Liquid Ring Water Pump Capacity Control 3

[lcms]

Gentlemen,

We are designing a new vacuum system to recovery condensable gases from a set of 9 tanks that works on a batch mode with unpredictable overall load (I mean, each vessel can go from maximum load at start of the process to the minimum load at the end of the process in a independent manner. Therefore the overall load to the vacuum central can go very fast from minimum to maximum load depending on the production schedulle).

We intend to use two liquid ring water pump working on a step control (0-50-100% capacity), but we would also like to have a gradually control action within this limits (I mean, from 0-50% and from 50-100%).

My question is: - Could I use a speed control device (RELIANCE, ALLEN BRADLEY, whatelse...) on the water ring vacuum pump to proceed this fine tuning adjustment? In this case, what would be the acceptable range to work within (I am quite sure that the liquid ring vacuum pump have a minimum and a maximum shaft speed required to work well, but I would like to kbow how to discover these limits).

I would like to thanks in advance for your help, and send my congratulations to all people involved on this forum that has been of great help to engineering people, according may opinion.

Regards,

LCMS

 

[Torricelli]

lcms,

The limit is the tip speed to stay between a min. of 18 m/s to get a stable liquid ring and a max. of 22 m/s because of the erosion and because of the power absorbed (drastically increasing with the rotating speed).

In my opinion there is no money to save by speed control on small (=up to 3000m³/h) liquid ring vacuum pumps.

hope it helps

ET

http://www.vacuum-guide.com

 

[lcms]

Dear Torriceli,

The information you gave us really helps. Thank you very much.

Unfortunately our pumps have already been working very close these limits (~ 18,7 m/s). Therefore we have realized that it shouldn’t be a good idea changing its speed. Besides, the power consumption has gone so high (every time the pump works at low capacity) that lead me to make some attempts (like reduce the shaft speed) to decrease the power consumption.

Well, since my first figure does not seem to be a good idea, I would like to know your opinion about recirculation when the capacity fall inside this limits (0-50 or 50-100%).

I look forward to hear from you soon.

Best Regards,

lcms

 

[Torricelli]

lcms,

you should obtain a huge improvement by installing a pre-condenser handling the vapours while the (very smaller) vacuum pump will handle the non-condensables only (just the nearly constant leakage). A spray type condenser is relatively unexpensive. You may use the same liquid for the condenser as for the liquid ring vacuum pump.

ET

http://www.vacuum-guide.com

 

[quark]

Minimum velocity for a non collapsing condition can be obtained by balancing weight of the sealing fluid inside the vacuum pump at any instant of time to the centrifugal force. Use the ring inner diameter for redundancy. (i.e v = (rg)^1/2) where r is the ring inner radius and g is acceleration due to gravity.

I have been speaking to various vsd manufacturers and vacuum pump manufacturers for past 5 years regarding speed control but nobody gave strong suggestion sofar.

My suggestion would be to increase no. of pumps and operate them one by one.

If your main aim is to recover condensible gases then a precondenser may not be of great advantage.

Regards,

 

[abcmex]

a general information:

http://www.graham-mfg.com/downloads/27.pdf.

As Torricelly mentioned the key of a vacuum system is to have only to handle the inert gases. Additionally it is as much as important to have a tight system.

The sequence to add tanks to the system should be looked at otherwise the vacuum would break each time. Perhaps a two vakuum header would help (one high volume low vacuum and a second at a higher vacuum).

RGS

 

[lcms]

Dear Friends (Torriceli; Quarc and abcmex);

I really appreciate your valuable help, with all great suggestions herewith presented.

The article suggested by Mr. abcmex. (Write by Mr. Joe Gagliasso - Graham-mfg) confirm my previous alternative of controlling my liquid ring pump capacity recycling part of the flow when I have minimum requirements.

The suggestions of Torriceli and Quarc are also great, but unfortunately don’t apply to my case. Let me explain better: - I use the condensable gases recovered on the water seal by using the mixture (water+condensate gases) later on my process (that’s why it isn’t a good idea to remove them before). Regarding the use of smaller pumps, as suggested by Mr. Quarc, I had already thought about it, but it will depend on my production schedule and how this affect the performance of the vacuum pumps; I mean: - I will see what happens with the bigger pumps, and if the flow remains near one of the control limits (0, 50 or 100%), then I will decide if I have to install a new smaller pump, and how should the capacity control be done.

Sincerely Yours,

lcms

 

[quark]

I don't totally rule out precondensing the condensibles. Generally when the ratio of partial pressure of vapor to the total pressure is more than 0.5, precondensing is economic.

Non condensible gas recycling should be limited to the minimum flow rate required by the vacuum pump(you can read it from the performance curve) to prevent discharge side cavitation, to optimize power consumption(incase you precondense the condensibles).

Sequential pump operation can be done by a vacuum transmitter in the common suction header.

I would strongly suggest speaking to Sterling Fluid Systems.

