Centravac Chiller has wide condenser approach 1

[Centravac]

We have a issue with one of our chillers and I am hoping that someone can shed some light on it. One of our R11 Chiller is showing low condenser leaving water temperature and high Condenser approach. Here are the specs taken yesterday.
Condenser entering water temp = 34.6 F
Condenser leaving water temp = 68.8 F
Sat Evaporator Temp = 37.8 F
Evap rfgt pressure = -7.9 psig
Saturated Condenser temp = 94.1 F
Condenser Pressure = 6.6 psig
Condenser approach = 25.5
.
This Chiller has been out of service since 08/22/07 and we put it in service last week. All other indications are normal. Purge is 0 and unit seems to be working fine. We never had an issue with micro fouling, which can also cause this condition. Any thoughts?

 

[gepman]

I am not very familiar with centrifugal chillers or R-11 but I will give you my opinion. I believe it is one of three things:

1) Condenser fouling (as you mentioned).
2) Non-condensibles in the condenser. I assume that what you meant by purge is 0 is that there are no non-condensibles.
3) Something is causing your condenser to flood (i.e. liquid is backing up and covering the heat exchange (condensing) surface).

It would be nice if you could get a reading of the liquid temperature as it comes off the condenser. Assuming that you don't have a subcooling section in your condenser the liquid temperature should be close to the SCT. Sometimes I use an IR thermometer on the trap of the condensate pipe after the condenser to get a measurement (although that is not the most accurate method).

Yorkman or imok2 probably can help on this.

 

[Centravac]

Gepman. Thanks for the feedback. When you say condenser fouling, I am assuming that you mean micro fouling (minerals, scale, algae) vs. Macro fouling (leaves, mud). I emphasize on this distinction cuz I think if its macro fouling, it will block the tubes and condenser leaving water temp will increase (same as low flow condition) and micro fouling will not restrict the flow that much and will just reduce the heat transfer and the condenser leaving water temp will go down. Am I correct ruling out blocked tubes?

 

[gepman]

I didn't understand what you meant by micro fouling but I do now and I did mean micro fouling.

Look at the heat transfer equation Q=UA delta T. Micro fouling (as you define it) would affect U. Macro fouling (as you define it) would affect A. How this affects condenser water flow depends on how your condenser water supply is set up and number of passes in your condenser. If you had a constant pressure water supply then macro fouling would increase the pressure drop across the condenser and reduce water flow overall but water to an open tube might increase somewhat. If you had a typical centrifugal pump supply then as some of the tubes became blocked the head on the pump would increase reducing overall flow but the flow at each open tube would increase (of course depending on the shape of the pump curve). If you had a constant volume supply (like a positive displacement pump then the overall flow would be the same but the flow to each tube would increase dramatically and so would the pressure drop across the condenser. You might have some type of control scheme which tries to maintain a constant leaving water temperature, a constant condensing pressure, or some other type of algorithm. You need to look at how your water flow is controlled also to help diagnose your problem.

 

[quark]

If you are sure about the cooling water inlet temperature, then the temperature difference seems to be high. Check for the flowrate, first. The pump may be malfunctioning, a corroded impeller, a valve is blocked or the valve sealing element might have losened from the spindle etc.

Scaling tends to decrease the temperature difference.

Did you make any changes in the system before putting into operation?

 

[absrbrtek]

Why a 34F delta T across the condenser? VFD? Are these temps off the panel or thermometers? What is the actual liquid line temp comming out of the condenser, not the saturated temp? Why 34F entering condenser water? You running a free cooling setup on a htex also?

 

[imok2]

The pressure/temperature relationship of the head and SCT is good so it doesn't look like non condensibles, however I believe you have fouled or contaminated tubes. Minerals, scale, mud, algae, and other impurities increase thermal resistance and degrade overall performance such as a hi condenser approach. It should be a simple matter to inspect the condenser tubes. Every plant I was responsible for I had installed a base plate and a removable inverse type swing arm reinforced with a gusset so we could hang a chain fall to remove the heads.

