described can only be a good thing.

Hoping to get elec. avaiable today at the maltings.

May I ask what setup you are currently basing your calcs on fridgie?

Spliting hot water, recovering heat from one fraction to input into the remaining stream to heat the glycol?

Or recovering heat from the warm return stream to put back into the full flow incoming hot stream?

For sake of I will from now on call 50-55deg C water hot, and 40-45 deg C water warm.

Merci
Bon jour,

You may have gathered by now, that all I am really interested is the process. the process determines the solution.(concept at this stage "10") Now you need to look at the practical application "11", for example you have a spare MW of power, this less than either of the 2 possible options I have given, so we have re-look at the concept, and see how we can get the best use of the max MW of electrical power (more than likely less total heating output, but a higher COP). This continues all through the stages, but it is very difficult to pass on experience gained. thats why you need someone who can look at the practical implications.
Apart from you air airflows, figures are starting to balance (which will actually balance to meet the airflow load)
The actual heating process (in my mind) is non of the above, as there are to many losses in efficiency in the above system.

I do not think my french is up to scratch, and my english is not much better (Lancashire schooling for you)

[MannyCalavera]

Excellent cheese.

From a "green" and efficient point of view, and if one accepts that fuel oil will always be burned en mass (more or less), the highest COP choice would seem to be the most logical. I guess if i were to start writing a report in it now (which i aint, dont worry) the objective would read

"Heat pump feasibility: investigation into significantly increasing air pre-heat temperatures using a heat pump to recover energy and boost the thermal quality of the hot water loop and therefore glycol runaround".

I just read - you are thinking of another setup? I suppose there will be a whole range of different options. In terms of utilising exhaust air, i dont think that's worth looking into - far away from glycol unit, high up, exhaust stack, all sorts of pish.

[MannyCalavera]

Just saw your post there desA - I am aware I am extremely lucky to get the input of Fridgie and Pete - refrigeration engineers are not in abundance here from what i can tell, and university folks have no practical perspective.

I realise it is a relatively specialist area of engineering, and although i would like to be able to do the theoretical calulations etc myself, there is clearly much more thought required than a simple vapour compression cycle. Searched high and low for textbooks / literature on industrial heat pump design, to no avail - it seems to be in your folks heads.

I see it my job to present them with a decent amount of info which explains the benefits a heat pump would bring - and the fact it would save x thousand litres of oil per month, CO2 tonnage saved, etc.

Then from there in the future they would decide to seek consultancy on the subject. So i feel i need to gain an insight into design and operation, given the process parameters, that would be the starting point for trying to present the idea to anyone in control of capital expendature.

But yes - extremely grateful for all help thus far :D

[mad fridgie]

Excellent cheese.

From a "green" and efficient point of view, and if one accepts that fuel oil will always be burned en mass (more or less), the highest COP choice would seem to be the most logical. I guess if i were to start writing a report in it now (which i aint, dont worry) the objective would read

"Heat pump feasibility: investigation into significantly increasing air pre-heat temperatures using a heat pump to recover energy and boost the thermal quality of the hot water loop and therefore glycol runaround".

I just read - you are thinking of another setup? I suppose there will be a whole range of different options. In terms of utilising exhaust air, i dont think that's worth looking into - far away from glycol unit, high up, exhaust stack, all sorts of pish.
The air is an option, and you could use a stacked packed tower, but no I have not gone down this track.
Just better use of the energy available.
Your feasability statement would be incorrect.
"Heat pump feasibility: investigation into significantly increasing air pre-heat temperatures using a heat pump to recover energy"
would read correctly

I have given you all you need to come up with the process (just not in a single description) Giving you the direct answer does not help you learn (An argument i have with my wife when doing the kids homework, learn the method, not just the answer)
I will give you a clue, start with the air (even though I think your figures are wrong)and work backwards.
Keep up the good work, look at the process not the individual parts.

[desA]

Just saw your post there desA - I am aware I am extremely lucky to get the input of Fridgie and Pete - refrigeration engineers are not in abundance here from what i can tell, and university folks have no practical perspective.

I realise it is a relatively specialist area of engineering, and although i would like to be able to do the theoretical calulations etc myself, there is clearly much more thought required than a simple vapour compression cycle. Searched high and low for textbooks / literature on industrial heat pump design, to no avail - it seems to be in your folks heads.


What you'll find on RE are experienced folks who know how to both design & implement a heat-pump system solution.

The university folks generally only have theoretical solutions - good on paper only. People who have spent many years implementing RHVAC systems will be able to provide an operating, safe, robust solution.

In MadFridgie & Peter, you have two of the best around, in my view - combining both a solid theoretical knowledge with extensive experience.

Enjoy the learning experience. :D

[MannyCalavera]

Righto, I am going to take a day or two off from the heat pump study so's i can view it with fresh eyes all of next week. I will re-read the thread and your suggested method of process selection, and let you know my findings and thoughts fridgie.

Cheers

[MannyCalavera]

Quick friday afternoon question.

In a heat pump the working fluid is going to be cold after it is passed through expansion valve, or similar device - very cold, oui?.

Aim of, say, the PHE with your heat source is to evaporate the refrigerant.

My question: if you have the choice of 40 or 52 degC water at equal flowrates, does it make a huge difference which flow you would choose to provide the evaporation?

My thoughts being the driving force for evaporation (temperature wise) would be so large anyway an extra 10 degC in source temperature wouldn't effect it that much.

Also, what determines how cold you can get your heat source? E.g., due to the very low temperature of your saturated liquid refrigerant, could you take a warm stream of 40 degC down to the likes of 10 degC, providing you had enough flowrate of warm water?

That is all for now - have an excellent weekend

[mad fridgie]

In reverse, travelling for 42 hours to your neck of the woods, I hope I have good looking air hostesses (if not, then weekend will not be so excellent),
generally the limit on evaporation temps from the comp manufactures (without the need for a special machine)
Not very cold, where the energy mass balance reaches equalibrium.
The design determines the temperature. (you pick it).
List your requirements,
One was 25C returned back to the process (no need for the cooling tower)
Look at the process not the heat pump. (The heat pump is designed around the process and the capital cost verses running cost)
NOT best engineering.

[MannyCalavera]

I have looked at the diagram of the setup for a while now, not getting any major ideas. (it'll help to look at that schematic i drew)

What seems to make sense to me is using the warm water as the heat source. There is a hold up tank of it before it is returned to the dist. It is waiting to be cooled, but it still has a large amount of energy.

