View Full Version : Ammonia pump problems (brand new system)

11-08-2011, 02:00 AM
wuts up fellas? new to the forum, and fairly new to the industry, a year in it. hence my profile name (nh3newbie)

anyways having issues with a tekkoku pump cavitating during sanitation/cleanup. its a one stage system, two frick screws, recirc. with two pumps one used as sec. rvs level probe with one high level float switch aswell as a modulating LFV. now pump usually cavitates when the vessel level reaches 50% or higher and at the same time the mod. valve is almost completely shut. i get a low amps alarm then it goes into cav. auto mode. now the setpoint on the rvs panel is set to a 1 min delay after this to automatically start pump again. if not the sec pump comes on. but its happened that at times this happens pump will never come back on and neither will the sec. and pump stays in cav. mode and cant be cleared. now this happens during cleanup only. there r 5 evaps in process room set at 44 deg. and 2 in the cooler set to 33 deg. and 1 in the dock. if any one has any suggestions please help . thanx

11-08-2011, 02:13 AM
Cavitation can be a difficult one to sort out on occasion.
Is the suction pressure falling just before the cavitation occurs? This can happen if the load falls off faster than the screws can unload.

11-08-2011, 03:24 AM
as per NH3, definitly a cavitation problem, which will prematurely damage the pumps. You will need at least the minimum 2 metres NPSH on pump as in clear liquid feed to pump, either raise the control level in accumulator, or unload compressors during sanitization cycle. This will reduce agitation/ boil up in surge drum/accumulator. Rapid pressure change in system will create havic. Alternatively do a pumpout of evaps during san cycle and start again afterwards.

11-08-2011, 09:05 AM
thanx for the replies guys, i greatly appreciate it. in response to nh3lvers post suction pressure doesnt drop much what i see happen is the compressor is almost fully loaded when the level is at 50% and lfv is shut . seems like compressor is not unloading fast enough. does tht make any sense?

11-08-2011, 10:09 AM
Hi nH3newbie.
How many of the evaps go into san/ cleanup mode at any given time.
You say the system is new?
What has the install / commissioning company said.
It definitely sounds as if you have control issues relating to the Suction Pressure and working level as the guys have said.
Simply put, you have a huge "Train" running flat out. Then you not only slam the brakes on but put it in reverse, with a huge heat load.
At least you have a lot to work with as you know when and why the cavitation is happening.
I also wold look at what settings etc activate / deactivate the auto cav- .
It is quite possible that the pump is staying permanently gas locked and therefor the pump won't come back on.
Why the second pump does not come on is a good question? Again what settings are controlling these.

Are the comps controlled by suction pressure or temps?
I would recommend that you enlist help to Analise your system.
Has this been a problem from new (How New?)

You have some very knowledgeable guys on the forum from your side of the pond.
(See above) :D
Have you thought about talking to any of them off forum?

For a newbie you have given good clear info sadly there are still many questions.
Good luck and keep us Informed please.

11-08-2011, 10:51 AM
Thnx for ur input grizzly . What activates the cav mode is whn there is low diff. Pressure between pump discharge pressure and vessel pressure or suction pressure. Now tht setpoint is at 15 psi w delay of 30 sec. So if the diff. Pressure gets to tht setpoint or lower for a time of 30 sec pump cavitates/shutdown due to low amps. Once this happens there is a strt delay of one min. After tht min pump should strt bck up.

11-08-2011, 11:00 AM
And as far as how new the system is it's a few months new ! New building new everything.

11-08-2011, 05:01 PM
What do happen with your evaporators during clean up? Do they shut off? If your plant shut off 5 evaporators simultaniosly, it can be reason of cavitation. Load should be changed gradually and suction pressure should be changed 1 psig per min.

