De-superheater design.

[R717Kid]

Does anyone have advice or learned lessons from installing a de-superheater? We are looking at installing a plate and frame heat exchanger in the discharge header of a system that uses only reciprocating compressors to capture heat for another process.

More specifics available if needed, but looking for general feedback.

Thanks,

[Sandro Baptista]

Will the de-superheater be above or below the inlet of the condenser?

[Sandro Baptista]

Is it's below a small vessel should be applied downstream the heat exchanger to colect the liquid condensate and the vapor will continue its own path until the condenser. The liquid collected at the bottom of that small vessel should be redirected to the the main liquid receiver.

Theoretically non liquid condensate should occur for the design conditions. But when the water is colder or the plant is at low capacity that could happen.

I hope this would give you an idea.

[RANGER1]

Without knowing specifics it doesn'treally save any energy or money .

Mainly because of pressure drop penalty on discharge line & large water pump required which costs money to run anyway .

[Sandro Baptista]

How large the recirculating water pump will be it will depend on the recovery heat you want and on the plate heat exchanger design.

[mad fridgie]

Without knowing specifics it doesn'treally save any energy or money .

Mainly because of pressure drop penalty on discharge line & large water pump required which costs money to run anyway .

You are right and wrong???
Most "F"in designers who have installed them have total mis understood the process (I see this all the time on large and small system)
Simply thay have designed these on a steady state heat transfer method. Cold water warmer water out, pressure drop on the refrigerant side at this is normally acceptable, but most heat recovery is a multipass system the water get entering gets warmer and warmer, reducing heat transfer and increasing the pressure drop.
De-superheaters should be sized on refrigerant pressure drop with "ZERO" flow of water.
Yes is bigger and more expensive.
The water side, you can size on velocity, to reduce the chance of fouling. The water pressure drop will also be low. so pump need to move large volume with little head.

[R717Kid]

The heat exchanger is being installed lower than the inlet of the condensers. In fact, in the engine room within ten feet of the HPR. The engineers don't count on condensate being produced, however, my belief is that if it is possible, it will happen. That said, I am asking for a small vessel to collect the "possible" condensate and implement a high side float.

The engineers are planning to install a 3 way valve (expecting low load conditions) to bypass the exchanger. And it is a small exchanger.

The cold side is heating glycol (closed loop) which in turn heats waste water prior to a treatment facility. Pump speed will be controlled via vfd's based on the glycol TD.

All of these variables reinforces my concerns for condesate formation.

[Sandro Baptista]

There is no problem at all if there is condensations of vapour R717. Don't see that is necessary a high side float because the small liquid vessel will be connected to the liquid receiver and discharge the liquid for this one.

The installation of the 3 ways valves it's just to isolate the heat exchanger for the case of service and the third one will bypass the heat exchanger (so nothing as to do with "expecting low load conditions").

[Sandro Baptista]

I'm sorry...I fargot, you should also count with a fourth valve lo isolate the discharge liquid line from the small liquid vessel

[R717Kid]

There is no problem at all if there is condensations of vapour R717. Don't see that is necessary a high side float because the small liquid vessel will be connected to the liquid receiver and discharge the liquid for this one.

I tend to disagree. This is not intended to be used as a condenser. Assuming the engineers didn't use the advice of the vessel to collect condensation, there would now be the possibility of vapor propelled slugs heading for the condensers...a particular concern of mine after having seen the aftermath of a similar event.

The installation of the 3 ways valves it's just to isolate the heat exchanger for the case of service and the third one will bypass the heat exchanger (so nothing as to do with "expecting low load conditions").

The 3 way valve will be computer controlled to bypass the exchanger when head pressure is under a given value. I.E. low load conditions. Important when production is down and ambient conditions are 0 degrees F, and the discharge from the only compressor running never exceeds 150 degrees F during those conditions. Again, this is not to be used as a condenser.