You could take out a fairly steady flowrate (close to the stream flowrate, maybe a jot less) of warm water and pass it through a PHE, then back to the tank.

Using this energy stream to do the evaporation it would cool substantially (also a bonus).

You would pick a desired refrigerant capacity/volume based on the minimum expected flowrate of warm water expected, to ensure it would always give the evaporation duty. This would not upset the warm water return in any way.

Current glycol HX's have an excess flow of water (easily bringing the glycol up to hot water temp), however i suspect in winter this will not be the case and we will need all the hot water. So for that reason i think it would be best not to split the hot water as previously thought.

Where to dump all these hundreds of kWs of energy?

PHE in the line of the hot water directly dumping the energy. This flowrate is variable, however if this is the case and you cannot achieve 100% condensation of refrigerant, does this cause a problem?

In this setup you could regulate the flow of warm water through the evaporation PHE which would be easily controlled. You could drop it X deg C depending on what worked out to be most efficient.

You said start with the air flowrates - not sure why. (though i am not questioning your skill ^^) I had done some thermo calcs recently, ~ 80,000 kg/hr/batch is required. As the kilns are paired the demand for air should be constant and consistent.

Air temperature is the biggy. % RH varies also however this just relates to kiln time. Temperature will effect the glycol return temperature. This will effect the hot water outlet temperature.

So say in winter, air colder, glycol cooled more, warm water out is cooler, into evaporation PHE cooler, then increase flowrate of warm water through PHE.

I think the good thing about the plant is that the demand for hot air is always there in roughly the same quantity. As i said before, air is desired at 65 C for half the kilning, and then ramped u p to 85 for the remainder. So a target air pre-heat of 60-65 should be aimed for.

In my mind this seems to make sense and seems operable. I will try and crunch more numbers on it now. Please let me know why what I have just explained is totally inoperable and wrong. :rolleyes:

Manny

[MannyCalavera]

Ok Fridge have a good journey, wont expect any feedback for a few days.

Enjoy

[Peter_1]

Sorry for the absence but I have to finish a new course for a new 1st year and the agreement was that it must be finished on 31 august. Started a little bit late with it.
I now have +/- 100 pages, +/- 30 to go.
I then will rejoin this thread.
I haven't read the posts of last week.

[mad fridgie]

I have looked at the diagram of the setup for a while now, not getting any major ideas. (it'll help to look at that schematic i drew)

What seems to make sense to me is using the warm water as the heat source. There is a hold up tank of it before it is returned to the dist. It is waiting to be cooled, but it still has a large amount of energy.

You could take out a fairly steady flowrate (close to the stream flowrate, maybe a jot less) of warm water and pass it through a PHE, then back to the tank.

Using this energy stream to do the evaporation it would cool substantially (also a bonus).

Yes Correct

You would pick a desired refrigerant capacity/volume based on the minimum expected flowrate of warm water expected, to ensure it would always give the evaporation duty. This would not upset the warm water return in any way.

No (but you are thinking correctly, need to control refrigeration capacity to meet minimum load)

Current glycol HX's have an excess flow of water (easily bringing the glycol up to hot water temp), however i suspect in winter this will not be the case and we will need all the hot water. So for that reason i think it would be best not to split the hot water as previously thought.

Correct do not split the flow

Where to dump all these hundreds of kWs of energy?

PHE in the line of the hot water directly dumping the energy. This flowrate is variable, however if this is the case and you cannot achieve 100% condensation of refrigerant, does this cause a problem?

Not sure what you mean here, (i think what you are taliking about is load limiting)

In this setup you could regulate the flow of warm water through the evaporation PHE which would be easily controlled. You could drop it X deg C depending on what worked out to be most efficient.

Correct

You said start with the air flowrates - not sure why. (though i am not questioning your skill ^^) I had done some thermo calcs recently, ~ 80,000 kg/hr/batch is required. As the kilns are paired the demand for air should be constant and consistent.

This is the primary load, so you need to work out the best way to introduce energy into the air stream, because this can be improved, if so it then effects all other stages.

Air temperature is the biggy. % RH varies also however this just relates to kiln time. Temperature will effect the glycol return temperature. This will effect the hot water outlet temperature.

Correct

So say in winter, air colder, glycol cooled more, warm water out is cooler, into evaporation PHE cooler, then increase flowrate of warm water through PHE.

Correct

I think the good thing about the plant is that the demand for hot air is always there in roughly the same quantity. As i said before, air is desired at 65 C for half the kilning, and then ramped u p to 85 for the remainder. So a target air pre-heat of 60-65 should be aimed for.

Maybe (ideal), maybe not (practical)

In my mind this seems to make sense and seems operable. I will try and crunch more numbers on it now. Please let me know why what I have just explained is totally inoperable and wrong. :rolleyes:

Manny

At this stage do not determine what type of heat exchanger you are using "PHE" just call it H/E (there are other types and should be designed to suit the application, not presumed just to be PHE)

Well done!! Now you are think of the process and how each part interacts with each other.

cheers
Mad

[MannyCalavera]

Excellent! Very helpful breakdown, much appreciated.

I see what you mean about air flow and current radiators, will need to do some looking into those.

Next week will be number crunching time based on observed flowrates and temperatures expected. Will report back then with info's.

Cheers,

Manny

[MannyCalavera]

"PHE in the line of the hot water directly dumping the energy. This flowrate is variable, however if this is the case and you cannot achieve 100% condensation of refrigerant, does this cause a problem?"

What I mean is if the hot water feed to maltings varies flowrate, say a decrease, will it have sufficient capacity to condense (& cool) vapour refrigerant? Perhaps at a reduced flowrate it would just increase in temp but still have the driving force to condense refrigerant.

I suppose it depends on condensing temps.

[mad fridgie]

"PHE in the line of the hot water directly dumping the energy. This flowrate is variable, however if this is the case and you cannot achieve 100% condensation of refrigerant, does this cause a problem?"

What I mean is if the hot water feed to maltings varies flowrate, say a decrease, will it have sufficient capacity to condense (& cool) vapour refrigerant? Perhaps at a reduced flowrate it would just increase in temp but still have the driving force to condense refrigerant.

I suppose it depends on condensing temps.

