Lithium Bromide/Water absorbtion chiller

[tchnikal]

We have 4 Carrier Mod 16JT-150 LiBr/water absorber chillers, 1,500 ton capacity each, using steam at 110psi to the generator. Design is 9.9 lbs of steam per ton. We're experiencing loss of capacity, can only squeeze 65% at full load conditions. It almost appears as though condensate is backing up and not able leave the machine fast enough through the drain heat exchanger float valve. Only two of the four machines have these symptoms. However, condensate from all four absorbers discharge into a common header. The condensate header is not presureizing. I'm inclined to believe the problem is internal to the machine. The steam flow control valve will open 100% to compensate for full load, but the steam flow will not exceed 65% of full flow capacity which would be 15,000 lbs per hour. The steam flow differential across the steam control valve decreases, telling me the flow has decreased. The drain heat exchanger float valves were inspected and checked out OK. We stopped using octal alchohol (wetting agent)because we were told there were side effects. However, the problem existed even before we stopped using it. There is a bypass around the drain heat exchanger float valve, which when opened will increase capacity, but efficiency suffers as a result and it also creates steam impingement on the drain heat exchanger tubes causing them the fail. Would appreciate any input that would help solve our capacity problem.

 

[briand2]

"The steam flow differential across the steam control valve decreases, telling me the flow has decreased".

I assume you mean the "pressure differential", not the "flow differential". If this is so, then the decrease would not imply a decrease in flow; it would imply that the valve had opened further.

Are you sure that the steam pressure upstream of the steam control valve is remaining at its design value?

Regards,

Brian

 

[Cooky]

Has a solution analysis been carried out, if so, what are the results?( very important as an internal condition indicator with the following; inhibitor type and PPM, alkalinity expressed as LioH, total undissolved solids PPM and dissolved Copper PPM). Are both cooling water temperature/ flow rate and chilled water flow rate for each machine close to design?(obvious I know) Condenser and absorber tubes are not fouled? Absorber loss is below 3degC (5degF)Are refrigerant levels correct?(indicated by but not restricted to Cycle Guard valve operating frequently) How much Octyl Alcohol has been added to each machine in the past?( you say you have STOPPED using Octyl Alcohol?)

Taking into consideration some assumptions without the above information, loss of capacity may be but not directly a result of the following:

Fouled condenser/ absorber tubes, cooling water temp/ flow rate and chilled water flow rate. Non condensables in absorber giving high absorber loss caused by air leak or excessive out gassing as a result of accelerated corrosion due to lack of inhibitor( high temps with Chromate inhibitor and Octyl Alcohol cause inhibitor drop out(Molybdate inhibitor depletes rapidly naturally and requires replenishing frequently). Excessive refrigerant levels will cause Cycle Guard valve to operate more regularly if not continuously, this results in a large if not total loss of capacity depending on how much excessive refrigerant present, excessive refrigerant can be taken on board from three places - leaking condenser tubes, leaking absorber tubes and most likely the generator or condensate heat exchanger. Excessive Octyl Alcohol will create a film on the surface of the solution in the absorber blocking its ability to absorb the refrigerant.There is also a slim possibility that excessive Chromate inhibitor with Octyl Alcohol has formed a gel and blocked one or more of the heat exchangers?


There is one clue in the information however, that when the condensate heat exchanger is bypassed, steam consumption is close to design? this leads me to believe that the problem may be the heat exchanger itself for whatever reason? Whatever the problem is, in most cases it something simple.

Another thing that could be done is to plot the Equilibrium diagram and compare that to design, this may point out the area to start looking.

The most relevant things are covered in the Carrier IOM. Unless you are experienced with fault finding on absorption chillers, I strongly reccomend you speak to your local Carrier office.

I hope this information helps.

Cooky (Carrier)