Help! Rankine Cycle Turbine: Finding Low Cost Feed Pump Advice

[kartsyont]

Hello! first time user, in desperate need of advice:

My company (start up) has built a 5 hp Rankine cycle generator with a thermal efficiency of about 33%, it works fantastic in an open cycle, however 6 engineers and many pump specialists later, and i am still at a loss...

My system uses r-134a as a working fluid, and i am looking for a pump that is relatively inexpensive that can pump liquid refrigerant (a very low viscosity fluid) at 125 psi with a variable flow rate depending on the btu output of the fuel i chose to burn, which can vary...

the problem is, i cannot find a feed pump that can prevent slugging out of the turbine...the turbine itself cost $350.00 to build, but i cannot find a pump small enough within the 350 dollar price range, i have looked at

1. double acting piston pumps: usually are too high pressure for my application (2000 psi+)
2. Variable displacement, pressure compensated vane pumps: cannot tolerate low viscosity fluids
3. pressure compensated piston pumps: also not low viscosity tolerant, expensive, and power demanding
4. 3:1 pneumatic diaphragm pumps: not truly independent pressure and flow, and how would i drive it? pneumatic line from the boiler to the condenser side? i think i would need a secondary pump at that point
5. Variable displacement, pressure compensated gear pumps: longevity and viscosity intolerant below 70ssu
6. Sliding vane pumps: PERFECT IN EVERY WAY FOR MY APPLICATION Longevity, efficiency, control of flow rate and pressure independent of each other AND can pump low viscosity fluids like R-134a. however, the pump in my pressure and flow rate range is $2500.00! more expensive than the entire system boiler and all!

i want the pump to start up and maintain the pressure differential while the turbine spools up to speed, after which, i would like the feed pump to be run off of the turbine via a belt and later a drive shaft. my pressure differential is 125 psi and my flow rate for the r-134a as a gas is approx. 50L/min.

I used an air compressor pump in a hermetically sealed container, the pump is supposed to have a max pressure output of 250 psi, but as soon as the pressure is equalized on both sides of the turbine, the pressure on the low side is too great and the piston pump stops working until i depressurize the condenser coil (low side heat exchanger), I'm at only 80 psi so it's not even pumping liquid yet and it is already starting to seize! I am sure i need a more powerful motor to drive the pump but i do not think that is the whole story. Does anyone know the answer?

also, i am almost positive that this pump will not work for a liquid, but is there a way to modify it? i have a large external one way valve, should i remove the bash valve in the cylinder head and upgrade the electric motor? i would like to use this pump if i can because it has my flow rate and it only cost $8.00. but if not, what pump would work?

was also considering the use of a air conditioning scroll compressor as i read it is pretty tolerant of liquid, however, tolerance implies a short term solution, i need a pump that will last, can this work or am i still up a creek without a paddle?

this 8 dollar pump was able to pump brake fluid in a test, which is many times more viscous than liquid r-134a, that is why i chose to use it, but that test was conducted at atmospheric pressure.

 

[edgepflow]

An eccentric disk pump may be suitable due to low NPSH requirement and also ability to handle a liquefied gas. Similar idea as your sliding vane. You may require either eccentric disk or sliding vane and have to front the cost.

 

[Vespa71]

The feedpumps of steam-turbine-plants are often multistage-sentrifugal-pumps. No seals required, just sentrifugal-force driving the liquid. Must be well balanced, but has cheap construction. When flow decreases in, so does power-need (less to throw). Vacuum-cleaners drives air-suction by similar principle, i beleve. I don't know if such pumps are aviailable in your scale though.