Printable View
We have an ammonia plant running a freezer. On startup in the morning it keeps shutting down on high liquid level. After a few attempts we have managed to get it running, then it runs all day with no trouble. We had to shut the surge tank off this morning to get it pumped down enough and now its running. The only thing I can think of is either a check vallve causing the evaporator to fill up and flood or maybe a TX valve.
Any Ideas please???
I had this kind of problem once and it was due to a check valve being installed to high in relation to the dump trap. Are you pumping the liquid or using hot gas dump trap?
Buckiesr
Have any of you ever heard of "Supachill Technologies Inc.,"? Can I attach files to a post on here?
Buckiesr
Iceman007.
With the little info you have given us.
I can think of a thousand options. But you say that the plant is going out on high level. ( on the surge drum ?)
Start with basics, what does the liquid level switch
control? What level switch is indicating high level
Check it's operation.
Maybe a liquid line solenoid valve. Which could be passing thereby causing the surge drum to overfill, despite the level control switch not being made.
Maybe the liquid level controller is mounted where oil from the surge drum drains into it and your levels are "confused". Due to the high proportion of oil within it.
How often are oil drains carried out?
We had a system where we only gained proper control once we had fitted an oil drain line.
From the bottom of the level controller into the oil seperation drum.
Maybe the level controller needs resetting. The red flag type can be reset by passing a magnet over them ( vertically up and down). You will see what I mean when you do it.
I have seen level controls "confused" or as we used to call it "Ghosting". When the valve controlling liquid make up from the intercooler, opens to quickly. And the inrush of refrigerant causes. Violent gas "flashing" and thereby setting the high level Alarm off. Incidently this can happen in whenever refrigerant is being metered into a vessel.
Basically though I would assume yours is question of poor level control. Look at the basics first.
I hope this post has been of some help?
Grizzlyhttp://www.refrigeration-engineer.co...ons/icon14.gif
Couldn't post back any earlier, connection was faulty.Quote:
Originally Posted by Grizzly
The high liquid alarm is in the surge drum. It is only there on startup, well for thfirst hour. Usually it gos out 2-4 times and I had to close the valve off to pump it out today to keep it going. Once it gets off it will run with no problems all day. It's a Frick screw on a blast freezer. The oil I don't think has been drained for a while. I had a look through some old service sheets and I know at the end of September a sample was took for testing but I can't find the results. Also a small shaft seal leak was repaired. At this time a diaphragm and orifice was replaced to a valve which seemed to solve the problem but it has returned in the last day or two. My own thought was a liquid line solenoid passing causing the drum to overfill, I thought if the problem was TX valve related it would happen not only on startup. I don't want to have to chane a load of bits to get it runnning but. maybe it needs a good overhaul as the 40,000 hr is getting close. It's a single stage recirculation system if that helps. I can't find a liquid pump either although I had only a quick look at the high side components as the freezer room was out of bounds. I shall return tomorrow and have a better look, as I will be able to start and stop the plant with no one in there. The closest I got to the coolers was the header and TX valve on top......
Thanks for the inputs
Iceman007.
It certainly sounds as if there is an issue with the liquid make up or Level controller.
Although just as a thought when you get chance have a look at the state of the plates / evaporator.
I look forward to hearing how you get on.
Cheers Grizzly
That sounds like the suction pressure is pulling down too fast during startup. At this time the liquid is simply laying in the surge drum, until the compressor starts. As the suction pressure starts to decrease the liquid begins to boil. If the suction pressure decreases quick enough, the boiling tuns very violent and CAN cause flooding of the compressor. It is during this violent boiling period where the liquid level rises and CAN cause the high level float switch to trip.Quote:
Originally Posted by iceman007
As you have stated, once started it runs OK. That tells me it is only related to the start-up condition and what I describe above.
You can try to slow down the loading rate on the compressor so that the suction pressure does not change quickly. That should solve your problem.
Hope that helps.
The fact you say start up makes me suspect migration to the colder vessel. It's something to look at. If you can isolate that vessel during down times
do that for an eliminating troubleshoot test.
Ice man is right if the compressor is flooding. If it's an accumulator ( espicially low side). different situation.
Hi Iceman007
You say it's a liquid recirculation system but you also say it has TX valves and you can't find the liquid pumps.
If it is a liquid recirc system the pumps will be at the bottom of the accumulator, If there are no liquid pumps the surge vessel will be a suction trap, If you are getting a high level in this vessel it will be the TX valves overfeeding and the liquid solenoids on the rooms leaking.