11-08-2011, 05:49 PM
After watching it closely last night nh3lver u hit the nail right on the head. Suction pressure drops at the exact moment 2,3,and at times 4 psig in a matter of sec. And compressor is fully loaded and doesn't unload in time so the diff. Pressure drops below it's allowable setpoint and time delay therefore causing the cavitation and shut dwn of pump. At this point if comp. Still hasn't unloaded and pump two kicks on whn it does it can't get really goin due the suction pressure so it alarms n shuts dwn. Now at times. It will wrk perfect. What can I do.!!!!!!:0

11-08-2011, 06:02 PM
Hi nH3newbie.
I agree with Sergei which is my first question for the very same reasons as he lists.
I have had a thought and that is if as we all suspect the drop in load is to sudden.
Causing the bubbles to collapse in the drum creating the situation where the pump cavitates.
Then dependent upon what controller you use (Unisab which is from the same stable as Frick ).
There may be a second set point.
Which could be utilized by using an initiation signal prior to clean down.
Other options would depend upon the system you have, IE on clean down is the comp used to purge the evaps?

Have you tried different surge drum working levels and monitoring what they are prior, during and after clean down.
As a better understanding of the working levels would help.
Try extending the cavitation cutout time and see what happens.
Hope this helps.

11-08-2011, 06:20 PM
Just had another thought!
How close to clean down do the evap liquid line valves shut?
Could you extend the time that they shut before going onto clean down?
Staggering them would be even better.
This would smooth out the surging within the system.
Which may help eliminate you issue, as we say don't make waves!

I would love to see your commissioning data.
Or was this system commissioned without full load available.
In fairness not an unusual scenario.

11-08-2011, 06:24 PM
As I mentioned, the load should be chaged gradually. 5 evaporators should be switched off one after another with time delay. Probably, 2 min delay will be OK. It would be even better if you have pump out function. Liquid solenoid close first, fans keep run for 5-10 min and switch off as well.

11-08-2011, 06:32 PM
Only two of the five have the liq. Shut to thm basically pumped out and fans off. These two stay tht way through the entire shift. They do this via the setpoints in program tht hav been set by the plant manager and doesn't wnt any one messing w. But I hav also been shutting an additional evaporator manually whn I get there before sanitation to hav one less unit calling for liq. During this time. I hav messed w the levels aswell and hav the same issue. Also extended cave. Tim to max. Allowable evn dropped the lo diff. P. Setpoint and raised time delay to it's max of 30 sec. The controller is an DVD controller . What's this sec. Setpoint u mentioned About an initiation signal??? Please explain. And it seems I hav tried everything and this system is beating me. Errrrr

11-08-2011, 06:35 PM
Thts rvs controller sorry*

11-08-2011, 06:51 PM
If you have a plant manager that is dictating the set points.
He's the man!
sorry but unless the site engineers IE you, are allowed to set the set points.
It's his problem!
Unless he is going to agree to let you try and resolve the issue by working with you.
Then I doubt I can help you.
One some systems the controller can operate around two suction set points.
Where the comps control to either suction set point dependent upon which is activated.
I have not heard of a rvs controller, I know Vilter do a VRS twin screw controller.
Are you saying that whilst the sanitation ( Is this CIP / sterilization with steam of plate heat exchangers?) 2 plates remain flooded.
Is so I would assume the heat load fluctuations would be huge.
There in may lay your problem?
What is the process / product?
If this is a room where steralisation is taking place then reduce the no of evaps that shut down during this period and stagger there defrost times?
Not sure how your system is operating now?
Please advise.


25-08-2011, 11:59 AM
Check the NPSH with the pump manufacturer. Is thier any min / max orifice instaaled? Any vortex on the pump suction?

26-08-2011, 06:06 AM
do you have minimum flow in your system or not .
what is the suction pipe size for your pump and slope of that.
can you set time of compressor to load and unload . (in your case time of to load your compressor should more longer than times of unload time) may be 2 minutes for load time and less than 30 sec. for unload time

01-09-2011, 11:18 PM
If you have not resolved your cavitation issue yet you should talk to Mr. Keith Butler at RVS directly. RVS is part of the Evapco company. They use a specific set point program for each pump and vessel combination. Slowly changing the load on your system is nesscessary for stable operation. To quote a seasoned operator " It's a violent world and ammonia don't like violence". Check the bypass line exspansion valve, it may be choked down too much. "Canned" pumps like to work under load.

03-09-2011, 01:10 AM
Has anyone checked the pump rotation. I have been court like that on new systems, goes fine during certain conditions then goes pear shaped at lower suction conditions.