 

[imok2]

Quark makes a good point about condenser flow, Hoped you checked it

 

[Centravac]

Inlet Condenser temperature is correct. This is just river water-cooling used as heat sink. River water temp changes from 34 F in winter to 90 F in summer. Also, Quark. When you talk about delta T, I am assuming you are talking about delta T across the condenser. In my experience, this delta T across the condenser is low for these chillers. We induce false load in winter months by throttling the condenser entering water valve. This is done to keep the condenser pressure in band and to prevent oil loss. Other Chiller, which is running off the same line, is running closer to normal with following specs.
Condenser entering water temp = 34.6 F
Condenser leaving water temp = 78.2 F
Sat Evaporator Temp = 39.5 F
Evap rfgt pressure = -7.6 psig
Saturated Condenser temp = 91.1 F
Condenser Pressure = 5.4 psig
Condenser approach = 12.8
.
These measurements are from the microprocessor but our analog gauges that tell us about the actual condition (for example, cond pressure) confirms the above readings.
.
imok2. Regarding your comment about fouling, if the tubes were plugged due to fouling or were restricting the flow through the condenser tubes, we would see the increase in condenser leaving water temp. This is not the case. As I have stated earlier, we induce the false load on these chillers and based on my experience in the past years and the way other chillers are operating right now, condenser leaving water temperature should be higher (around 80 F). If we experienced micro fouling (scaling, minerals and algae), then our condition would make sense but never had an issue with this king of fouling and believe me, we track this closely. My thoughts are that the pass partition plate is erroded and some of the inlet condenser water just going straight the outlet side and indicating low condenser leaving water temp. We had seen some erosion during our last clean up.
.
Any comments.

 

[RossABQ]

Was the unit out of service with a wet condenser, or dry? In either case, I think you have answered your own question: similar units that were not off-line are doing fine, the only problem with this unit is condenser range. Something was growing in the condenser for three months! Without condenser water flow rates, you really can't verify this. I'd suggest using dP on the condensers to compare flow rates at least qualitatively. If these are like most CentraVac's, the ports should be there to get this info.

 

[gepman]

Condenser water flow rates and dP across the condenser would be good information to have as others have pointed out. Is the value for SCT (Saturated Condenser Temperature) a measured value or a calculated value based upon compressor discharge pressure? Where are you measuring it if it is a measured value? What does "Purge is 0" mean? How is your condenser water flow controlled?

 

[Centravac]

I don't have the pressure drop across the condenser at the moment but I can answer other questions.
.
Saturated Condenser Temperature is the measured value and pressure is the calculated. We also have a pressure gauge and it's reporting the same pressure as calculated condenser pressure. The temperature element is located on top of the condenser.
.
Purge 0 means that our purge unit is not taking out any non condensables from the chillers. This coould be due the bad purge unit but since the measure Condenser pressure matches calculated condenser pressure, I dont think that is the case.
.
Condenser water flow is controlled by a Control Valve that opens or closes based on condender pressure. In winter we throttle a hand valve to limit the intake of cold water to the condenser and thus artificially increase the condenser preesure. That is why you see such high delta T.

 

[imok2]

I still believe you have a lack of heat transfer due to Minerals, scale, mud, algae, and other impurities which is easy to determine. Isn't it time to clean the condensers? If the tubes were clean then you might say that the water velocity is to high because your about 10 degrees below the other chillers from what you say

 

[gepman]

The problem could either be fouling as imok2 says or a plate on the heat exchanger head failing and allowing water to bypass the heat exchanger as you say. Fouling is much more common but either way you will need to open up the HX to see (unless you chemically clean the tubeside).

