AWHP superheat & sub-cooling

[desA]

On some systems, the superheat at the compressor inlet can be considerably higher than at the TXV bulb, causing the compressor to run hot. Liquid is injected into the suction line, or sometimes directly into the compressor, to reduce the superheat and cool the compressor.

Thanks, Gary.

How effective are these injection systems over a large operating envelope, as, say, for a heat-pump?

[desA]

http://www.ra.danfoss.com/TechnicalI...Subcooling.pdf

An interesting article by Danfoss on the influence of subcooling on the system.

I'd be interested to hear your comments on their view.

[Gary]

Mmhh... an interesting point. That's got me thinking.

How would the frequency affect operation of the compressor? Would this be in terms of motor heat? At 60Hz, does the compressor rpm change, compared to 50Hz?

Very interesting...

With the same number of poles, a 60hz motor runs 20% faster.

[desA]

^ You may be onto something here. Very, very interesting. Brilliant.

In your experience, what could be the influence of the SAME compressor running at 50/60=83.3% of the US designed speed?

[Gary]

http://www.ra.danfoss.com/TechnicalI...Subcooling.pdf

An interesting article by Danfoss on the influence of subcooling on the system.

I'd be interested to hear your comments on their view.

I didn't find it all that interesting.

There is an old saying, "Dazzle 'em with brilliance or baffle 'em with b*llsh*t"... lol

They make simple concepts difficult to understand, because if they said it in plain language it wouldn't be all that impressive.

[desA]

^

I'd say that the take-away values for me, from the article were:

1. The recommended evaporator inlet quality condition should be in the range 0.2 < x < 0.3 .
2. Sub-cooling should be set in order to make (1) possible.
3. "Under constant marginal conditions, the compressor capacity increases when subcooling increases".

These statements should be debated to see whether they hold water, or not, in terms of an AWHP, where 0.2 < x < 0.4 under reasonable subcooling.

[Gary]

^ You may be onto something here. Very, very interesting. Brilliant.

In your experience, what could be the influence of the SAME compressor running at 50/60=83.3% of the US designed speed?

Given the same Te,sat and the same Tc,sat, less RPM gives you less mass flow, which reduces system capacity.

[Gary]

^

I'd say that the take-away values for me, from the article were:

1. The recommended evaporator inlet quality condition should be in the range 0.2 < x < 0.3 .
2. Sub-cooling should be set in order to make (1) possible.
3. "Under constant marginal conditions, the compressor capacity increases when subcooling increases".

These statements should be debated to see whether they hold water, or not, in terms of an AWHP, where 0.2 < x < 0.4 under reasonable subcooling.

Increasing subcooling increases mass flow. If the heat load is marginally low, the valve may hunt. Well... yeah. The flow is too much for the load.

That's what smaller orifices are for... or bigger fans.

[desA]

Originally Posted by desA
^ You may be onto something here. Very, very interesting. Brilliant.

In your experience, what could be the influence of the SAME compressor running at 50/60=83.3% of the US designed speed?

Gary:
Given the same Te,sat and the same Tc,sat, less RPM gives you less mass flow, which reduces system capacity.

Ok.

Now, with less refrigerant mass-flow, what would be the effect on the compressor internal heating, friction, motor heat, that may allow Te,sat & Tc,sat to be pushed up slightly higher (from the US baseline of Te,sat=13'C, Te,sat=65.6'C)?

The answer to this could provide the link to the 60Hz versus 50Hz theory.

[desA]

Hmmm... the transfer would cool the liquid line, but that heat is transferred right back into the refrigerant on the other side of the TXV, to be carried through the evap. I don't see where there would be any gain there. Perhaps some by eliminating flashing.

Placing the VIC at the evap outlet transfers the heat from liquid to suction without carrying that heat through the evaporator.

Checked the second option:
VIC -> evap

There is a very narrow band at extremely high dT,liq where this is feasible, but, outside that, it increases the evap duty.

Scrapped...

I'm now back to trying to figure out why the first option:
evap -> VIC

is giving me such a modeling headache. Bear with me on this, I've only done around 1000 simulation runs so far... lol

[desA]

VIC feedback

The answer for the instability boils down to the following:
1. Cp,hp = 182036.5 J/kmol.K
2. Cp,lp,v = 84538.5 J/kmol.K
3. hfg,lp = 1.987e7 J/kmol

The huge differences between the latent heat of evaporation (hfg,lp) & the various specific heats (Cp) is causing all kinds of local numeric unhappiness in the simulator. I'll continue working on it.

[Gary]

Ok.

