I have seen it somewhere, but has anyone any articles/papers for recommended suction pressure rate of decrease in a plant pull down, or in general to avoid liquid pump cavitation.
Sometimes there is a rule of thumb etc.
Thanks in advance
Printable View
I have seen it somewhere, but has anyone any articles/papers for recommended suction pressure rate of decrease in a plant pull down, or in general to avoid liquid pump cavitation.
Sometimes there is a rule of thumb etc.
Thanks in advance
Ranger, scrolled all the way up from the bottom, because it turned into a book instead of a quick reply!!
Hope you enjoy the reading, not sure if I answered your question, but I think I tried to cover most bases :D
------------------------------------
By pull down, do you mean after a long shut down, or after heat load has been added, like starting a "warm" evaporator?
By cavitation, do you mean cavitation that may over time damage the pump, or cavitation so severe that the pump fails to deliver pressure?
After a shut down where the plant is at ambient (usually around 10C around here), I usually to a TLAR (That Looks About Right) assessment.
Last plant I pulled down after a week shut down was 4 metric tons of NH3, 2 x XRV204145.
Took me maybe 3 hours to take it down from 14C to -42C, running the compressors in manual with the suction valves throttled to keep the suction pressure at 0.5 - 1 bar.
But I never run the pumps continuously while pulling down.
I'll open the valves to a few consumers, then I'll start 1 or 2 pumps intermittently for a few minutes just to get the level down in the LP drum.
Once the pressure is down I start 1 pump on the consumers I have opened, then
Ranger, you know this next bit, but I'm still adding it for other readers :)
-----------------------------------------------------------------------
As long as the liquid in the Low pressure receiver is saturated, there will be no boiling (bubbles), and the plant is in a stable state (on paper, there will always be bubbles :)).
Let's say the pressure is -40C, the LP receiver, the liquid and everything is at -40C, therefore, no bubbles.
-----------------------------------------------------------------------
if you lower the pressure, boiling will start, and the more and the faster you lower the pressure, the more boiling and bubbles you will have, and the chance of the pumps cavitating increases exponentially.
I can't really give any rule of thumb, wish I had one so I could tell our apprentices an easy way to judge it.
For me, it's experience and a TLAR assessment depending on the size of the plant.
When pulling down a plant after a shutdown, my rule of thumb is "never allow the suction pressure of the compressor to be above 1 barg (Unless the manufacturer says it should be lower). This way I don't overload the compressors or the condensers.
When it comes to how fast to load a compressor on a plant that is in operation, again it comes down to a TLAR assessment.
If the compressor loads to slow, the client will complain, if the plant loads to fast and overloads the condensers or makes the refrigerant pumps cavitate and cut out, the client will complain.
So for me, it all comes down to looking at a few things:
-Plant size
-refrigerant charge
-how many consumers
-what kind of consumers (storage, blast freezers, cold rooms, etc)
-condenser capacity (sea water cooled or air cooled)
-compressor size
-compressor motor size
For example:
-if it was a plant that operated just freezing storage, I would have the compressors react slow, and keep the suction temperature +/- 0,5C to prevent any upsets or trips on low pressure because of the sudden drop after one of the rooms or evaporators finished a hot gas defrost, or got pulled down to setpoint after a defrost.
-if it was a plant with multiple blast freezers, I would have the compressors react fast.
Blast freezers (on the plants we oversee), are usually pulled down like freezing storage, fans off.
Product is loaded in, then the fans are started, and I would want the compressors to get up to 100% as fast as possible to arrest the increase in pressure before it goes to high, because if it goes too high, it will also start to affect the other blast freezers.
On most freezing plants, the recurring reason has been not enough liquid in the LP receiver.
--------------------------------------------------------------------------
The only plants where I have had problems with the refrigerant pumps cavitating to the point of loosing pressure and make the compressor cut out on LP is on RSW plants with spray chillers.
In spray chillers, the liquid charge is all in the drop leg, and there is no liquid in the shell/tube part.
A refrigerant pump takes liquid from the drop leg pumps it up to one or two spray bars (depending on chiller size), with special nozzles, that are placed above the tubes in the shell.
Now, if the compressor loads to fast and starts demanding more gas than the liquid sprayed across the tubes can produce, the refrigerant in the drop leg will start to boil.
When the liquid in the drop leg starts to boil, the refrigerant pump will start to cavitate and will loose efficiency, thereby not pumping as much liquid as it should, , but the compressor is still demanding the same amount of gas, so now the liquid in the drop leg is really boiling.
The amount of gas bubbles in the refrigerant increases, the pump fails to pump any liquid to spray over the tubes inside the chiller, the suction pressure collapses and the compressor stops on LP alarm.
This was a major issue when we first started delivering RSW plants with spray chillers, because Inverters were still expensive and the plants would be sold with star/delta starting gear.
the problem was to get it slow enough, without having the client complain that it took to long.
Also, those last millimeters when the slide on the screw compressor closes gave such a drop in suction pressure that it could be enough to trip the plant.