In industrial systems we usually calculate SCT from the discharge pressure measured at the (evaporative) condenser but it is best to measure both in order to see if purging is required (or if you autopurger is working). How does it measure the condensed liquid refrigerant temperature if the probe is at the top of the condenser where (I would assume) the gas enters and has not yet condensed and is probably superheated? (Please humor me since I rarely see a chiller, most of my work is industrial).

 

[imok2]

gepman , It's been my experience thet chemical cleaning will have minimal effect on mud,algie but have a good effect on mineral deposits. also

"The temperature element is located on top of the condenser"

WoW, how does that work!!

 

[quark]

Why should you worry about the condenser approach if you are turning down the capacity of the condenser, deliberately? I don't bother as long as the discharge pressure is with in control and the secondary refrigerant temperature is normal (i.e chilled water or brine).

The condenser approach gives you indication about the condenser performance at design flow condition. As your prime interest is to maintain the pressure in the system, the cooling water flowrate may not be sufficient to cool the liquid to the required temperature.

If your other system is running with same cooling water flowrate but with different refrigerant parameters, then check for scaling on condenser tubes for the faulty chiller.

If you forsee bypassing of the cooling fluid, it is better to throttle the outlet valve (infact, I would always go for controlling the outlet valve).

As already said, temperature sensor at the top of the condenser is not proper. You should check the approach with respect to subcooled refrigerant temperature (i.e at the bottom)

If you can place two pressure gauges one at the top and second at the bottom of the condenser then the difference indicates the presence of noncondensable gases and thus effectiveness of your purge unit.

If both the cooling water inlet and outlet are from one side of the condenser then check the gasket between dome and tubesheet for possible bypassing.

 

[gepman]

quark

If you forsee bypassing of the cooling fluid, it is better to throttle the outlet valve (infact, I would always go for controlling the outlet valve).

If the bypassing is internal to the condenser I don't see how changing the throttling from the inlet to the outlet will affect the bypass. The flow will still split to have equal pressure drops between flow through the tubes and flow through the leak. Centravac suspects a leak in the pass partition in the head (or dome as you call it). You had a good idea in pointing out that it could also be a faulty seal of the gasket between the baffle plate and the tubesheet.

You should check the approach with respect to subcooled refrigerant temperature (i.e at the bottom)

I wouldn't expect much subcooling on the shell side of a condenser unless the liquid refrigerant level in the condenser was over the tubes otherwise the liquid refrigerant would not be in equilibrium with the gas at that pressure. If there is liquid refrigerant backing up over the tubes then in addition to what you state I would also check the level of this backup as I suggested in item #3 in my first response. Liquid refrigerant over the tubes of the condenser basically reduces the size of the condenser. I tried looking at the Trane Centravac manual but it wasn't very detailed in how the liquid gets from the condenser to the evaporator.

If you can place two pressure gauges one at the top and second at the bottom of the condenser then the difference indicates the presence of noncondensable gases and thus effectiveness of your purge unit.

I don't see how this does anything but measure the gas pressure drop on the shell side of the condenser. You need to measure the SCT (temperature of liquid refrigerant) and compare that to the pressure (at the same point depending on how accurate you want to be). The liquid must be saturated (not subcooled) otherwise it would tend to indicate non-condensibles even if there aren't any.

 

[imok2]

Without the water pressure drop acrossthe condenser I believe were pis...g in the wind!

 

[quark]

gepman,

The first comment is rather tongue in cheek as I overlooked the pass partition leak. I suspect this may not be the case as there is a higher dT across the cooling water circuit.

The pressure should be in equilibrium in a flooded type condenser irrespective of liquid level, in the absence of noncondensable gases, as the discharge pressure of the compressor varies as per the condenser pressure.

The method I mentioned is a standard one for manual purging. We can always ignore the refrigerant pressure drop across a condenser when the system is running.

As per your reasoning, systems designed with subcooling of the liquid refrigerant should always err in indicating noncondensables. However, this can easily be overcome if you take pressure and temperature readings in the liquid region.

I wish a very Happy, Prosperous and Peaceful New Year to you all.