As long as we know this occurs then it can be handled in a number of ways, this then comes down the practical application and what benefits/ disadvantages may or may not occur.
The compressor do not have infinite working pressure ratios or preesure limits (in some case these determine the control methodology)

[MannyCalavera]

Ok so a varying hot water flowrate (the stream to recieve the heat from the heat pump) would not cause an operational tits up scenario? Excellent

[mad fridgie]

Ok so a varying hot water flowrate (the stream to recieve the heat from the heat pump) would not cause an operational tits up scenario? Excellent
Going back to air flow, driving force for the varying stages of heat transfer, re look at you quote above, then come back, clue " do not look at heating the water with the heat pump"

[MannyCalavera]

Ok I will re-look through and see what else I can come up with based on what you're saying fridgie. Not much mental capacity today, birthday yesterday and whisky does not sit well ontop of alot of red wine apparently.

Will get in touch later in week.

Cheers, Manny

[mad fridgie]

Ok I will re-look through and see what else I can come up with based on what you're saying fridgie. Not much mental capacity today, birthday yesterday and whisky does not sit well ontop of alot of red wine apparently.

Will get in touch later in week.

Cheers, Manny
There is a saying never mix grape and grain.
heres another clue, in a perfect world we would use the condensor of the heat pump to directly heat the air, this would give maximum efficiency (this is not a ruled out option, but info indicates that this would be an unlikely option, for practical, technical control reasons), so what would be the next best method?

[MannyCalavera]

Direct condenser use to heat the air had crossed my mind Mad, however if the refrigerant vapour is super-heated, is it not the case that poor heat transfer coefficients between air / vapour would cause an issue?

I am not sure how superheated the vapour would be, if it was a large amount or just sitting above condensing temperature.

I presume you are indicating towards direct glycol-refrigerant heat exchange.

From that viewpoint I can see a couple of methods; 1) Use the hot water as a source, and use refrigerant on the glycol loop to heat it.

2) Keep the hot water-to-glycol PHE as it is, and then use the warm water as the heat source of the heat pump, and use this high temperature refrigerant to further heat the glycol

Tomorrow i will sit down and think more about why this is more / less advantages & efficient.

Cheers.

[mad fridgie]

Direct condenser use to heat the air had crossed my mind Mad, however if the refrigerant vapour is super-heated, is it not the case that poor heat transfer coefficients between air / vapour would cause an issue?
No this not the issue with the direct system, it more to do with the amount of refrigerant and where it sits (hard to explain in a few words)
I am not sure how superheated the vapour would be, if it was a large amount or just sitting above condensing temperature.
The process determines the superheat temp and the amount of usable energy, remember one of the first questions do you want temp or energy.

I presume you are indicating towards direct glycol-refrigerant heat exchange.
Correct

From that viewpoint I can see a couple of methods; 1) Use the hot water as a source, and use refrigerant on the glycol loop to heat it.

2) Keep the hot water-to-glycol PHE as it is, and then use the warm water as the heat source of the heat pump, and use this high temperature refrigerant to further heat the glycol.
Correct, now you are getting it, but do not at this stage discount the other options, as you need to look at yearly load profiles and how changes in the air temp effect the process loops

Tomorrow i will sit down and think more about why this is more / less advantages & efficient.
Also remember that we have power limitations, (1 MW) so it now about best use of energy not temperature.

Cheers.
Keep up the good work

[MannyCalavera]

"Correct, now you are getting it, but do not at this stage discount the other options, as you need to look at yearly load profiles and how changes in the air temp effect the process loops"

yes good point sir, I guess i need to come up with a couple of spreadsheets for the different setups, with all the yearly process condition changes and assess how each one would respond and have a punt at how effective it would be.

Hope the UK weather is treating you well, very nice up north!

[MannyCalavera]

Got a fair document of all the process variables now, in terms of flow and temperatures and operating times. There is physically room also on site around the HX / hot water recovery shed.

I bet you won't like this question; where should i start when thinking about compression?

Need to compress refrigerant to a pressure such that it will condense T,c above T,sink. That's all i know.

Compressor type (screw?), ideal compression ratio's, superheat temps achieved - any guidance on this area of design?

Perrys Chemical Engineer handbook has the theory but again I sense it will be unpractical :rolleyes:

Any input appreciated.
NB i am still thinking about the process first and pump second, fear not.

Cheers,
Manny

[mad fridgie]

Got a fair document of all the process variables now, in terms of flow and temperatures and operating times. There is physically room also on site around the HX / hot water recovery shed.

I bet you won't like this question; where should i start when thinking about compression?

Need to compress refrigerant to a pressure such that it will condense T,c above T,sink. That's all i know.

Compressor type (screw?), ideal compression ratio's, superheat temps achieved - any guidance on this area of design?

Perrys Chemical Engineer handbook has the theory but again I sense it will be unpractical :rolleyes:

Any input appreciated.
NB i am still thinking about the process first and pump second, fear not.

Cheers,
Manny
You are right, do not like the question,
The question can not be answered, unless you know all process variables.
This will determine the type of compressor, also possible refrigerant choice, all these effect SST, SCT, superheat, sub cooling thermal amd mechanical loss. Which all change as the processes change.
Not much help at all, sorry!!!

[MannyCalavera]

No worries Fridgie, saw that coming. If i were to submit a reply / word doc with a detailed breakdown of all the process variables with the degree of variation depending on process / seasonal changes / other factors, along with numerical values that i have given before (but now refined), and a detailed physical layout, would you be able to (with your experience in the industry) look at that and go

"so this is changing, this is constant, you have this much energy available, this is what you require, blah blah" and from that be able to describe a setup which would be suitable?

I have worked through this data the last two weeks and will compile it for next week, upload it here also for perusal. It will list all variables which effect the glycol system / magnitudes of flow and all foreseeable variation, seasonal or otherwise.

It would be easier if you could transfer information like in The Matrix, so I wouldn't need to keep asking questions here :o

"Want some more?"
"Hell yes."

[mad fridgie]

No worries Fridgie, saw that coming. If i were to submit a reply / word doc with a detailed breakdown of all the process variables with the degree of variation depending on process / seasonal changes / other factors, along with numerical values that i have given before (but now refined), and a detailed physical layout, would you be able to (with your experience in the industry) look at that and go

"so this is changing, this is constant, you have this much energy available, this is what you require, blah blah" and from that be able to describe a setup which would be suitable?