Now, with less refrigerant mass-flow, what would be the effect on the compressor internal heating, friction, motor heat, that may allow Te,sat & Tc,sat to be pushed up slightly higher (from the US baseline of Te,sat=13'C, Te,sat=65.6'C)?

The answer to this could provide the link to the 60Hz versus 50Hz theory.

I think the key is the mass flow. In order for the 50hz compressor to pump the same mass flow (same compressor load) as the 60hz compressor, the 50hz model would have to increase Te,sat and Tc,sat.

I think we have found the difference between the baselines, in which case we can comfortably raise our Te,sat to 15C.

[Gary]

An example:
Take an establishment which only uses most off its water (21 m3) during a peak time of 19h00 - 21h30. Hot water storage tanks are extremely pricey in this region - more than the heat-pumps, for instance.

Current system as follows:
1. Small inventory of hot water, via small fuel-fired boiler - pump-around. Start of cycle 30'C - end ~65'C - hold when hot. Enough for off peak usage.
2. At peak usage, hot water storage not enough, so heat mains water directly from 30'C to 65'C on the fly - for use during peak time only.
3. Cost of fuel incredibly high, need to go to alternative energy source.

Proposal (by consultant):
4. To install a heat-pump in either of two configurations:
4.1 Direct heating of small volume flow of water, through heat-pump (direct), into hot-water storage tanks - sufficient for peak usage. Heat-pumps operate off-peak, taking advantage of off-peak electrical rates. Fuel-boiler on make-up call if water storage begins to run short.
4.2 Pump-around heating of larger volume flow of water, through heat-pump, into hot-water storage tanks - sufficient for peak usage. Heat-pumps operate off-peak, taking advantage of off-peak electrical rates. Fuel-boiler on make-up call if water storage begins to run short.

Should we elect to go the direct heating route, ideally the supply of fully heated water would be sufficient to keep the tank level from dropping too low during periods of peak water usage.

But if it is not sufficient to maintain a minimum tank level, we need a means of bypassing the water regulating valve to provide full flow through the condenser. Perhaps a dual float switch and bypass solenoid.

As the tank level drops, the first float switch turns on the AWHP in feedwater heating mode. Should the level drop further, the bypass is energized by the second float switch, bringing the level back up (with partially heated water) and then de-energizing the bypass.

[desA]

Originally Posted by desA
Ok.

Now, with less refrigerant mass-flow, what would be the effect on the compressor internal heating, friction, motor heat, that may allow Te,sat & Tc,sat to be pushed up slightly higher (from the US baseline of Te,sat=13'C, Te,sat=65.6'C)?

The answer to this could provide the link to the 60Hz versus 50Hz theory.

Gary:
I think the key is the mass flow. In order for the 50hz compressor to pump the same mass flow (same compressor load) as the 60hz compressor, the 50hz model would have to increase Te,sat and Tc,sat.

I think we have found the difference between the baselines.

Now this is making sense. I'm going to have to get my head around how to model this, to let the simulator determine these values. Out come the thermo books...

Now we're cooking with gas...

[desA]

Should we elect to go the direct heating route, ideally the supply of fully heated water would be sufficient to keep the tank level from dropping too low during periods of peak water usage.

But if it is not sufficient to maintain a minimum tank level, we need a means of bypassing the water regulating valve to provide full flow through the condenser. Perhaps a dual float switch and bypass solenoid.

As the tank level drops, the first float switch turns on the AWHP in feedwater heating mode. Should the level drop further, the bypass is energized by the second float switch, bringing the level back up (with partially heated water) and then de-energizing the bypass.

Wouldn't the cold water then mix with the hot water & we'd end up with similar results to the pump-round route? By pump-around route, I mean tank->pump->hp->tank, not the external hot water loop (Grunfos-style).

[Gary]

Wouldn't the cold water then mix with the hot water & we'd end up with similar results to the pump-round route?

No... it would be warm water mixing with hot, not cold water mixing with hot... and only enough to maintain a minimum level. After the peak usage the tank would gradually be filled, then if needed the system would go from feedwater heating mode to maintaining tank temp mode.

By pump-around route, I mean tank->pump->hp->tank, not the external hot water loop (Grunfos-style).

The circulating loop shown in the Grunfos diagram is a side issue here. Its sole purpose is to maintain hot water in the lines so that when you turn on the tap you almost immediately get hot water as opposed to waiting for the hot water to come down from the roof.

On the plus side, the hot water mains become part of the hot water storage system. In effect it is like having a larger storage tank.

It should also be kept in mind that while the user is waiting for hot water to arrive from the roof he is wasting water down the drain, water which must be replaced and heated.

If the circulating pump were used, I would recommend that it be integrated into our system, perhaps becoming the tank temp maintenance pump.

