Quick fix for this leak? McQuay R134A centrifugal chiller

[kaon]

(it's a cross-post (http://www.hvac-talk.com/vbb/showthread.php?p=1973525#post1973525)... sorry I'm that desperate for competent help)

Hi all,

My chiller developed a serious leak at the flange joint where cooling gas exits the motor casing. (see pics)
The leak was naked eye visible and audible from a distance. But we cannot tolerate much downtime right now... so all we did was to tighten the bolts and the leak slowed down to abt 10%. And we add refrigerant as needed.
It is a 220 ton McQuay PEH063, year: 1996.

The local McQuay/Daikin guys want to take 3 days to do the job.
We want to DIY or use a third party if there is a faster, cheaper way. To do it with the rfrg in place. (do I need to don flame-proof gear now?)

One thought is to use epoxy-type sealant with a nipple fitting to provide an gas escape route during curing. And then close the nipple once the epoxy has reached full strength.

Any suggestions?
This chiller will be retired within one year, so longevity of the fix is not necessary. It is more important to minimize downtime.

PICS:
http://img83.imageshack.us/img83/3452/mcquaypehrefrgflowdiagrbr6.png
http://img525.imageshack.us/img525/8647/img00234lg7.jpg
http://img257.imageshack.us/img257/848/img00242fe0.jpg
http://img300.imageshack.us/img300/821/img00243vc1.jpg

[taz24]

Hi all,

My chiller developed a serious leak at the flange joint where cooling gas exits the motor casing. (see pics)
The leak was naked eye visible and audible from a distance. But we cannot tolerate much downtime right now... so all we did was to tighten the bolts and the leak slowed down to abt 10%. And we add refrigerant as needed.
It is a 220 ton McQuay PEH063, year: 1996.

The local McQuay/Daikin guys want to take 3 days to do the job.
We want to DIY or use a third party if there is a faster, cheaper way. To do it with the rfrg in place. (do I need to don flame-proof gear now?)

One thought is to use epoxy-type sealant with a nipple fitting to provide an gas escape route during curing. And then close the nipple once the epoxy has reached full strength.

Any suggestions?
This chiller will be retired within one year, so longevity of the fix is not necessary. It is more important to minimize downtime.



Hi kaon

You are between a rock and a hard place.

There are no easy ways that I know of to sort this problem. The quickest would be to shut the system down, remove the refrigerant from the area local to the leak and replace / repair the offending item.

Where exactly is the leak? Is it on the joint or behind the joint where the pipe is fitted to the flange?

There is a lot of rust there and I can't tell if it is the flange joint or the pipe behind it.

Three days to shut down does seem excesive but they are the guys on the ground and with the expertise so I would listen to them.

Offer to pay them to work round the clock to carry out the repair with a bonus on completion if within a set time.

Down time now will be cheaper than to bodge the job and continualy have it recurring.

All the best

taz

.

[fridgemandan]

i agree with taz, a rough job will bite you in the ass in months/weeks to come, get the repair done properly, i would condsider replacing the rusted pipe sections, while the system is down. from my own experiance there is nothing more disheartening than to take a system of line to carry out a quick fix and run into problems in the futures. let us know how you get on good luck

[kaon]

Where exactly is the leak? Is it on the joint or behind the joint where the pipe is fitted to the flange?Thanks Taz.
The leak is at the lower part of the flange joint, tightening the two bolts helped.
At least that's where I feel it with my fingers. And reported by my team.
Will personally do a bubble test to see for sure.

[ptsac]

Kaon, that flange will have an o ring. It will need replacing. You could isolate the king valve and motor cooling etc and pump the refrigerant into the evaporator. That would remove the majority of the liquid, and then transfer the remaining vapour with a transfer unit to the condenser. You would be relying on the discharge check valve holding.
The o ring material is buna nitrile which is compatible with R134a. McQuay estimate of 3 days seems to be a fair estimate as i would expect the refrigerant charge to be about 350 Kg and they should be able to provide the correct size o ring to make a proper repair.

[kaon]

Kaon, that flange will have an o ring. It will need replacing. You could isolate the king valve and motor cooling etc and pump the refrigerant into the evaporator. That would remove the majority of the liquid, and then transfer the remaining vapour with a transfer unit to the condenser. You would be relying on the discharge check valve holding.
The o ring material is buna nitrile which is compatible with R134a. McQuay estimate of 3 days seems to be a fair estimate as i would expect the refrigerant charge to be about 350 Kg and they should be able to provide the correct size o ring to make a proper repair.
Soap bubble test confirmed leak is at the flange joint, mostly on the lower side.
We closed the king valve, then ran the compressor, the chiller tripped after abt 3 minutes when the saturated evap pressure dipped below 300 kPa.
Then we quickly closed the motor cooling supply valve (under the condenser barrel).
The sat evap pressure and sat cond pressure equalized again within 30 seconds... so either the discharge check leaks or there are some other lines that need to be shut off.

Is it safe to run pump down with the motor cooling supply valve shut off?

[ptsac]

Hi Kaon,
I believe that you could shut the motor cooling and liquid line and pump over two times with a decent time between starts. The problem lies in whether your oil is refrigerant cooled. The heat build up is fairly quick and will trip at 60c. If it is water cooled no problem. The next problem is that from start the vanes will be fully closed, thus preventing a decent volume of refrigerant being moved. There is a manual procedure for vane operation that could be used, but the vanes must be closed on start. Maybe jumper the output relay for vane open after the motor is in delta. You stated that the evaporator pressure went to 300kpa the correct mech pressure switch setting would be about 200kpa.
The evap and cond pressures on the microtech will equalise as the disch pressure is read above the discharge check valve. If you put your own guage on to the condenser vessel this should indicate if the check valve is holding. All this can be a bit daunting as the compressor starts to surge and there is the worry that the motors too hot. There is also the doubt of the check valve. My experience has led to me to believe that removing the charge into clean reciever cylinders is perhaps the safest method. I still believe that McQuay estimate of three days was fair.

Best of luck

[ptsac]

http://www.mcquay.com/mcquaybiz/literature/lit_ch_wc/IMOM/Im307-3.pdf

I think that page 9 of this document explains better. Hope this helps. I am not sure how to post a link but where there are spaces should be underscores _

[Taaz1275]

kaon,

assessing the advices from all I believe should now give you an idea of what you should do to save $$$$, I believe that you are concerned about downtime which is $$$$ as well, weight out your options and with all the brilliant advices that you have got thru this forum you should be able to make up a report to support your recommendations of what is best needed to be done in this situation saving $$$$$ in the end...buisness is all about making $$$$$...so the best option is the one that would cost less..save the company more $$$.Your explanations with options provided with figures will surely give a solution to your problem.

[sarmad]

hi
pls i have a***** leak in power cable connection with stater terminal with my mcquay centrifugal chiller model wsc 087
if any one know how to fix it?
thanks alot

[ptsac]

hi
pls i have a***** leak in power cable connection with stater terminal with my mcquay centrifugal chiller model wsc 087
if any one know how to fix it?
thanks alot
No quick fix I am afraid. Refrigerant out, motor cover off, disconnect cables inside and out. New terminal(s), maybe taper thread or parallel thread with "o" ring. Use loctite 577. Be careful tightening as spark plug types are prone to stressing, best done with a socket.