I have worked through this data the last two weeks and will compile it for next week, upload it here also for perusal. It will list all variables which effect the glycol system / magnitudes of flow and all foreseeable variation, seasonal or otherwise.

It would be easier if you could transfer information like in The Matrix, so I wouldn't need to keep asking questions here :o

"Want some more?"
"Hell yes."
The simple answer is "yes"
I suspect that you are still a lttle away from this stage, but i will have a quick look, but you are now moving into info that is in the commercial field "well you have been it it for quite some time"
By now you should have come up with some very basic savings and a rough acceptable return on investment (not the same as the equipment cost, but a limit to the cost, give or take), so I do not see me giving you the design method, but a results method, prop. not what you would ideally want.
cheers
Mad

[MannyCalavera]

Doing some cost evaluation.

Mad I recall you used a figure of ~ £400k + for the cost of heat pump plant. Is this realistic?

I presume capital cost doesn't really increase with heat pump load, this would just be an operating cost?

For the sake of doing some evaluation is it possible based on the type of system I have described to give a rough power requirement for compression? 200-300kW, for example?

Any punts welcome, or other related capital / operating costs to take into account.

Cheers

[mad fridgie]

Doing some cost evaluation.

Mad I recall you used a figure of ~ £400k + for the cost of heat pump plant. Is this realistic?

I presume capital cost doesn't really increase with heat pump load, this would just be an operating cost?

For the sake of doing some evaluation is it possible based on the type of system I have described to give a rough power requirement for compression? 200-300kW, for example?

Any punts welcome, or other related capital / operating costs to take into account.

Cheers
Capital cost really does change with heat pump load.
If you are looking at 300kw power draw, then you should be well in with 400,000 pound.
At this stage you could calculate on a COP of 4 (1200Kw), this would give you an approx saving of 24 pound an hour
cheers
Mad

[MannyCalavera]

Done some quick cost work;

Electricty per kWh = £0.08

Based on mass&energy balance, gaining +1degC air preheating all the 3 paired loops (6kilns) saves £38,000 annually

Say power consumption of heat pump was 200kW
8000hr working year; £125,000 annually on elec.

Being optimistic and assuming a COP of 6,

200kW in = 1.2MW out

Average hot water flow to be assumed ~ 200m3/hr

Temperature lift due to 1.2MW ~ 5.2 degC

Savings on fuel oil ~ £205,000

Annual saving; £76,000

Very rough and ready calcs but a fair saving annually.
Any comments?

Cheers

[mad fridgie]

have you an updated cost of the fuel heating i thought t was 4 pence Kwhr

[MannyCalavera]

11.7 kWh/litre

~ £0.47 per litre

£38,000 comes from air mass flowrate across the 6 kilns

each batch ~ 90,000kg air per hour

4 batches at any one time

[mad fridgie]

11.7 kWh/litre

~ £0.47 per litre

£38,000 comes from air mass flowrate across the 6 kilns

each batch ~ 90,000kg air per hour

4 batches at any one time
0.47 / 11.7 = 0.04pound/kwhr

So 1.2mega watt saving *8000 hours * ).04 =
Yeary saving on fuel 384000 pounds,
less electric running cost
125000 pounds,
actual saving 250,000 pond

[MannyCalavera]

With you on elec Fridgie, substantially higher fuel savings than i am getting.

May I ask your assumed/calculated CoP?

At this stage I am to present initial findings to some people in the company.

I feel i have got a good enough understanding of design considerations to present the case to them for use of a heatpump. Work will be ongoing until xmas and i'm sure by then i can present a more detailed case of the proposed setup(s).

Thanks for all the guidance up till now Mad, very much appreciated. Any developments, i will post here.

Regards

[mad fridgie]

With you on elec Fridgie, substantially higher fuel savings than i am getting.

May I ask your assumed/calculated CoP?

At this stage I am to present initial findings to some people in the company.

I feel i have got a good enough understanding of design considerations to present the case to them for use of a heatpump. Work will be ongoing until xmas and i'm sure by then i can present a more detailed case of the proposed setup(s).

Thanks for all the guidance up till now Mad, very much appreciated. Any developments, i will post here.

Regards
Calcs have to been made using same data,
you stated 8000 hrs run time, at 200Kw and COP of 6 (I have no idea where this figure came from) so you must also calculate the fuel savings based upon the same run hours. 1.2Mw * 8000hrs. To be conservative
I would use your 8000 hr run time (seem reasonable)
I would how ever use a COP of 4 or 300Kw power draw to produce 1200Kw of heat energy, the you can convert to an annual pound value.
cheers
Mad

[MannyCalavera]

Been reading an online pdf "Industrial heat pumps for steam and fuel savings". It outlines a method to estimate COPs and savings for a system for mechanical heat pumps.

Basically you estimate your heat pump inlet temperature (function of heat source temp) and the heat pump outlet temperature (you can set this, but need to factor in sufficient approach temp for the condenser).

CoP is then estimated as T,out / (T,out - T,in)
, in degrees Rankine.

Take an efficiency of 70% from ideal CoP, and this gives you an approximate CoP based on the temperatures you have specified.

I have done this and get 6.1.

Then Work,in = Q,in / (COP - 1)

which turns out to be 180kW (compressor power i presume).

Q,out = Q,in + Work,in


From this method it seems that the inlet and outlet temperatures of the heat pump (compressor) play a large role in efficiency. Is this fairly true?

I see what you mean about needing all the process variables tied down, playing around on Excel, even small changes in flow / temp massivly change efficiencies.

Cheers

[mad fridgie]

Been reading an online pdf "Industrial heat pumps for steam and fuel savings". It outlines a method to estimate COPs and savings for a system for mechanical heat pumps.

Basically you estimate your heat pump inlet temperature (function of heat source temp) and the heat pump outlet temperature (you can set this, but need to factor in sufficient approach temp for the condenser).

CoP is then estimated as T,out / (T,out - T,in)
, in degrees Rankine.

Take an efficiency of 70% from ideal CoP, and this gives you an approximate CoP based on the temperatures you have specified.

I have done this and get 6.1.

Then Work,in = Q,in / (COP - 1)

which turns out to be 180kW (compressor power i presume).

Q,out = Q,in + Work,in


From this method it seems that the inlet and outlet temperatures of the heat pump (compressor) play a large role in efficiency. Is this fairly true?