Regards,

 

[Torricelli]

Dear Quark,

You could even recommend a precondensation.
There is 3 good reasons here to do it

1. – Reduce the volumic flow (= use a smaller vacuum pump) taking advantage of the condensation:
the liquid ring vacuum pump is a volumetric pump. The volumic flow to handle after a precondenser will be a small part only of the total initial volume.

2. – Better pump efficiency with a cool seal liquid:
absorbing the heat of condensation in the pre-condenser, we will reduce the heat load in the vacuum pump,
= decrease the effective ring temperature
= increase the efficiency of the pump, which again leads to a smaller pump.

3. - keep the seal flow within an acceptable range:
lcms told us that he was running the pump relatively slowly (18m/s tip speed) but however he was concerned about the high power absorbed. The explanation for the high power absorbed is not the speed but very probably the fact that the pump is not designed to handle this amount of liquid (seal liquid + condensate) through the discharge port.

Also the recovery of the condensate is not in contradiction with the precondensation.

Hope it helps

ET

http://www.vacuum-guide.com

 

[pmwinter]

ICMS,
Trying to centralize your vacuum pumps on batch vessels (batch reactors) can be difficult if you need to maintain vacuum on all the other vessels if you are bringing even just one on line. If you are trying to maintain vacuum on the 8 vessels at, say 25"HgV, but need to bring a large ractor on line (from atmosphere), there will be a very high, short term load on the central system. Unless you drastically oversize the vacuum pumps, you will get a spike in pressure on the 8 vessels, unless you control the vacuum on each vessel (limit the flow, and thus the evacuation rate on the 1 vessel). Also, unless you design the control system properly, you could get backflow(contamination) into the other vessels.
You should really contact someone with good application experiance.

Wintek Corporation.
PMWinter

 

[katmar]

Hi lcms,

The fact that you re-use the gases condensed in the seal water is not a reason to reject pre-condensing. In fact you are very lucky that the gases can be re-used after they have been absorbed into the water.

This set-up is very common in the sugar industry, where boiling pans and evaporators operate under vacuum. The vacuum is almost always created by a combination of a direct contact (i.e. barometric) condenser and a liquid ring pump because, like your process, they can feed the "sweet water" back into the process.

The condenser actually does most of the pumping by condensing the condensible gases, and the liquid ring pump removes the incondensible gases. Direct contact condensers (as pointed out by Torricelli) are relatively inexpensive, and they can operate at extreme turn down rates.

You can use water as your direct-contact liquid in the condenser, just as you would in the liquid ring pump, and feed this into your process without any changes. The water from the vacuum pump could still be fed to the same point to recover any slippage from the condenser.

My own experience with liquid ring pumps that have to be designed for very wide operating rates is that it is very easy to over size them and finish up with a severe cavitation problem. The best way that I have found to solve this problem, and to control the pressure accurately, is to bleed air or recycled process gases back into the suction of the pump. This is confirmed in the article by Joe Aliasso of Graham Manufacturing, and I see he calls it an "anti-cavitation or suction pressure control". **Never** control be throttling the gas inlet to the pump as this forces it to operate at maximum vacuum all the time, and is a recipe for cavitation.

With a pre-condenser you will most likely find that you need only one liquid ring pump and your control system will be very simple.

A disadvantage of direct contact condensers is that their running costs are fairly high because of the water circulation, but this will be off-set by having to run only one liquid ring pump.

 

[lcms]

Dear Friends,

I think I can see now your point of view regarding precondensing direct contact to help me to better control the liquid ring vacuum pump by reducing its flow requiring. I will try it as soon as I can (now I am testing the pressure control for each vessel which pmwinter have mentioned, and it seems to be allright thus far).

Thank you very much for your kind help, Gentlemen.

Reagards,

LCMS

 

[rmw]

As already mentioned in earlier posts, the liquid ring is limited by the inlet temperature of the seal water. Pre condensation is beneficial as mentioned for capacity reasons, but it also reduces the heat load to the LRVP.

All the latent heat is absorbed by the seal fluid as a result of the condensation. This is in addition to the mechanical heat load input due to the driver and internal friction of the liquid ring.

If you have a limited supply of colder water, having seen some plants that could use limited amounts of "city" water or well water for the process, but you have ample amounts of "not so cold" water, from a variety of sources, cooling towers, cooling ponds or pools, etc, then you can do the pre condensation, direct or otherwise, with the warmer source water, and still greatly limit the vapor and heat load to the LRVP's, helping their efficiency, and reducing their tendency to "derate" due to higher seal water temperatures.

Have you ever looked at a "derate" curve for your LRVP?? They aren't happy campers at temperatures much above about 60F or so. A LRVP would just as soon boil its own seal water, and pump that vapor to the higher pressure (atmosphere in this case) as to pull vapor and non condensables from your process.

The precondensation recommendation was a good one.

rmw