PDA

View Full Version : Could a compressor work in reverse?



hookway
01-10-2008, 06:13 AM
I have an application that provides compressed gas. Is it conceivable that a compressor could be modified to operate in reverse? That is, to accept compressed gas and turn the pistons, I expect the values would have to be altered. I would greatly appreciate any feedback on this topic.

Grizzly
01-10-2008, 06:57 AM
Personally
I can't see it being easily achievable!
There is a lot of things to consider besides just the valves.
Which in themselves are a major hurdle as they are nothing more than a one way valve.
Simply put the whole gas flow through, and the balance of pressures therein are direction specific.
Rather than a recip what about a Turbo?
More importantly why?
Grizzly

US Iceman
01-10-2008, 01:59 PM
Is it conceivable that a compressor could be modified to operate in reverse?


Yes, the item you are describing is called an expander.

A compressor compresses vapor from a low pressure up to a higher pressure where the vapor can be condensed by cooling.

An expander has high pressure gas flowing into it and by extracting work (high pressure to low pressure expansion) power can be created by a generator.

As a simple concept consider these points:

A compressor has an electrical motor to provide the power.

An expander has a generator to create electrical power.

nike123
01-10-2008, 03:32 PM
http://tinyurl.com/4q6ey6

Grizzly
01-10-2008, 03:54 PM
Guys!
Am I being corrected, in that you can convert a recip?
(not a criticism I am interested)
Or are you agreeing that this type of kit exists?
Grizzly

NoNickName
01-10-2008, 05:09 PM
You need reed valves upside down.

chemi-cool
01-10-2008, 05:14 PM
Its a generator.
I doubt if you can get something out a recip. compressor.

US Iceman
01-10-2008, 06:17 PM
A recip. compressor would have to have the valve positions reversed to operate as an expander. So... you cannot use a standard refrigeration compressor for this purpose without a lot of work.

A centrifugal compressor becomes a turbo-expander.

A screw compressor becomes a rotary expander.

Neither have valves to worry about.

hookway
01-10-2008, 08:48 PM
I don't want to use a turbo because I expect it'd need more pressure and there may be problems sealing the vanes. I appreciate the feedback, I considered the option of using external valves. I notice turbo expanders seem to be part of very large refrigeration units but they don't seem to be an easily available part. It seems using a pneumatic tool is a good idea, depending on the friction - any comments?

Brian_UK
01-10-2008, 10:43 PM
Not quite in the same league but I know in the old days people have converted four stroke engines into steam engines by building their own valve gear for them so I suppose anything is possible (but is it worth the effort?)

US Iceman
01-10-2008, 11:24 PM
hookway,

There are some firms who make various expanders depending on the capacity requirements. However, you need to remember any of these will not be a cheap piece of machinery.

Are you just looking into expansion of the gas only or expansion with power recovery?

Alex Harvey
02-10-2008, 03:14 PM
Very unlikely that the compressor would be able to be converted to run in reverse. So the answer is no to the question.

But there are 'gas expanders' which have a different function to generate power by reducing the gas pressure. Which may be what you have in this case.

In theory its a compressor in reverse. But in practice the engineering is very diiferent.

hookway
02-10-2008, 07:07 PM
hookway,

There are some firms who make various expanders depending on the capacity requirements. However, you need to remember any of these will not be a cheap piece of machinery.

Are you just looking into expansion of the gas only or expansion with power recovery?

Right. I will see how far I can get first, having arranged for the valves to be reversed in a compressor. I am interested in power recovery, with a small capacity as a test of principle.

John_Galt
04-10-2008, 08:53 PM
<snip> It seems using a pneumatic tool is a good idea, depending on the friction - any comments?
For small scale testing, a pneumatic tool might be a good choice. Not much engineering required.

The one issue with a pneumatic tool will be that pneumatic tools are "total loss" devices. In normal operation, the compressed air that goes in is released to the atmosphere.

If you're compressed gas source is of a nature that venting it to the atmosphere isn't an issue, then a pneumatic tool would be a good choice.

If you need to recover the gas after expansion, that might be more difficult with a pneumatic tool. It's not impossible, just difficult. You could design a sealed chamber to hold the tool, with a seal around the shaft to get the useful rotational energy (work) out of the chamber. A well sealed plexiglass box would probably work if you needed it.

The other issue might be what the gas you are using will do to the internal components and seals in the pneumatic tools. They are a little forgiving of certain contaminants. Oil is not a problem. Moisture, while not good, won't cause an immediate failure. Pure nitrogen (with a bit of lube oil mixed in) will run a pneumatic tool basically forever. CO2 works as well. But I wouldn't run anything corrosive or or even slightly inflamable through a pneumatic tool.