I see what you mean about needing all the process variables tied down, playing around on Excel, even small changes in flow / temp massivly change efficiencies.

Cheers
You can have what ever COP you want, comes down to what numbers you put "in", add what does this come down to, you have guessed it "The Process"
Refrigeration is all about compression ratios.
So here is a question to keep you thinking and is topical
Is better to produce 1MW of heat with a COP of 6 or produce 4MW with a poor COP of 4
(Back in NZ)

[MannyCalavera]

4MW with a COP of 4 would have larger savings. I see your point there.

I have come to realise that conditions won't vary as much in winter as i had presumed. Cold air over the glycol radiators won't remove a much larger amount of energy from the glycol. Delta T across glycol likely to change only slightly.

Anyway I have a question. Being in the heat pump business you must keep a close eye on the relative cost of electricity versus fuel oil.

And of course this has a large influence on heat pump economics and profit. I know you are in NZ but have some UK knowledge - over the next 2, 5, 10 years - do you reckon this will swing in favour of elec:fuel oil? Or will the CO2 aspect be the larger driver?

[mad fridgie]

4MW with a COP of 4 would have larger savings. I see your point there.

I have come to realise that conditions won't vary as much in winter as i had presumed. Cold air over the glycol radiators won't remove a much larger amount of energy from the glycol. Delta T across glycol likely to change only slightly. Correct, work is need on this area, (large savings can be made here)

Anyway I have a question. Being in the heat pump business you must keep a close eye on the relative cost of electricity versus fuel oil.

And of course this has a large influence on heat pump economics and profit. I know you are in NZ but have some UK knowledge - over the next 2, 5, 10 years - do you reckon this will swing in favour of elec:fuel oil? Or will the CO2 aspect be the larger driver?
This is only an opinion, I suspect that the difference between the price of a raw heating fuel (any) and the price of electricity will become somewhat closer, as focus is given on renewables (wind, wave, solar and I also think nuclear will rise again) The CO2, well your guess is as good as mine, I think we should look at protecting resources, more than focusing on a reason for protecting them.

[MannyCalavera]

Aye.

The other thing that had become fairly clear was the need to minimise the hot water flow through the glycol heat exchangers.

This is so that the Q,in (based on Q,out + W,in -> source and compression respectively) results in the maximum temperature lift of the hot water.

I.e. 1.5MW in the condenser would result in a greater temp rise of a 180m3/hr stream, instead of say 210m3/hr.

The 3 PHEs are oversized so i think they would handle hotter water to bring glycol outlet temp up to its inlet temp (within reason). That make sense?

Finally - with a boosted hot water temperature, a new equilibrium would be established in the system, and undoubtedly the warm water out would be hotter. This has to be a good thing for a heat pump evaporator point surely? (temp lift smaller).

Best regards.

[mad fridgie]

It is a fine balance between energy and temperature,
why do you want to heat the water, you want to cool it!, you want to heat the glycol.
Remove as much heat as possible from the distillary water, directly into the glycol, then further reduce the water by refrigeration (lets say to 25C as required for the return), use HOR of the refrigeration plant to heat the glycol directly, we do comprimise the Heat pump COP, but have a greater amount of FREE energy.
Remember, think of the process as a whole, each loop effects other loops.
Only at the very final stages, can you look at specifics,
At the start you had very big loops, these now should be starting to reduce, to where an equalibrium is being reached in each loop, final control of the heat pump, finishes the loop balancing

[MannyCalavera]

I recognise heating the glycol directly with refrigerant is the most efficent - but i was thinking about practical operation.

Incoming hot water is 1 line, if it was getting heated, would require 1 condenser., one control loop.

As is the 3 glycol loops are independant of one another (not fed from a main line), that would require 3 condensers (each with potentially different loads, (different radiator performances, etc).

Internal heat exchanger reconfiguration at the distillery is going on - hopefully this will boost the water temperature to the maltings by 2/3 degrees, and also work is being done to smooth out the flow pattern.

[mad fridgie]

I understand, that your are looking now the practical application, which is good! I am unable to detrmine the correct design scenerio, without knowing the load profiles, thus total potential savings, which then limits the capital allocation, plus the physical limitations of the installation.

[MannyCalavera]

Just to confrirm Mad, when you talk about "load profiles" - this is the expected duties/range of variation on the condenser and evaporator?

[mad fridgie]

Just to confrirm Mad, when you talk about "load profiles" - this is the expected duties/range of variation on the condenser and evaporator?
Not quite, it is load profile of the process(s) then indirectly the refrigeration limitations.
I believe this not just a case of adding a heat pump (and is the simplest option) to truely save while you have the oppotunity (without going overboard) focus on the whole.
You also need to look at things such as life of the system(s), for example if you had a single system/heat exchanger, how long to do you think it would take to get a replacement and what would be the cost of the plant not being in operation.
It is all about finding a balance, which is not as easy, as saying it.
You should have reasonable idea on a range of savings, thus a limits on a range capital expense,
You then start to move money around the different potential engineering solutions, and hopefully you reach a viable solution. (my gut feeling is you can do quite a lot) improved air heating, heat pump and control systems.
Mad

[desA]

Manny, you really do need to bite the bullet, fly MF over to see you & let him set you on the right path.

I see that he has been extremely kind in working around your technology probes, but, I must honestly say that you must both be very close to your respective IP boundaries.

Fairness should prevail at this point. Surely?

[r.bartlett]

Manny, you really do need to bite the bullet, fly MF over to see you & let him set you on the right path.

I see that he has been extremely kind in working around your technology probes, but, I must honestly say that you must both be very close to your respective IP boundaries.

Fairness should prevail at this point. Surely?

I doubt Manny is in a position to sign off such a request on a purely theoretical study.They would need to be nearer to finalizing it for that to happen.There are such things as pictures, movies conferance calls Skype etc that will cover most things for now.

Plus I suspect someone would point out that using a company from the other side of the world would be fraught with difficulties with regard to accountablity and as such would assume they would prefer someone more 'local' ?

[desA]

^ Well, so said. It would be wonderful to have seen someone 'local' take the trouble to answer so patiently as has MF & with as much depth of experience.

My 2 pence worth, mind you.

[r.bartlett]

^ Well, so said. It would be wonderful to have seen someone 'local' take the trouble to answer so patiently as has MF & with as much depth of experience.

