ORC Waste Heat Scroll Expander DIY

[voltech444]

I bought a A/C scroll compressor from a 2005 Toyota Prius. I want to convert it into a expander to use in a ORC waste heat recovery system. If I can get it working on the bench, then it's my dream to put it on a hybrid vehicle, most likely another Prius since I own 2. The electricity it generates can be fed back into the battery to significantly increase the efficiency. From what I understand a scroll compressor and expander are nearly identical, the scrolls are the same, there's a check valve that needs to be removed from the compressor and a few other small modifications. These compressors for hybrid/electric vehicles are perfect because they have a brushless permanent magnet motor which also makes a great generator. From information I've read, the Prius compressor has a maximum output of 4.3Kw.

To increase efficiency I could connect another expander in series with the first to make a 2-stage expander system. I plan on using the condenser from the Prius and the 2 electric fans that come with it. I need to design and build the heat exchanger that goes in the exhaust. I also want to utilize heat from the engine cooling system, so perhaps a small heat exchanger that connects to the engine coolant system. I'm also comparing if I should have the refrigerant flow through the heat exchangers directly or use a liquid coolant loop with automotive anti-freeze. Having only the refrigerant flow through heat exchangers is a simpler system with less parts, so that's what I'm leaning towards.

The liquid refrigerant is pumped by a small pump (centrifugal I think, there's probably a few options for that) into the first heat exchanger (this exchanger has coolant from the ICE flowing through it, it goes before the ICE radiator to get the most heat) it then goes into the second heat exchanger in the ICE exhaust system. Then it is expanded through the scroll expander which generates electricity which is fed back into the hybrid battery. Then the refrigerant goes through the condensor where it is cooled down and turns back into a liquid and the cycle starts over.

The motors on these compressors are 3-phase permanent magnet AC. The AC power would have to be rectified into DC and then go into a DC-DC converter that can boost it to the proper voltage to feed into the battery.

These expanders could be used to generate electricity from any waste heat source and solar. There are literally thousands of these for sale from junkyards, and most of them are just sitting there, they are so reliable they are not having many failures in vehicles, and if they do they car is well over 200k. I've always wanted to get a scroll expander to build a waste heat ORC system, but they're expensive starting at $5k for a 5Kw. I see an opportunity here to make use of all these hybrid scroll compressors that are just sitting and deteriorating in junk yards (it makes me so sad , but excited if we could find another use for them)

I'm not a mechanical engineer, I run my own company repairing hybrid and electric vehicles, so I have to approach this by just building and trying things. I want to figure out as many of the details I can before I start putting things together, but this will probably be mostly by trial and error.

Any input into this matter is greatly appreciated, thank you for reading!

 

[jrmichler]

Sounds like a fun project. I suggest that you put together a system schematic. Indicate on the schematic everything you know about temperatures, pressures, and flow rates. Then take the expander and connect it to compressed air and prove that you can make it spin. Then measure how much electricity you can generate. Post everything, and we will help you take it further.

 

[voltech444]

Ok I will do that. I was also thinking of using compressed air to test; there's a few videos on youtube of people doing this (converting scroll compressor to expander and powering with compressed air) so it can definitely be done. I will put together a schematic and post it here; i'll also make a picture of the entire system. I'll start seeing what information I can get about the compressor and details on the flow rate and pressures. I would like to use R134a as the working fluid just because its safe, cheap and available. But there might be a better fluid I could select, i'll have to figure that out once I get the specifics. I want to get thermal insulating wrap to put around the exhaust manifold, maybe on the catalytic converter too, insulate all the way to the heat exchanger, and insulate the exchanger too.

There's a lot of information in these links to papers I found:

http://iopscience.iop.org/article/10.1088/1757-899X/147/1/012147/pdf

https://ir.library.dc-uoit.ca/xmlui/bitstream/handle/10155/142/Oralli_Emre.pdf?sequence=1

http://orc2011.fyper.com/uploads/Fi...xpanders for small scale ORC applications.pdf

https://academic.oup.com/ijlct/article/6/3/200/680918

 

[voltech444]

https://www.sciencedirect.com/science/article/pii/S1364032113000592

This one too

 

[NTL2009]

@voltech444 - Interesting, any updates?

These scroll compressors look to be pretty efficient ( ~ 70% in this application?), but how much heat can you expect to remove from the cooling system? I agree, exhaust gas would be a tougher nut for DIY - high temperatures, and need to avoid restrictions which would increase back pressure, so the cooling fluid would be easier/safer to tap into.

I'm just having trouble picturing 180F coolant coming in, and exiting at say 125F (absorbing ~ 1/2 the temperature delta of 180F to ambient 70F). Can you pull that much heat out (I really don't know)? But if you could, and if we assume ~ 25% of energy lost is in the coolant (probably less, I think some of that is lost through the engine surfaces), and we capture half of that, and get 70% efficiency converting to mechanical, and maybe 80% efficiency delivering it, we are at roughly (.25 * .5) * (.7) * (.8) = 7%. Not insignificant, maybe an added 2-3 mpg?

Is 125F coolant returned to the engine too cool? Would that hurt efficiency if the engine is designed to run 'hot'? I really don't know how far a standard radiator/fan drops the temperature, the only temperature I'm aware of is the exit temperature at the thermostat (before it gets to the radiator).