And pneumatic tools are designed to have some lubricating oil run through them. A few drops in the inlet when you first hook them up will satisfy the lubricant requirements for a while. A better solution is to use an oil metering device that adds a drop or two of oil to every few cubic feet of air that passes through. Do an internet search for "air tool oiler" for a listing of such devices. Small ones suitable for use with a single device are available at reasonable prices.

SIGNSTU
01-11-2008, 01:59 PM
Remember oil pumps are direction sensitive!

Peter_1
01-11-2008, 07:21 PM
Remember oil pumps are direction sensitive!

Not those I know

Alex Harvey
03-11-2008, 05:27 PM
Would be interesting to see how you get on with this theory test and the swap of the valves. You may not get much usefull power. Is this a semi-hermetic compressor?

Keep us all posted and good luck!

glenn1340
17-12-2008, 09:29 PM
My field of work is mainly in air compressors. The problem as I see it is a recip may stall when the first piston reaches bdc. Probably easiest to use a dedicated air motor (that`s supposing it`s ok to use on with the gas available).
Main consideration, of course, is price. I don`t `spose these things are cheap.

Remember...,
The pen is mightier than the spanner

wambat
18-12-2008, 12:54 AM
This may be of some interest to you:

Scroll compressors are mass produced for the refrigeration and air conditioning markets. Energetix takes these scroll compressors and with minor modifications run them in reverse to operate as a scroll expander. When in normal operation they are driven by a motor to compress a gas, however, once modified to operate in reverse compressed gas is injected into the scroll which then expands, rotating the scroll expander which in turn drives a generator to produce electricity

WINJA
18-12-2008, 06:51 AM
you maybe able to use a 471 supercharger off a truck , i bought one for my RX3 coupe many many years ago and for a laugh put it on the end of the exhaust to see what happened it spun quite fast and had a lot of torque suprisingly, the one i had was a staright lobe off a 2 stroke diesel and it had sealed bearings, im just throwing this out there in case your looking for a cheap do it yourself type option im not saying it will work

3dtooling
22-12-2010, 04:15 AM
I have been toying with the idea of making my own turbo expander gen unit for a few months for an ORC unit to make electrical power, it would be nice to start with a good process engineer, CAD designer, and then after all the math is hammered out I could just build a unit, I have a full CNC, shop with mastercam, fab shop that we could make one of these units, any takers for the engineering? 100 KW, gen set, ORC, using biomass boiller with an oil loop.

NoNickName
22-12-2010, 08:59 AM
Been there, done that. I reversed an open screw compressor, by inverting the thrust bearings. No need to design anything: just reverse the bearings and the pipes.
It works. It's not optimised, but it works.
On the shaft you can connect an alternator, or another open drive compressor and a double-shaft motor.
Since the mass flow is the same, the compressor and the expander are the same unit.

DTLarca
22-12-2010, 10:46 AM
Reversing the valve plate on the recip wont work.
Here's an piston steam engine animation:
http://science.howstuffworks.com/transport/engines-equipment/steam1.htm

chilliwilly
22-12-2010, 10:05 PM
If your trying to convert a compressor into an electric generator. How are you going to cause excitation from the AC induction motor to enable it to generate power?

NoNickName
23-12-2010, 12:26 AM
If your trying to convert a compressor into an electric generator. How are you going to cause excitation from the AC induction motor to enable it to generate power?

Are you asking me or the poster?

Dacien
28-12-2010, 03:42 PM
Just remove the valve, does not need

chilliwilly
28-12-2010, 06:19 PM
Are you asking me or the poster?

Anyone involved in this thread, but if you know the answer, then please enlighten us all.

If your generating power, stick to the tried and tested methods. ie turbines and other basic prime movers. Their design efficiency will far exeed any compressor operating in reverse.

NoNickName
28-12-2010, 07:39 PM
I'm not sure I fully understand your question, but I will try to explain. A motor sized for the compression work is connected to the compressor (open type in this case). The motor will have two shafts from both ends.
The other shaft is connected to the expander. As soon as the expander starts to make work, the load on the motor will reduce correspondingly. The motor power input can be reduced substantially.

Alternatively, the expander can be connected to an alternator or DC generator (depending on the size of the unit), in turn connected to an dc to ac inverter, and returned into the grid (provided the necessary equipment for connecting to the grid are installed and approved by the grid operator) or used in the plant as auxiliary power supply.

chilliwilly
28-12-2010, 07:46 PM
Expander... do you mean a receiver?