My 2 pence worth, mind you.

I don't doubt MF's technical experience or ability. I just suspect a bean counter will have his input and question why they need to fly someone 1/2 way round the world. If that can be answered sufficiently then no doubt MF will get a BA ticket through the post at some point in the future. :D

[MannyCalavera]

I appreciate your point desA, but as a student on placement I have no hope of securing cash to bring a consultant like Mad over. If I did then i would have when he was in the UK recently - and that would have been damn useful.

And Mad knows i am extremenly grateful for all his advice. I am now in a position where I can make rough estimates to economic potential, while baring in mind the whole process and operability of the heat pump thanks to his input.

Bartlett hit the nail on the head regarding my role to the company in this however. Getting a consultant in to look at the problem is a fair way down the line. The first stage is for me raise this to their attention and explain how it would work and what savings it could offer. And for this I need more than a basic explanation.

If it weren't for these wonderous forums & MF I would still be at stage one, on Excel trying to work out power of compression based on theoretical equations out of Perrys Handbook.

I.e. desA, I am far from taking the help for granted - but I have no weight to throw around $$$ wise.

Cheers

[desA]

This thread is an interesting read. Keep it up everyone. :)

[MannyCalavera]

If and when this project becomes more than a study (which im sure in the future it will given carbon neutrality aspirations) I will be sure to ship over some 24yo cask str single malt, fear not.

Regarding your previous thoughts mad: if some of the heat pump kit were to fail or require maintenance, it would have little impact on the kilning process.

The system would switch to use exhaust air-to-glycol cycle as a means of pre-heating (for the down-time of repair your steam use - the primary heating - would ping up as this pre-heating is small compared to hot water use), however production would not be effected.

You would just have to burn X thousand more litres of fuel oil to compensate.

[mad fridgie]

Hi, Manny did make it clear where his limitations lie (no worries there), always willing to accept a bit of Malt.
At some stage, a consulatant will be required. But before this is to happen, there is a need to prove a range of savings (which is what manny is doing), as far as an actual optimised designed process, no offfence given, manny does not yet have the experience (but does have the brains).
When you look at systems like this, you do need to visit site and get a "feeling" for the process. Something which is hard to do electroniclly.
As far as cost, it is not the cost, its the benefit. (as far local, earlier i did say it would via my UK sister company).
back to tech, manny you mis understand my point about service, most recovery systems do not effect the actual processes, only save money, but it is the cost in money terms when a recovery is off.
Thats why you look at multiple plants, with relativly standard equipment. (also it may help in the load diversity)
Focus on your load profile and then the savings, work on COPs of 4 for the heat pump (i know it could be better),
Mad

[mad fridgie]

I have been nudging with atoffe hammer, now lets use a sledge hammer, forget about the heat pump for now. imagine you have infinatively (can not spell) size air to glycol and glycol to water heat exchangers. Work out the theorectical savings, 100% for free.

[MannyCalavera]

Ok so the pre-heating radiators have infinate area, the glycol-water PHEs have infinite area.

Currently glycol can basically reach the hot water temps (oversized PHexchanger, excess of hot water).

Assuming we have an infinite size radiator (with excellent heat transfer coeff) and turbulent flow over the pipes&coils, air temp should be able to reach glycol temperature.

However the seasons come into this hypothetical: 0 degree winter air -> 50degC would save considerably more than 18degC summer air.

However: if we go with the average temp (10 C) and assume with an infinite sized radiator this can reach glycol temps of almost 50 C (air=25kg/s = 1MW).

Here glycol (13kg/s) would cool to ~29 C.

That is a lift of 40degC on the air. (from 10 to 50 C)

Currently, average lift observed is ~ 25 degC

Extra 15 degC approximately equates to £570,000.

[mad fridgie]

Ok so the pre-heating radiators have infinate area, the glycol-water PHEs have infinite area.

Currently glycol can basically reach the hot water temps (oversized PHexchanger, excess of hot water).

Assuming we have an infinite size radiator (with excellent heat transfer coeff) and turbulent flow over the pipes&coils, air temp should be able to reach glycol temperature.

However the seasons come into this hypothetical: 0 degree winter air -> 50degC would save considerably more than 18degC summer air.

However: if we go with the average temp (10 C) and assume with an infinite sized radiator this can reach glycol temps of almost 50 C (air=25kg/s = 1MW).

Here glycol (13kg/s) would cool to ~29 C.

That is a lift of 40degC on the air. (from 10 to 50 C)

Currently, average lift observed is ~ 25 degC

Extra 15 degC approximately equates to £570,000.
I have not checked your numbers, so lets say that they are OK. 570,000 *3 years for reasonable payback, that is a s**t load of surface area, of course the theory is impossible, but you should by now understanding what I am talking about. Understand the process! Find the technical/capital compermise, then look at what energy is left, to be used by the heat pump, which can be used to heat the glycol, further increasing coli duty and thus air temp.

[MannyCalavera]

Radiators are copper tube, aluminium fin - and perform quite well from what i gather.

Obv. you will always have limited heat transfer across them as a function of heat transfer coefficients and pressure drop.

Observing the system, 1degC glycol increase results in a 1degC air increase (quite consistently)

Although 1degC glycol does not equal kW-wise 1degC air pre-heating, it performs this way.

So as a basis for calculation, i am assuming if you have glycol 5 degC hotter, air pre-heating is +5 degC.

This doesnt mean the glycol has 5 degC worth of heat taken out, however.

Perhaps when you are talking temps of 60-70, the radiators wont perform quite as well.

[mad fridgie]

Radiators are copper tube, aluminium fin - and perform quite well from what i gather.

Obv. you will always have limited heat transfer across them as a function of heat transfer coefficients and pressure drop.

Observing the system, 1degC glycol increase results in a 1degC air increase (quite consistently)

Although 1degC glycol does not equal kW-wise 1degC air pre-heating, it performs this way.

So as a basis for calculation, i am assuming if you have glycol 5 degC hotter, air pre-heating is +5 degC.

This doesnt mean the glycol has 5 degC worth of heat taken out, however.

Perhaps when you are talking temps of 60-70, the radiators wont perform quite as well.
Your present system is wrong! so forget about it!
You have a million quid to design new heat exchangers (LOOKING AT SAVING 2/3 OF MAX POSSIBLE SAVING) this type of engineering is all about the money capital verses return.