[desA]

But if it is not sufficient to maintain a minimum tank level, we need a means of bypassing the water regulating valve to provide full flow through the condenser. Perhaps a dual float switch and bypass solenoid.

As the tank level drops, the first float switch turns on the AWHP in feedwater heating mode. Should the level drop further, the bypass is energized by the second float switch, bringing the level back up (with partially heated water) and then de-energizing the bypass.

Went back & re-read what you said. Apologies, I misread.

So, basically, there would be two water speeds through the condenser - the slower one being that which maintains Tc,sat=75'C, as heating mode.

The second allowing the water flow to speed up through the condenser, as water requirements call for it (dropping tank level). This would then pull down Tc,sat a little, off peak.

Good point.

[Gary]

I'm now back to trying to figure out why the first option:
evap -> VIC is giving me such a modeling headache.

Bear with me on this, I've only done around 1000 simulation runs so far... lol

Since the superheat at the bulb remains the same, the only real change is the drop in liquid temp. The effect should be the same as having an independent subcooler in the liquid line.

[desA]

Originally Posted by desA
Ok.

Now, with less refrigerant mass-flow, what would be the effect on the compressor internal heating, friction, motor heat, that may allow Te,sat & Tc,sat to be pushed up slightly higher (from the US baseline of Te,sat=13'C, Te,sat=65.6'C)?

The answer to this could provide the link to the 60Hz versus 50Hz theory.

Gary:
I think the key is the mass flow. In order for the 50hz compressor to pump the same mass flow (same compressor load) as the 60hz compressor, the 50hz model would have to increase Te,sat and Tc,sat.

I think we have found the difference between the baselines.

desA:
Now this is making sense. I'm going to have to get my head around how to model this, to let the simulator determine these values. Out come the thermo books...

Now we're cooking with gas...

I've been doing some calculations based on the 50 Hz versus 60 Hz assumption. I have a little cleaning up to do in my calcs, but it is looking very, very likely that the frequency is responsible for the change in performance specs.

Now, if this is the case, & these systems are indeed running at 50 Hz, whilst tests were done at 60 Hz - why on earth have the manufacturer's technical staff not picked up on this? Is this a case where the technical staff are glorified salesmen who actually don't have a clue about what they are spouting?

Next, what sense can we make of this in our developments?

[Gary]

Next, what sense can we make of this in our developments?

We can use Te,sat=15C and Tc,sat=75C.

[desA]

Originally Posted by desA
Next, what sense can we make of this in our developments?

Gary:
We can use Te,sat=15C and Tc,sat=75C.

Agreed. Boy, I'm relieved.

Now - how on earth would a warranty claim be handled in the event of a commercial application? I'll bet when it's submitted for checking on warranty claim, the compressor technical staff will NEVER understand the frequency difference 50Hz versus 60 Hz, rpm at 2900 instead of 3500, power nominal 6700W versus 8250W etc. They would most likely always fall back on the safest option for them.

In cases like this, does the OEM heat-pump supplier, for instance, state that the unit was operating as per recommended operating values, but let the operator actually have the option to move the settings up to take advantage of the Te,sat=15'C, Tc,sat=75'C relaxation?

I doubt that the technical staff at these huge compressor operations will ever have the intellect, or desire, to solve this one.

[desA]

What I may very well end up doing is developing a position paper on this aspect & having a long, long chat to the regional technical representative for the compressor outfit at the upcoming trade show we have here in BKK next month.

This could be the subject of a nice talk... Seems I've been given some technical talk time on this show, perhaps it's good time to raise some healthy debate...

[mad fridgie]

MOP valves, act similar to a CPR, at little extra cost, I do not ever remember having any problems, another option with out the need for valving is to control the evap fan, cycle on and off (cheap option and proven) many methods of control.
Double port valve, made by Sporlan I believe, has to internal expansion ports, 1 small and 1 large, if super heat is higher both ports are open as the super heat drops then the large shuts first and controls on the small. Ideal for large variations in TXV pressure differential.
50-60Hz, 20% more capacity on 60Hz, thus more work undertaken by the motor!
As a designer of similar products, my advice is focus on what the end goal is "heating water"

[desA]

MOP valves, act similar to a CPR, at little extra cost, I do not ever remember having any problems,

For your MOP option, can I ask you your experiences on the following:
1. Spares availability (international);
2. Control of pressure, using a temperature signal (MOP bulb);
3. Delays due to tube wall, bulb response, TXV response.

another option with out the need for valving is to control the evap fan, cycle on and off (cheap option and proven) many methods of control.
Double port valve, made by Sporlan I believe, has to internal expansion ports, 1 small and 1 large, if super heat is higher both ports are open as the super heat drops then the large shuts first and controls on the small. Ideal for large variations in TXV pressure differential.

