Clean Condenser Running too hot??

[Gary]

Yes this line. A whole is drilled/burned into the line at the higher level and it pulls from here. I think that this tube is pulling from around the middle of the line (think of the cross section of the pipe) and it is flashing. So that is one of the reasons I think there is a gas lock in this line. Usually this interstage line is connected to the liquid line (after the reciever in my case)

I agree. Not a gas lock, but possibly vapor feeding the interstage TXV.

What is the interstage superheat?

[Gary]

The water in is 78F and the out is 90F. The condensers are getting at least 40gpm each.

Hmmm... it occurs to me that with SCT of 122F and approach of 32F, there should not be enough heat transfer to get this dT of 12F.

I recall there is a formula for calculating BTU output given dT and GPM. Anyone have that formula handy?

I'm thinking that the actual water flow may not in fact be 40GPM or even close, or the dT would be much lower given the high approach.

[Gary]

I'm wondering if this could be a case of 60hz pump motors running on 50hz power supply... or the pumps rotating in the wrong direction... or undersized piping... or reduced flow through valves/fittings?

[Gary]

Correct me if I'm wrong here:

BTU = GPM x dT x 500

40 x 12 x 500 = 240,000 BTU

So... is this system moving enough BTU to do the job?... OR is the GPM considerably less than 40?

[Chseybold]

The flow is ok, the minimum flow for the cooling tower water distributer to rotate is 80gpm. It rotates and the pump is sized ok. 60hz power

[Chseybold]

Also if there wasn't enough water flow, wouldn't the water temp differential be much higher then 10?

[Chseybold]

Unfortunately the client is getting very nervous as this system is not working properly. Finally upon doing some research I think the problem is with the liquid seal on the bottom of the condenser, I posted another question about this in the fundamentals section. If there isn't a proper seal no matter what I do I won't be able to subcool the refrigerant...

[Chseybold]

and supply liquid without flash gas in it. Thanks for all the help by the way. I think I might just take a crap shoot and send him another condenser to test.

[Gary]

Also if there wasn't enough water flow, wouldn't the water temp differential be much higher then 10?

Yes it would... if the heat were getting transferred to the water.

You stated that trying to clear the sightglass causes to SCT to rise. This would transfer more heat to the water, which would give us a higher differential, which would seem to confirm insufficient water flow.

Given a clear sightglass, perhaps we could look at all of the temps/pressures and figure out the underlying problems.

I would add refrigerant until the subcooling is 15F, then take an entire new set of pressure and temperature measurements.

[Peter_1]

Just joining this thread: I should recover all the refrigerant and vacuum the whole system. This is a job for only some hours.

You're then sure having no non-condensables (NC) in the system.

Check temperature with pressure of your recovered refrigerant and determine if refrigerant was eventually contaminated with NC.

Recharge the refrigerant in a liquid phase and leave some lbs in the bottle to be sure that eventual NC remains in the bottle. Calculate roughly how much you may charge in the system.

I doubt the condenser is fault, you should have then 2 faulty condensers.

[Chseybold]

Given a clear sightglass, perhaps we could look at all of the temps/pressures and figure out the underlying problems.

This is one of the problems, the sight glass can't clear!! Not even with an overcharge!?

[Chseybold]

Peter the problem is that there are not 2 but 4 of the same condensers!! So the only thing I can think is that they are defective or piped wrong (which I do not think is the problem) I'll tell them to test for non-condensables again, because that is a strong possibility but due to how many of the machines there are you would think that at least 1 of the 4 would work!

[Gary]

Let's look at this from a different point of view:

40GPM x 10dT x 500 = 200,000BTU

If we assume that the pump is indeed moving 40GPM, then it is delivering 200,000BTU with the rest of system not working properly.

What if the rest of the system were working properly? Just taking a wild guess, the output would be something more like 300,000BTU... in which case we would need a 60GPM pump in order to maintain the 10F dT.

What is the heatload?

What is the capacity of the compressor?

At the current SST is it indeed outputting 200,000?

Do you have performance charts for the compressor?

[Gary]

Given a clear sightglass, perhaps we could look at all of the temps/pressures and figure out the underlying problems.

This is one of the problems, the sight glass can't clear!! Not even with an overcharge!?

Of course the sight glass can be cleared... and charging to 15F SC is NOT an overcharge.