Another way to check what type of vessel it is have a look and see how many float switches are on the control column. If there is only one and no solenoid feeding the vessel it will be a suction trap.
Hope this helps
Paul
Sorry I am way low on data to parse here...retract my last outbursts
Went back this morning and started it up myself. No problems at all. Suction pressure pulls down gradually, as it does, setpoint in panel changes and compressor loads and unloads OK. Drum had liquid in because of the ice around the inlet and sight glass- I had thought it was a pumpdown system ? There are no pumps and surge drum is like a trap feeding a header (this has just one TX valve) and another TX valve on the surge drum inlet from receiver just after the LLSV. Hot gas from oil seperator passes through the cooling tower and returns to the receiver. On shutdown I had perhaps naively expected to see a pumpdown cycle, but no reduction in suction pressure, just shuts off. The original thought was the LLSV not shutting off and allowing the drum to fill. This problem has only been going on for a few days, and only during pulldown after startup. I had wondered about the suction pressure dropping too quickly, but it was taking over 30 minutes to drop from 700kPa to 100 kPa, with a final suction of 65kPa after 40 minutes or so. Usually the plant runs under full load for much of the day, hopfully its overhaul should improve efficiency. Not that imortant but compression ratio 16:1 or so (discharge pressure 1230kPa)
Can't think of anything else
Back again. Did you happen to notice what the evaporator was like?
Basically we are all saying the same thing yourself included.
What US Iceman was saying about the refrigerant
Boiling can be a real problem with Ammonia vessels, especially if the control band between working levels and high level is small.
Quite common in UK. where for economic reasons the surge drum/ economiser ect are sized on the small side.
Anyways the state of the evaporator ie if it is iced up or not defrosted properly. Can cause the wet return to create exactly what US Iceman was describing. Also see my referance to "Ghosting."
Maybe just maybe the defrost stratergy / useage cycle needs looking at?
You would be amazed at the number of times Operations have missed or cut short a defrost cycle.
There is certainly enough of us watching this one and hopefully with everyones help the problem will be resolved.
Keep us all posted iceman007 as it's good to get a nice "juicy" Industrial problem for a change.
Cheers Grizzly
Hu Guys:)
it's a surge vessel and it is surging on startup, quite normal. A partial pumpout before shutdown would help. Also shuting a hand valve on the liquidline would rule out a passing solinoide.
Loading slowly helps too, but I remember reading it does load slowly.
Could be a false high, with the foaming effecting the level indicator. Measure the suction superheat and watch the discharge temps to determine if there is any liquid carryover during startup.
You may well have a high level and still no liquid carryover, in fact this should be the case as the high level cut out is there to protect the compressors.
Kind Regards Andy:)
I shut off the hand valve on the liquid line as a test to either identify or rule out the solenoid. On MOnday all will be revealed !!!!
Thanks
It would be worthwhile to make note of the start-up liquid level in the surge drum and compare that to the high level shut down float switch.
The boiling in the surge drum can cause the liquid level to rise and cause the float switch to trip. It doesn't mean the compressor floods out, it just means the float switch tripped.
Happens all too often...:o
I had a similar situation with a surge drum. When the system was shutdown, the liquid in the surge drum would migrate to the evaps in the freezer area. The level in the surge drum would then lower and call for make up; this would continue to happen until the temps equalized. Then on startup, the liquid would rush back and cause a high level and a shutdown. Our solution was to set the level such that the makeup solenoid valve would not open during the shutdown period, then reset to normal after startup.
Hope this helps,
Lance
here, i sense a serious issue on liquid level controls.
we have similar experience. but, we knew the problem is from carry-over lube oil stays at low temp receiver.
we also recognise the overfeeding from liquid supply line during liquid level control in the event of flooded design.
we have physically seen minor particle cause improper closure of electronic valve seat.
we also have experience with lube oil inside HLL float switch.
all these is the effect/s of poor maintenance!
I thought I might let you know what happened with this problem. I send two of our ammonia techs back to site with a new valve. They changed the solenoid on the liquid level make up line and started it up. All ran fine, and every morning it started with no problems, then......... after a three day shutdown, they started up the compressor, and twice it went out on high liquid level, so the problem has got better, but still not disappeared altogether, because it is still there after an extended shutdown (3 days). I was quite hopeful that a replacement solenoid would be the answer, but not to be, so now I have to rethink again.