[Sandro Baptista]

I'm sorry I'm not understanding. Something is missing
You can and you should control the condensing pressure of the refrigerant plant actuating on the condenser not on the de-superheating. The de-superheating can represent about 10% to 30% of the total heat rejected...so unless you have sometimes water too much cold and with only...lets say about +15ºC and 10% (it depends on the design of the exchanger) of the total refrigerant capacity running only in this case it could justify that. But I think you want hot water so that problems of too low head never should happens.

[R717Kid]

Sandro,

No worries. The head pressure will be controlled via condenser controlls. That is, condenser fans and pumps will stage off at lower head pressures.

There are times, depending on what is happening with production, that colder temperatures can be introduced into the waste water and therefore into the cold side of the heat exchanger. This is not frequent, but does occur.

I guess all in all I am looking for past experiences and what not to do or tips for success.

Thanks,

[RANGER1]

You are right and wrong???
Most "F"in designers who have installed them have total mis understood the process (I see this all the time on large and small system)
Simply thay have designed these on a steady state heat transfer method. Cold water warmer water out, pressure drop on the refrigerant side at this is normally acceptable, but most heat recovery is a multipass system the water get entering gets warmer and warmer, reducing heat transfer and increasing the pressure drop.
De-superheaters should be sized on refrigerant pressure drop with "ZERO" flow of water.
Yes is bigger and more expensive.
The water side, you can size on velocity, to reduce the chance of fouling. The water pressure drop will also be low. so pump need to move large volume with little head.

It all sounds good but at what point do you decide its viable or not with payback,or do you do it because you have conscios about environment .
-Big heat exchanger
-water pump
-tank , pipework ,etc
-cost of electricity

Each case would have to be weighed up individually .

[mad fridgie]

It all sounds good but at what point do you decide its viable or not with payback,or do you do it because you have conscios about environment .
-Big heat exchanger
-water pump
-tank , pipework ,etc
-cost of electricity

Each case would have to be weighed up individually ,but if this is your core business I guess every case is good .

You are correct it has be viable, the first thing you need is ensure is that there is sufficent load (or use for the hot water) which again many forget.
I tend to say any thing with under 2 year paypack, is a real goer and should be, 2-5 years really depends upon the client and how they view their long term business, over 5 years basically is a no go.
if you look at 4 year payback, you are getting a 25% return on investment, what do you get from the bank?

[Magoo]

Generally I work on a 15>20% heat recovery of total heat of rejection, any more and the system balance goes out the window and all hell breaks loose.
Talk to and listen closely to mad fridge, he has been there done that and perfected it.

[R717Kid]

ROI is 2 years. Load is sufficient on the secondary side as well as consistant. The premise for using heat from this area is to offset the cost of fossil fuel required to heat the water due to plant production increase.

Again, just looking for do's and dont's on this particular topic.

[Magoo]

Then definitly talk to mad fridge, he has a system that will pre-heat boiler make-up water and save truck loads of dollars.
Basically recovering energy for cooling and re-cycling it back into your process, say like you pay to heat it then you pay to cool it, remove the utility companies costs between.

[mad fridgie]

Do you have a desired temp and kw rating. Which is of greatest importance.
Have you calculated the NH3 pressure drop, at max refrig load and minimum heat load.
What effect does this have on the refrigeration running cost (direct and indirect)
Are you sure that you heating load is constant, do not take a diversivied average, for heat exchanger selection (OK for calculating savings)
You can install a diff pressure relief valve, as a precaution.
Cheers Magoo, have a doughnut.
Ranger "yes" water heating and heat recovery utalising refrigeration is my primary business, but certainally not, are all potential jobs a goer. On average my predicted savings are 6% under what the actual savings are.
The first thing I always do is look at the recovered application, it has to make at least some form of fiscal sense.

[RANGER1]

If it is PHE you would not want gaskets but fully welded because of high temps .

[R717Kid]

Yes, higher temps and I am also thinking that cyclic pressures associated to reciprocating compressors can come into play as well.

As far as size, I'll get all the detailed information today for review.

[Sandro Baptista]

If it is PHE you would not want gaskets but fully welded because of high temps .

My opinion is that is not necessary. Semi welded plate exchanger is enough even for recips where the discharge temperatures are higher that for screws.