[mad fridgie]

I don't doubt MF's technical experience or ability. I just suspect a bean counter will have his input and question why they need to fly someone 1/2 way round the world. If that can be answered sufficiently then no doubt MF will get a BA ticket through the post at some point in the future. :D
If you want the best, you pay for the best;)

[MannyCalavera]

I will have a good think about capital vs return then, and open up the concept of new exchangers / radiators, etc to see what would result in most efficient process.

I will get the managers drunk and get them to sign a contract for your concultancy eh

[mad fridgie]

I will have a good think about capital vs return then, and open up the concept of new exchangers / radiators, etc to see what would result in most efficient process.

I will get the managers drunk and get them to sign a contract for your concultancy eh
Sounds like a good reason to get drunk, and that can not be all bad. lol
On system like this, the design evolves and that is OK

[MannyCalavera]

Random heat pump question.

Is it the case the case that after compression the system condenser would just be one heat exchanger unit?

Or are there cases when it is more efficient to have a seperate de-superheater, condenser and subcooler in series?

[mad fridgie]

Random heat pump question.

Is it the case the case that after compression the system condenser would just be one heat exchanger unit?

Or are there cases when it is more efficient to have a seperate de-superheater, condenser and subcooler in series?
Process specific!

[MannyCalavera]

Very quick question here on Tuesday morning.

Clearly the latent heat of evaporating your refrigerant constitutes the bulk of the energy removal from your source.

But i presume your heat source would also sensibly heat it?

Approach temperature depending on the process and system of course :o?

[mad fridgie]

Very quick question here on Tuesday morning.

Clearly the latent heat of evaporating your refrigerant constitutes the bulk of the energy removal from your source.

But i presume your heat source would also sensibly heat it?

Approach temperature depending on the process and system of course :o?
Superheat (sensible) of some level is required, and "normally" relates to requirements of the refrigeration machinery, but is calculated into the total evaporator load (vapour has poor heat transfer properties)

Did you get your managers drunk, when are you flying me over, LOL?!!!!!!!!

[MannyCalavera]

I am working on it!

Been reading a few journals, apparently heat pumps are used quite alot in the lumber drying industry. Got some fairly good information on operability from a control point of view, startup/shutdown considerations (liquid in compressor, etc).

Found these guys up north

http://sylow.ife.no/hybridenergy/hybridheatpump

Seems to be a *fairly* similar type of scenario - but isn't MVR heat pump.

Will get in touch more next week so you can pick at my calculations if you have a moment!

[mad fridgie]

Drying basicallly is the the same, you need energy, it is the process that changes. so it would seem that timber and malt are same, but with malt, you do not need to worry about, about strength, twisting, warping and all the specifics required for different timber types. With the whiskey making processes each will have their own bit of magic at the different steps.

[MannyCalavera]

G'day to you.

Got a couple of ancient books on heat pumps now (D.A. Reay) - very helpful.

I have now realised what would be the most efficient thing to do (i think). Use the hot water at 50C directly as the heat source. Forget about heating glycol with water.

Set a design target of achieving a glycol temperature of 70 degC. On this basis aim for a condensing temperature of 80 degC (gives a decent sized dT). 70 degC glycol will give pretty close to 60 degC air temperatures (and this is the desired starting temp, do not want higher to begin with).

Select an evaporating temperature (remove X degC from the hot water) giving a sufficent dT based on the amount of cooling which is necessary to give you sufficient Q,in + W to the glycol.

It just dawned on me that the air across the radiators will never remove a huge amount of energy from the glycol, so why bother heating with water -> then heat pump?

Glycol heated soley with heat pump to boost glycol to 70 degC would be the best thing surely. I'd expect glycol return temps to be around the 45/50 degC mark.

System would still have good control on the evaporator with hot water flow control.

Got any thoughts on this mf?

[desA]

Got a couple of ancient books on heat pumps now (D.A. Reay) - very helpful.

Dave Reay is a very helpful man, indeed.


Set a design target of achieving a glycol temperature of 70 degC. On this basis aim for a condensing temperature of 80 degC (gives a decent sized dT). 70 degC glycol will give pretty close to 60 degC air temperatures (and this is the desired starting temp, do not want higher to begin with).

Condensing temp of what refrigerant? You may want to review this aspect a little further.

You may want to look at what happens with condenser/evaporator failures if you do decide to discard secondary circuits. Perhaps I'm just reading this thread all wrong?

[MannyCalavera]

Condensing temperature of what indeed. Further study required on this part, alot of the old literature deals with refrigerants that may not be socially acceptable now :mad:

Looking at a P vs. Enthalpy chart of ammonia, 80degC is far from critical properties, but it's at 40bar (seems quite high, i gather 20-25 is more like what is aimed at).

I don't think you are reading wrong, but if the heat pump exchangers went tits up, the primary heating medium (steam radiators) would ramp up to cope. Would they be able to cope with heating (the now)cooler air to 60+ degC? Very good point. They might struggle without pre-heating.

However, the system can always revert back to it's old recovery (which uses the exhaust humid air to heat the glycol a wee bit). This just requires the flick of a switch / control valves (so to speak).

Any thoughts on refrigerants for this temperature range appreciated.

[mad fridgie]

The glycol loop is more than likely there for practical reasons, how cold does it get? what happens to your air heat exchanger, if there is no heat flow. (kiln is off).
Of course direct heating is better than indirect! to be determined!
Earlier on I told you that your heat pump should heat the glycol directly
As far temps go they are just pie in the sky figures, until the air heating process has been determined,
The heat pump is the very last thing to sort out.
Simply what is the temperature of the fluid entering the evap?

[mad fridgie]

I have just re read you last two statements, it could be constrewed in a number of ways.
I think you are again moving away from your whole plant process.
You have water at 50C plenty of it.
What is the required temperature of the water back at the distilling process, what happens if this colder! Does this improve the distilling process.
You already have some from of heat exchange, using your exhaust air.
Understand that the compression ratio determines the efficiency of the heat pump. Very early on I asked do you require energy or temperature, again this can only be determined by knowing the air heating process, including diversity of the kilns.
As far as refrigerant choice, what will be the working conditions, whos going to fix it, are components of the shelf, is it safe, if there is a leak will taint your product, what is the capital cost.
it is very difficult to find one that will be top in all areas.