I am still unclear as the nature of this system. You have mentioned TX valves and that is not commonly used anymore on flooded systems (I am thinking of the Sporlan Levelmaster) except on small systems.
If you could scan a drawing of the system and post it, that would help us all.
i've few Qs to ask.
1] alarm occurs after compressor stabilize?
2] any alarm delay timer for high level alarm?
3] how often oil is refilled?
4] is this a liquid pumped system?
5] this alarm only happen when there is a prolonged stoppage? continuous running does not pose such alarm?
thank you for such brain storming...
Hi, NH3LVR :)
Could not agree more with you;)....I have to admit it is hard to discuss something not knowing how it looks...one photo can say.....Quote:
Originally Posted by NH3LVR
Best regards, Josip :)
Iceman 007.
You are definately making us all think with this one.
I agree with the last few posts. Something strange about this one?
I am inclined to think that it is something that over an extended shutdown is affecting the liquid level.
So lets start looking for abnormal influences.
Did you change the whole valve or just use a repair kit.
Could your guys have missed slight wire drawing of the valve seat?
I to am confused as to the referance to TXV valves?
surely the "Txv" should be refered to as Metering device?
IE solenoid controlled, when the freezer reaches temperature how is the liquid feed to it shut off?
As you originally mentioned "flooded evaporator"
If this is not pumping down on reaching setpoint or its liquid line shut off valve, (quite often just a thermostatically controlled solenoid valve) is passing .
You would have on start up a evap full of liquid surging back to the surge drum! Andy refers to it as does Others.
Have you tried widening the band between working level and high level on the alarm control.
Have you looked at the defrost stratergy ie is a partially blocked evap compounding the problem.
Keep on plugging away you will get there and it will probably be something stupid!
But most importantly thanks for and please carry on
letting us all know what's happening.
Grizzlyhttp://www.refrigeration-engineer.co...ons/icon14.gif
slow the compressor load rate and let system stabilize. The liquid in evap floods back to low pressure zone.
magoo
if PLC controlled have a first start a day function that restricts compressor load rate, that reverts to rapid response after seeing fully loaded position and half decent accumulator conditions. If pump fed, ignore fluctuations in differential pressures until optimim conditions stabilize. Soft control if starting from hot start.
Hell I've moved to improved poster.
Hi, Iceman007 :)
.....this is only one more idea...some (bad;)) experience from the past....Quote:
Originally Posted by iceman007
.....This is a freezer system (-45*C), thus you must have some second stage system (-10/15*C) and those compressor/s are working together with compressors/s on (-35*C to keep temperature in cold rooms) even when your freezer not....freezer remains open (for cleaning...) and all ammonia within evaporate and return to belonging freezer separator where we still have low temp ammonia - it is like a sink for ammonia from freezer evaporator, freezer separator remains cold also due to low pressure caused by high stage system working....seems the system is overcharged a little (due to small separator) or your level switches are installed too low....
you can help yourself a little by switch on your freezer pump before start compressor and push some liquid back into freezer....
of course it is possible to have a leak on charging line...BTW...is your feeding line coming from -10*C separator with lower pressure or from receiver through heat exchanger with high pressure (do you have a coil within -10*C separator as a heat exchanger...there you can have a small leak;) enough big to cause troubles in stand still conditions and almost not noticeable in working conditions)......
....this is only a guessing (we do not have a scheme of your plant)....maybe your system is not like I think it is...and maybe I am completely wrong;)
Best regards, Josip :)
Hi Iceman,
Diagram surely will help. Had be like each other problem and solved only slow pickling compressor when ammonia much cooks. Supply valve won't also defective, altitude float doesn't go determine in report- necessary see.:confused: Cheer Josef
Hi Iceman
Had a similar prob,em on a ammonia pack a few years ago. I assume that you have checked all the basics like sol valve passing etc. The system that I worked on was a pump circ and we found that if you start the pumps up for 60 secs before compressor start up this reduces the level in the vessel and allows the initial surge of ammonia to be below the high level tripRegards
Iceman
Welcome to the forum Bruce.Quote:
Originally Posted by Bruce Skinner
Does Lancashire equate to Ex TDG Manchester.
If so you were one of my bosses and used to carry one of the biggest bunch of keys I have ever seen.
If you are the very same Bruce pm me and I will update you.
Cheers Grizzly AKA Steve Adamshttp://www.refrigeration-engineer.co...ons/icon14.gif
Sounds like liquid is migrating during the night.
if you have a shut off valve to isolate the vessel close it during off time and make sure reliefs are there for protection.