Luckily 99.99% of the plants nowadays have inverters :)
Sorry I couldn't give you a rule of thumb
Tycho,
Thanks for detailed response.
I guess I was generalising with my question.
Every plant different & understand your plants with pull down each day.
I was really thinking of general production where reasonably stable, unless a chiller is turned on. Mainly chillers & boning rooms.
Cornell suggest 1 psi minute, so to make nice round metric number 10 kpa sounds good to me, if you were to put a number on it.
If something like a beef chiller is turned on plant would have to react quick enough to get on top, but unloading would have to be reasonably quick so to not overshoot & possibly cavitation liquid pump.
Big plants, little plants, has to be in proportion to this starting point.
https://www.cornellpump.com/site/wp-...OUBLESHOOT.pdf
I think that pulling down 1 psi per min is good number but in real life is not easy to achieve this rate.
I want to share my experience with one of the plant. It has 3 low pressure vessels. V3 pressure 45 psig for cooler and dock. V2 5 psig for holding freezers. V1 -5 psig for one holding freezer and 2 blast freezers. If no need for blast freezering V1 and V2 are on common. Than BF should be started, suction separated and freezing process can be initiated. Pressure in V1 can be pulled down by 3 ways.
1. Pressure pull down only for the vessel V1 and BF will be started after pull down. This is head ache.
2. BF started up and pressure will be pulled down with load of BF. It can be done gradually without any problem.
3. Reduce set point for the holding freezer connected to V1, run 1 or 2 evaporators of this freezer and pull down pressure in V1. When pressure is low, start up BF and reset set point for holding freezer. It will go gradually as well.
My point is that lowering suction pressure should be done with additional load and it will go smoothly or/and it should be done by the smallest compressor.
Thanks SegeiQuote:
Originally Posted by Segei
as always :DQuote:
Originally Posted by RANGER1
1 psi/minute doesn't sound like a lot, 10 kpa, I will agree with.Quote:
Originally Posted by RANGER1
I agree that, if it is a plant that is reasonably stable, where they might add a slight heatload every now and then, then the pull down can be more gentle.
The plants I work on, fish factories on ships and blast freezers on fish factories on land, on these plants you have to use the TLAR and find the proper pull down and look at the motor current, condensing pressure and suction pressure, and try to find some kind of balance where you get the motor current up as close to the limit as you can, while trying to maintain the suction pressure and not overload the condensers :)
some systems are undermentioned, so the operators can pile it on, but some may choke on the condensers or the motor current so you have to teach them to ease the load on.
I've been working with the Kvaerner KS-E5 pumps, semi hermetic and very sturdy, the guy who designed them has one of the prototypes at his house running a fountain (if he is still around).
and more recently I've been working with the CAM Hermetic pumps, and I can't say that I've ever had to discard or replace any major components on any of those pumps due to cavitation.
The KS-E5 pump is a vertical pump, and I had one once that had a rag wrapped around all 5 impellers from inlet to outlet, it was still giving pressure, but it was pulling excessive amps.
I did everything I could, but in the end I had to open it, and it ended up being 2 hours of slowly prying pieces of this rag that had wrapped itself around just about all the impellers out.
Not being smug or sarcastic, but in my 20 odd years, I have never come across a refrigerant pump that was damaged due to excessive cavitation. Loss of pressure due to cavitation, yes. then you have to check if there is enough refrigerant, or... if the compressor is loading to fast :)
Tycho,
I also think if system stable, then add a lot of load, a compressor can load quicker than 10 kpa because the liquid in vessel will be subcooled,, no rapid boiling of liquid in pump suction.
If you overshoot & drag down suction from it's stable point, then cavitation is more likely.
We do a number of plants for hamburger paddy's & pull down every day.
Went away from liquid pumps in recent years to surge drum on each evaporator.
I agree with you, but In my time I have never seen a refrigerant pump damaged from cavitation, so I disagree on that, and therefore don't see any reason to limit how fast you pull down a plant :)Quote:
Originally Posted by RANGER1
The loading/unloading of the compressor should be adjusted so that it will meet it's suction regulator setpoint, and don't bypass it.
If a compressor loads so fast it pulls down the suction so fast it bypasses it's setpoint someone has made a big mistake.
Compressor loeding should always be adjusted so that the suction pressure/converted to temperature can be arrested at the setpoint, and should never bypass the setpoint by more than 0,4C
in my opinion :)
Tycho,
Pump cavitation is a headache in a production facility, as it upsets critical room temperatures required for quality control.
some plants have machines that are to big, or added onto & not quite as you would like.
I have seen a number of damaged pumps over the year due to cavitation, low liquid levels, controls that don’t work, oil & crappy design.
Usually bearing issues on hermetic pumps & sometimes rotor hits stator.
Hermetic Pumpen good but no bearing indication like Japanese brands we use, Nikkiso & Teikoku which have electronic indication.
Does not happen every day, but now & then.