[Gary]

Here is my current theory: The evaporator isn't working properly because there is not enough refrigerant to clear the sightglass and feed liquid to the evap.

You haven't cleared the sightglass because the head pressure goes too high.

This brings us to the big question. Why does the head pressure go too high when the sight glass is cleared? There are only a few possible answers:

A problem with the condenser.

A problem with the pump.

Both.

I would add refrigerant until the SC is 15F.

If the head pressure and dT are then excessive, I would install a 60GPM pump.

Then see what the numbers tell me about the condenser.

[Gary]

Finally upon doing some research I think the problem is with the liquid seal on the bottom of the condenser...

That's just another way of saying that there is not enough refrigerant in the system... it means the same thing. When the sight glass is clear, there is a liquid seal at the bottom of the condenser.

[Gary]

Next question: You say the head pressure goes too high when you try to clear the sightglass, but you don't tell us what the pressure went to. How high is too high?

And you say the sight glass cannot be cleared even with an overcharge. How do you know it is an overcharge? What is your definition of overcharge?

[Gary]

If there isn't a proper seal no matter what I do I won't be able to subcool the refrigerant...

This makes no sense. Can you explain your reasoning on this? Maybe I am missing something.

[Peter_1]

Which Bitzers are you using? I suppose a SGF30.2Y?
Those are giving 35 kW at -35°C. together with the absorbed power is this +/- 60 kW= +/- 204.000 BTU

The pump is giving 40gpm, is this a free flowing flow or is this with a certain back pressure?

[Peter_1]

Do you have a technical file of the used pump? flow/head pressure?

[Gary]

Please correct me if I am wrong.

As I understand it, the equalizer line and receiver have been removed.

These are major changes, so we need to know the new temps and pressures.

Without information, all we can do is offer wild guesses.

[Gary]

With incoming water at 78F:

If the dT is 10F and the approach is 32F, the SCT should be about 120F and the high side pressure should be about 310psi.

[indian]

Hiya guys have spent a little time reading your threads. May be wrong but to my mind there are possibly two problems here.

Firstly I would say that the equalization line must go. I think this may have allowed discharge gas to flow into the receiver mixing with the liquid causing flashing (was there a check valve? suspect not). Even if there was a valve in place the high water flow and big condenser with low load would allow liquid to form in the bottom of the condenser until the point where pressure forced it into the receiver along with discharge gas causing a mix in the liquid line i.e. flashing. Yet if the liquid built up in the condenser enough, as more charge was added to stop sight glass flashing, this would reduce the surface area of the condenser which would eventually cause lack of condensing and high head pressure as was noted originally. If you read second point will explain why I dont think there was a check valve on the equalisation line.
Removing the equalation line would stop this.

But then if the receiver is still in place we get a second problem. No one has mentioned head pressure control or water regulating valves. As has been said the condenser should easily handle the system duty at full load, at minimum load it is way too big with a great flow of cooling water flowing through it. Therefore the refrigerant is condensing in the condenser - as it should - but due to the uphill run it cannot get into the receiver so it fills up the condenser. Eventually the surface area reduces the system pressure builds it pushes liquid into receiver and fills it. Then once more as the system gets more refrigerant the condenser fills with liquid reducing the condenser area again and pressure builds but we now have a clear sight glass a full receiver and a shed load of refrigerant and an overcharge.

So for me get rid of equalisation line and receiver, decant and recharge just until sight glass stops flashing and possibly fit some for of head pressure control for low loads.

Possibly a complete load of rubbish but makes some sense in my strange world.

[Gary]

Hiya guys have spent a little time reading your threads. May be wrong but to my mind there are possibly two problems here.

Firstly I would say that the equalization line must go. I think this may have allowed discharge gas to flow into the receiver mixing with the liquid causing flashing (was there a check valve? suspect not). Even if there was a valve in place the high water flow and big condenser with low load would allow liquid to form in the bottom of the condenser until the point where pressure forced it into the receiver along with discharge gas causing a mix in the liquid line i.e. flashing. Yet if the liquid built up in the condenser enough, as more charge was added to stop sight glass flashing, this would reduce the surface area of the condenser which would eventually cause lack of condensing and high head pressure as was noted originally. If you read second point will explain why I dont think there was a check valve on the equalisation line.
Removing the equalation line would stop this.