[desA]

The process specification envelope needs to be determined before a detailed engineering phase can, or should ever, begin.

Manny - first set down your hot & cold streams, load patterns in stone, then re-visit the equipment required to get there.

[MannyCalavera]

Flexability is required with cold water Fridge - the temperature required can change depending on character of spirit. > 30 is too hot, 18- 28 is normal. Therefore cooling down to 30 would be ideal, but no less than 30. There is sea water cooling facilities in place to bring it to the exact requirement.

Air flow through the radiators is 24/7 - a kiln will be "off" for say 4 hours out of a 24 period, but at that time it's paired kiln will be in full whack so air flowrate is still there.
----
Trying to guess how the radiator performs at elevated temperatures is tricky, and so its really hard to know return temps that glycol would be at. I realise this is key information to know before any design can take place.

I am wondering if it is best to assume new radiators will be purchased which can be designed to best suit the new process?

[mad fridgie]

Flexability is required with cold water Fridge - the temperature required can change depending on character of spirit. > 30 is too hot, 18- 28 is normal. Therefore cooling down to 30 would be ideal, but no less than 30 NO. There is sea water cooling facilities in place to bring it to the exact requirement.NO waste of energy (do not disconnect as may be used when ambients are very high)

Air flow through the radiators is 24/7 - a kiln will be "off" for say 4 hours out of a 24 period, but at that time it's paired kiln will be in full whack so air flowrate is still there. If your sure that the rads never drop below freezing, then water is OK direct (it can be controlled, but its risk management)
----
Trying to guess how the radiator performs at elevated temperatures is tricky, and so its really hard to know return temps that glycol would be at. I realise this is key information to know before any design can take place.Correct

I am wondering if it is best to assume new radiators will be purchased which can be designed to best suit the new process? 100%
"S" is the key

[MannyCalavera]

Excellent. Let's assume this is the case henceforth - no point installing a great heat pump system if the delivery of heat is going to be hindered by some old piece of sh*t radiator.

My next thought would be, what would be best glycol temperature for the condenser. Glycol returning hotter or colder? (depends on radiator design and glycol flow, both of which we can change). Colder would strike me as being better, more driving force for the condensation etc.

1.2 MW is the required duty to raise mean ambient air from 10 C to 60 C given the average air flowrate in one radiator. This is the key requirement of the glycol.

I am in contact with a fabricator of radiators up here, hopefully he can advise me on temperatures which would do this duty so i can get an idea of what temperatures of glycol to expect to achieve 60 C air-on.

[mad fridgie]

Still moving a bit quick,
so now descibe your process (no numbers) keep it simple
I will start; we have 2 energy streams,"a' hot energy coming from the dist. process and "b"low grade heat from exhaust recovery
>
>
>

End desired results

[MannyCalavera]

- two energy streams, hot dist water and exhaust kiln air - exhaust air currently vented with no recovery - dist water currently used to heat glycol circuits - cooled dist water returned to cooling towers -------- - 6 kilns in question - operate in pairs - numbered 3/4, 5/6, 7/8 - 3/4 twice the capacity of 5-8 - air flowrate always fairly constant - 24 kiln cycle per kiln - first 12 hours air is 100% wet, final 12hrs moisture content reduces - whilst 1 kiln is in first 12hrs, paired kiln will be in second 12hrs - this kiln will recirculate 100% air to its paired kiln each pair has 1 pre-haeting radiator - glycol passes through radiator pre-heating kiln air - cooled glycol returned to hot water PHEs - 3 pre-heating radiators = 3 glycol circuits = 3 PHEs - glycol circuit can switch between exhaust air & hot water as a heat source - hot water almost always used preferentially - exhaust air not utilised ------------ heat pump - use hot water as a heat source - use a heat pump to delivery energy and raise the temperature of the 3 glycol circuits - increase pre-heated air to a temp of 60 degC pro's - savings on annual HFO - CO2 considerations a positive - cooling condenser water saving on cooling costs v.s. - capital / maintenance / operating cost of heat pump

[MannyCalavera]

apologies for format, copy paste didnt work :rolleyes:

[MannyCalavera]

Finally got Visio working - this schematic should make it clear what is going on

[MannyCalavera]

Basis:

1.25 MW heating required for: 10 C - > 60 C air at average 25 kg/s

Glycol delivering this duty at given flowrate: delta T = 23 C

1.25MW per radiator (3/4 = 5/6/7/8 air flow)
1.25 * 4 = 5 MW total required heat input

Hot water source in evaporator: 50 deg C
Evaporator drops it to : 32 deg C

18 degC delta T at average flow = 4.5 MW

Glyol heated to 70 degC in condenser
Drop over radiator = 23, so 47 deg C entering condenser

Air flow measurements required for each pair of kilns, profile over a 24 hour period can then be built up to give a load profile on the radiators.

So

evaporator inlet temp = 50 deg C
4.5 MW removed, outlet temp = 30 degC
Condenser temperature = 70 degC
Inlet to condenser = 47 degC
Current pre-heating: 10 - 37 degC
Improved pre-heating: 10 - 60 degC
CoP anyone?

[desA]

Suggestion - put your process information directly on the process diagram. It will be easier for us to make sense of your logic.

[MannyCalavera]

Good call desa.

Ignore the 75 C.

Obviously glycol flowrate can be altered, which will change the temperature it enters the condenser at. (i don't know whether cooler or hotter entering temp is best)

70-80 deg pre-heating required for radiators to be able to bring air up to 60 C. (fairly realisitc assumption based on similar radiators at different maltings).

Any input appreciated. Regards

[r.bartlett]

Did this die a death??

[desA]

I suspect that the hooch ran low. :D

[DTLarca]

Sheesh - I've read up to post #26 which for me is half way down the first page of 4.

Thought I'd take a break to watch a movie - however - my first thoughts here are multiple R134a screw circuits, thats if you want to use standard copper piping components, you would then go for circuits of about 500kW each to avoid exceeding the elastic limit of the size piping implied.

[mad fridgie]

In short, these plants can save copuse amounts of energy, but to to give maximum benefit, a full understanding of the factory processes is required. In my opinion this not just about heatpumps.
I would say for you boys in the UK there is a great oppotunity to make a new market but you will need to go with an open mind. Apparently there are 2000 of these plants in Scotland, all using a very similar system.