Many thanks for this. I do see some folks use the on/off option.

I must say though, the heat-pump response to fan is fairly quick - 30-60 sec. The fan is surely going to be cycling a fair bit on/off, to accurately manage the Te,sat drift towards the end of the heating cycle?

50-60Hz, 20% more capacity on 60Hz, thus more work undertaken by the motor!

Also, more motor heat.

As a designer of similar products, my advice is focus on what the end goal is "heating water"

Yes, it is - of course - but safely.

Can you comment on what you observe for the Te,sat over the course of a complete heating cycle? I'd imagine you are in NZ. What ambient temps do you experience over there? Have you worked at 35-40'C ambients? I'd love to hear your experiences.

[mad fridgie]

I presume you are designing a domestic unit, if so, then yoiu need to accept that within monentry constraints that ideal engineering goes out of the window! Difficult I know, but I also had to accept this.
If money is not an issue, then balancing the system becomes easy, speed control all motors, (comp, fan and pump) use electronic expansion valve.
My machine has been tested from
-10C to 43C, and water upto 98C (pushing it a bit though) normally 75C

[mad fridgie]

my favorite saying
"engineering is easy, doing for a price is hard"

[mad fridgie]

Also should remind you that a lot of the world requires a vented double barrier heat exchanger, and that causes its own problems

[desA]

I presume you are designing a domestic unit, if so, then yoiu need to accept that within monentry constraints that ideal engineering goes out of the window! Difficult I know, but I also had to accept this.

I will not prejudice system safety for cost-cutting. If system conditions force the compressor into compromise territory, then it is suicide.

I have seen this philosophy used by another colleague. Ended up in a something like +20% failure in the field. This is madness.

If money is not an issue, then balancing the system becomes easy, speed control all motors, (comp, fan and pump) use electronic expansion valve.

There are far simpler ways to do the same - mechanically. I hate electronics - terribly unreliable stuff, to be honest - especially in humid, hot climates.

My machine has been tested from
-10C to 43C, and water upto 98C (pushing it a bit though) normally 75C

Either you're using a high Tc,sat compressor, or you will face problems in the field. This is not something you can do on a domestic budget, I'm afraid. I'd discount that technical claim, to be blunt.

What Tc,sat maximum are you working to?

[desA]

Also should remind you that a lot of the world requires a vented double barrier heat exchanger, and that causes its own problems

There are better ways to manage the sanitary requirement, than a vented HE.

[mad fridgie]

I can agree long of the equipment is very important, I perhaps should of made it clear, that system stability and efficiency can be comprimised to ensure longevity, without the need to make a machine uncompetative
Re my machine, different fundementals used, to achieve desired results.
Not using special compressor!

[mad fridgie]

regs indicate a double barrier that is vented (does not need to be a single hx) A santitry coil in a cylinder is another method, but you always have to a delta T between to achieve required temps

[desA]

I can agree long of the equipment is very important, I perhaps should of made it clear, that system stability and efficiency can be comprimised to ensure longevity, without the need to make a machine uncompetative

Sometimes, sometimes not...

Re my machine, different fundementals used, to achieve desired results.
Not using special compressor!

You will not get 98'C water out of a standard, dometic-variant compressor, running within manufacturer's design limits - even if you use de-superheaters. Well, not for long, that is. You will be exceeding the safe limits of the compressor envelope. You can't beat thermodynamics.

The alternative is to inject electric heating along the way, or light a fire under it...

[desA]

regs indicate a double barrier that is vented (does not need to be a single hx)

A santitry coil in a cylinder is another method, but you always have to a delta T between to achieve required temps

True...

[desA]

We can use Te,sat=15C and Tc,sat=75C.

We continue along this track, then.

[mad fridgie]

I did state that 98C was pushing it a bit. No fire, no heater, there is more than one type of refrigeration!

[desA]

@ Gary:

CPR versus MOP

Can we explore the control methodology & merits of both schemes, as applied to a Te,sat value which gradually drifts upwards over the course of a water-heating cycle.

Te,sat can start at say 10'C on some systems & then drift upwards towards +19'C. For others, it can start at Te,sat=12.5'C, or even Te,sat=15'C then drift upwards.

With Te,sat control, we're surely controlling pressure, not temperature perse'.

How would MOP control interfere with the evap superheat control function?

[desA]

I did state that 98C was pushing it a bit. No fire, no heater, there is more than one type of refrigeration!

On a domestic budget? Ok, then, we'll let it go at that. Thanks for your kind contribution to the thread.