But then if the receiver is still in place we get a second problem. No one has mentioned head pressure control or water regulating valves. As has been said the condenser should easily handle the system duty at full load, at minimum load it is way too big with a great flow of cooling water flowing through it. Therefore the refrigerant is condensing in the condenser - as it should - but due to the uphill run it cannot get into the receiver so it fills up the condenser. Eventually the surface area reduces the system pressure builds it pushes liquid into receiver and fills it. Then once more as the system gets more refrigerant the condenser fills with liquid reducing the condenser area again and pressure builds but we now have a clear sight glass a full receiver and a shed load of refrigerant and an overcharge.

So for me get rid of equalisation line and receiver, decant and recharge just until sight glass stops flashing and possibly fit some for of head pressure control for low loads.

Possibly a complete load of rubbish but makes some sense in my strange world.

I agree your comments on the equalizer line, but as I understand it, this line has been removed.

As to your comments on liquid flowing up to the receiver:

1 psi of differential can push a column of liquid up 2.43 feet.

When there is sufficient liquid at the bottom of the condenser to form a liquid seal, the needed pressure differential will be immediately established and if there is sufficient liquid in the condenser to maintain that seal the pressure differential will remain and the liquid will continue to flow.

The sight glass will be clear, there will be a slight difference between the condenser pressure and the receiver pressure and the liquid level in the condenser will be just enough to maintain a liquid seal. Similarly, the liquid level in the receiver will be just enough to provide a liquid seal.

Most small condensing units have a vertical receiver mounted on the same base plate as the condenser, with the liquid flowing from the bottom of the condenser up to the top of the receiver. The added pressure differential is inconsequential.

But this is a moot point, because the OP has informed me that the receiver has also been removed.

[indian]

The theory there may be fine and well dont intend to kid you that I am fluent enough in math to argue the point.

Actually totally agree about liquid locking of the condenser to receiver, have seen hundreds of small condensing units where condenser outlet below receiver and every one works - but not seen too many that had a condenser that was, what - say double the size it needed to be.

The only thing is from my experience I know that I can take a supermarket pack system which has way too much refrigerant in it to pump down normally. Yet if I take the condenser and force the fans to run full speed I can back up all the liquid into the condenser without tripping the hp until its pretty damn full of refrigerant. Have done similar with compressor receiver sets emptying the receiver into the condenser simply by closing the condensate return and running with condenser on full speed.

In doing this all I was doing was simulating a low load ie reducing need or availability of liquid to system combined with (I know some of you guys dont like the term) over condensing the liquid - or excessively sub cooling it. In this way the liquid backs up into the condenser.

Even with the receiver removed this confirms my point if we have a low load ie valve is closing down (like pump down) combined with maximum condensing ability you will backup liquid into the condenser which will eventually hit a point where the reduced surface area in the condenser until head pressure shoots up and we have a problem - over charge. Think there may even be a bit of math in there regarding the volume of refrigerant if it is excessively sub cooled think its volume may shrink as well so you can get more in.

By the way am on your side have read other posts re ODP and GWP have studied Paleoclimatology, Geology at University and all the facts as most people belleive them are not FACTS.

[Gary]

In my previous post I was discussing a system that is properly charged.

When a system is overcharged, the excess refrigerant liquid backs up into the receiver and the condenser simultaneously while still maintaining the needed pressure differential. Being that liquid is backing up into the condenser, the active surface area is reduced and the pressure does indeed rise.

Subcooling tells the story, but we don't have the temperature and pressure measurements to evaluate this system properly. We can't see it from here and we don't have the measurements, so all we can do is guess.

[Gary]

Hmmmm... just another thought:

It has been a very long time since I worked on a 2 stage system, but it seems to me that the liquid is heavily subcooled at the interstage and thus the liquid line from the interstage to the evap needs to be insulated. Please correct me if I am wrong on this.

[indian]

Did'nt think it was two stage system just a two stage compressor so stages internal within the compressor just suction and discharge to condenser - at least thats what I took it to mean.

[Gary]

I am thinking you are probably right.

Pictures might be helpful.

[Gary]

Are you making any progress with this system?

[Chseybold]

I'm sorry for not following up with you guys. Unfortunately I never had the chance to go back down to South America to do the suggestions. But I think indian described the situation in his post. I wanted to remove the receiver completely and pipe it direct. But I never got the chance as the client has financial troubles.
Thanks gents for all the good advice, wish I could have implemented them and gave more data.

[Brian_UK]

Thanks for coming back and giving the update, that's life I suppose.