Innovative Cavitation Solutions for Low Volume, High Pressure Pumping

I used a Grundfos central heating pump for this purpose (impeller) however it only seemed to make the problem worse.

A central heating pump will have a maximum pressure rise of, say, 0.5-1 bar, which is small in comparison with the 20 bar pressure rise required of the gear pump. That it made the situation worse is therefore quite understandable.

There seems to be a problem with providing enough net positive suction head that cannot be seen here so far. Someone more knowledgeable is needed to comment on the characteristics of R134A at these temperatures and pressures, maybe someone with access to a copy of Mayhew & Rogers: Thermodynamic Properties of Steam - in the back of this tome there is a useful table relating to the pressure-boiling temperature characterisitcs of R134A. I'd lend you mine, though it's in a box in my attic, which is currently occupied by the Resistance.

I think you fighting an up hill battle because your not going to get R134A to stay liquid at 100c. What is the purpose of having the liquid at the generator?

And where at the generator? Are you trying to reduce heat?

Hi design engineer,

Basically I am responding out of interest because I have generally had problems with low pressure systems; this in comparison to my own experience with hydraulic systems which start at 200 bar and reach 2,600 bar.

I kow nothing of refrigeration systems fluids, but for what its worth, I would envisage that what you have attempted to do so far is reasonably sound in that you have tried to boost the low flow / hp pump. This is similar to the UHP systems when we use triplex pumps. However, I have never had to need to boost a gear pump as they usually can pull a negative head without priming.

In stating the obvious it then comes down to the choice of pump. Presumably a standard refrigeration pump will not give you the output pressure, so you could consider an intensifier or an air over fluid design from the likes of CAT or SC. Depending upon the larger input piston to smaller dia output piston these could meet your needs.

If you require any further info on these please reply. I have very limited access to this blog but will endeavor to respond ASAP

sorry, I meant to advise Haskel, not CAT

Since reading replies from other contributors, it seems that the fluid will turn to gas at 100 deg. C. If this is so and is acceptable, the pumps I have meantioned will still work OK with gas with slight modifications.

You could seek help from

www.staffshydraulics.co.uk

who diistribute SC pumps and manufacture thier own hydro and gas test systems

"The reservoir is above the generator, so I initially tried to use a gravity feed system. I constructed a transfer volume (vertical length of pipe with ball valves on either end) The plan was simple (1) open the top valve and fill the transfer volume (2) close the top valve and then (3) open the bottom valve to drain the fluid down into the high pressure generator before finally (4) closing the bottom valve and beginning the cycle again. This didn't work very well as the fluid vaporized upon entering the hot pipe and the upward moving bubbles inhibited the flow into the pipe. This also meant that I was constantly losing valuable thermal energy back to the low pressure reservoir".

I think the gravity feed system will work but add a loopback pipe (return pipe). Connect this pipe from the generator to the point just after the top valve. Provide this loopback pipe with another valve. Then revise your system of opening and closing the valves. After step 2 of the above procedure, open the loop valve,(this will pressurize your pipe to 30 bars but entering from the top), then open the bottom valve to drain the content. Repeat the same cycle.

The present system actually pressurized your gravity pipe to 30 bars also and would flow back to your reservoir chamber the instance you open the top valve. Another alternative would be to add a vent pipe (with another ball valve) to your gravity pipe and depressurize this to at least 10 bars by venting to atmosphere before filling up again. Additional pressure gauge have to be provided.

.

When you say you have a low pressure reservoir I presume for a safety point of view it will sustain the working pressure on the outlet side of the pump?

If so you could arrange a small bleed back line to the reservoir with regulation to keep a head pressure in the reservoir high enough to give liquid delivery to the pump inlet.

You need to operate your pumps at a lower temperature. Also, the use of a phase seperator before the pump will ensure your suction does not see vapor.

"You need to operate your pumps at a lower temperature. Also, the use of a phase seperator before the pump will ensure your suction does not see vapor."

Pressurize your supply tank/cannister, then you can forget the phase separator as you will have a net positive suction head at the inlet of the pump at all times.

Dear:

At 30C and 10Bar you are very close to the liquid-vapour saturation line.

At 100C and 30Bar you are in the middle of the liquid-vapour region.

Then, it is impossible to pump the refrigerant without having vapour bubbles in it.

This is the typical problrm of closed power steam circuits.

To reach 30Bar it is neccesary to add cooling to the pumping system to avoid the vapour region (in the P-H curve). Temperature will be around 75C.

If there is no small pumps, try adding recirculation back to the tank.

The P-H curve can be found in http://www.nzifst.org.nz/unitoperations/appendix11a.htm

Hope this will hel you

Cheers

Design Engineer,

I am having some difficulty understanding why you would be experiencing suction condition induced cavitation. My guess is that you are seeing vapourisation due to heat input from your pumps. I have only looked up one source on the web, but if it is correct then the vapour pressure of R134A at 30 degC is around 7.7 barg. If this is correct then you have at least 6 meters NPSHa plus whatever the difference in elevation is. This should be more than enough as the line losses at this sort of flow would be negligible.

I would not think that a gear pump would be suitable for this duty as the viscosity is very low so the slippage (and thus heat input) could be substantial depending on capacity.

I would guess that the centrifugal pump that you installed upstream would have a BEP flow substantially higher than that of the gear pump so it would be pumping very much to the left of the curve and at very low efficiency, again resulting in heat input to the pumped liquid leading to vaporisation and making the situation worse. (Suction recirculation cavitation also possible).

As stated by others, you will be pumping into a vapour phase at the conditions that you mention, but this should not matter. I would suggest that for the low flow high head application such as this that a diaphragm or piston pump would be more suitable.

Have you considered using a Scroll Pump also known as a Scroll Compressor? This may have some advantages in operation that your other pumps do not.

Check it out:

http://en.wikipedia.org/wiki/Scroll_compressor

The most efficient pump and the least likely to create cavitation is the Helical rotor pump.

I am not saying that a Helical Rotor Pump won't work in this application, but aren't they used primarily for pumping semi-solid slurries and other high viscosity materials? If liquid refrigerant is the media being pumped, the HRP may be overkill for the application, even if the proper specs can be attained.

They are also used quite extensively in the chemical industry, dairy industry, irrigation etc. Why would anyone deliberately use a less efficient, less reliable pump, especially if nothing else is working?