All the very best with your developments.

[mad fridgie]

MOP, reduces flow into the evap, thus increasing superheat, CPR restricts flow out of the evap basically flooding the evap, thus reducing superheat.
CPR superheat at the compressor can be calculated based upon the pressure drop across the CPR. Both stop control of evap superheat!

[Gary]

MOP, reduces flow into the evap, thus increasing superheat, CPR restricts flow out of the evap basically flooding the evap, thus reducing superheat.
CPR superheat at the compressor can be calculated based upon the pressure drop across the CPR. Both stop control of evap superheat!

I agree with all but the last. The CPR does not stop the control of the evap superheat.

[desA]

So, to partially summarise:

MOP, reduces flow into the evap, thus increasing superheat,

CPR restricts flow out of the evap basically flooding the evap, thus reducing superheat.

CPR superheat at the compressor can be calculated based upon the pressure drop across the CPR.

Is this a fair reflection of the MOP, versus CPR operational comparison?

Question 1:
1. How does the MOP reduce flow into the evap?

Question 2:
2. How does CPR restrict flow out of the evap? Based on what signal?
Answer 2 (please adjust):
Crankcase suction pressure pulls CPR valve open, against a curve - allows flow through according to valve characteristic at setpoint pressure.

Question 3:
3. What about evap pressure controllers? Could these assist in controlling drifting Te,sat?

[Gary]

The CPR will not flood the evap, although the superheat at the compressor inlet may drop a little. Essentially, there is very little, if any, change in superheat.

[Gary]

The CPR will not flood the evap, although the superheat at the compressor inlet may drop a little. Essentially, there is very little, if any, change in superheat.

On further thought, the superheat at the compressor may actually rise a little.

[Gary]

Question 1:
1. How does the MOP reduce flow into the evap?

It doesn't actually reduce the flow, it limits the flow, this much and no more.

Question 2:
2. How does CPR restrict flow out of the evap? Based on what signal?
Answer 2 (please adjust):
Crankcase suction pressure pulls CPR valve open, against a curve - allows flow through according to valve characteristic at setpoint pressure.

Okay... although again, the flow is limited, not restricted.

Question 3:
3. What about evap pressure controllers? Could these assist in controlling drifting Te,sat?

I don't understand the question. What sort of controllers did you have in mind?

[desA]

^ Thanks, Gary, for the above 3 posts.

So, essentially, the CPR & MOP are both restricting (not controlling) flow, not pressure, at a pre-selected cut-off point. Is this correct?

[desA]

Originally Posted by desA
Question 3:
3. What about evap pressure controllers? Could these assist in controlling drifting Te,sat?

Gary:
I don't understand the question. What sort of controllers did you have in mind?

Ok, what I understand is that the CPR restricts suction flow into the compressor, hence we see a small pressure drop, with associated temperature change. This works to control the compressor end.

Danfoss & others also make a range of devices, which work on controlling the evap end.

Example: Danfoss type KVP ; PKV/PKVS ; KVQ

The writeup for the KVP states:
" KVP evaporator pressure regulators are mounted in the suction line of refrigeration & airconditioning systems. They are used to maintain a constant pressure corresponding to a constant temperature on the evaporator."

The valve characterisitc is opposite to the KVL, in that capacity RISES from a setpoint (0%) towards 100% after P-band.

[Gary]

Ok, what I understand is that the CPR restricts suction flow into the compressor, hence we see a small pressure drop, with associated temperature change. This works to control the compressor end.

Danfoss & others also make a range of devices, which work on controlling the evap end.

Example: Danfoss type KVP ; PKV/PKVS ; KVQ

The writeup for the KVP states:
" KVP evaporator pressure regulators are mounted in the suction line of refrigeration & airconditioning systems. They are used to maintain a constant pressure corresponding to a constant temperature on the evaporator."

The valve characterisitc is opposite to the KVL, in that capacity RISES from a setpoint (0%) towards 100% after P-band.

An EPR valve does not allow the Te,sat to drop below setpoint pressure. The exact opposite of what we are trying to accomplish.

[Gary]

^ Thanks, Gary, for the above 3 posts.

So, essentially, the CPR & MOP are both restricting (not controlling) flow, not pressure, at a pre-selected cut-off point. Is this correct?

They both limit flow to maintain a pre-selected pressure, the MOP according to evap pressure and the CPR according to compressor inlet pressure.

[desA]

^ Thanks for clearing that up.

[Gary]

Done any test runs lately?

[desA]

I'll get back into the tests starting Monday. I needed to clear up an office backlog & needed to settle some technical questions. Basically there, now.