Can you please elaborate what you mean by expander.

NoNickName
28-12-2010, 08:34 PM
The refrigerant once liquefied (condensed) is expanded through the TXV. The amount of work that the expansion of the refrigerant is able to return, is equivalent (minus losses) to the work of compression.
If instead than a TXV, an expander is used (a screw compressor working the other-way-around, for example), the work of expansion can be converted and returned to the system of plant.

http://www.totalenergysolutions.ca/images/expander.gif (http://www.totalenergysolutions.ca/images/expander.gif)

chilliwilly
28-12-2010, 09:32 PM
What would the energy in over the energy out ratio equate to? It does look like a very innefiecient refrigeration system and power generation set up. I've only ever come across a flooded evaporator system like this years ago what we decommisioned because it was past its best. I wouldn't have thought of piping it through an expander/compressor of any kind for fear of causing a pressure imbalance.

Question

In such a system, how does the low pressure vapour condense back into a liquid when it passes through the condensor to fill the receiver? And how does the refrigerant superheat and expand from a liquid to a gas when its under pressure in the evaporator? And do I take it that the pump isn't a recip pot, that's if the condensation of the refrigerant turns back to liquid.

mad fridgie
28-12-2010, 10:45 PM
the best way to think of this, is the expander is an air motor.
To make it spin you need high pressure air (or refrigerant "R" in this case)
After the air(R) has passed through the motor it is at a lower pressure, you can determine the lower pressure, in the same way you determine the condensing pressure on a refrigeration system. (water temp, surface area, flow)
You now have liquid refrigerant, it is easier to pump a liquid than a gas/vapour (efficiency per mass)
the liquid is pumped to a pressure (as high as possible bur will still vapourise), where a waste heat stream (exhaust from an engine for example) to turns the liquid into a vapour. back to the expander (motor)
Just like a refrig system you choose where you want the evap and cond pressures to be

chilliwilly
29-12-2010, 12:48 AM
Yes I understand what is trying to be acheived here, its nothing new, there are loads of crazy ideas these days like these. There's an pneumatic motor that can be coupled to a high ratio block and tackle to cut down on the time it takes to lift a heavy object. And there's an hydraulic motor for low speed high torque loads. theses are tried and tested and they work.

The amount of losses and inefficiencies just aren't worth considering building a plant such as this, that would badly generate power and or cool a space. Its almost as if a suggestion is being made that the theory of perpetual motion has been acheived.

I can see how the high pressure refrigerant would cause the expander to rotate and output torque on a shaft/coupling. But if the expander is being used as a metering device, how does the condensor reject heat and subcool the refrigerant and condense it back into a liquid? There would have to be a cascade system to make the condensor work any where efficient enough to condense low pressure vapour back into a liquid.

The PT relationship doesn't correspond. And how does the liquid absorb heat in the evaporator and expand into a high pressure vapour? the superheat setting must be non existent. The fact that its under pressure will make it try to remain a liquid and defy the basic laws of Graham and Dalton, and the theory of refrigeration.

mad fridgie
29-12-2010, 01:06 AM
May be I should but numbers on this,

Water at 30C, condensing 40C (so far so good)
Elevate liquid pressure to boiling point 90C evaportor, heat source 100C.
The equation basically: can your expander produce more energy than the pump and condensor fans use! this what the world is playing with, how to convert low/medium grade energy in something that is useful.

NoNickName
29-12-2010, 01:10 AM
Actually, the refrigerants used in these type of circuits have a very favourable T/p ratio.
If you've got warm process water and an air cooled condenser, theoretically you could condensate into the air cooled condenser, pump the liquid into the expander, and evaporate the refrigerant into a process-water cooled evaporator.
It is infact inefficient, but these days we are scrapping the bottom of the barrels.
For some types of refrigerants, like steam itself (R718), the evaporation takes place at 100°C @ 1 bar, and condensation for example 180°C @ 10 bar.
You've got a 10:1 expansion ratio with a very favourable temperature approach: the heat sink can be anything colder than 180°C, and the heat source any fluid warmer than 100°C (sun thermal panels or process steam).
With ammonia, a system with -30/+50 would have a 20:1 expansion ratio with a liquid feed pumped circuit. Subcooling can be controlled in the condenser, but superheat is always obviously zero.

chilliwilly
29-12-2010, 01:40 AM
May be I should but numbers on this,

Water at 30C, condensing 40C (so far so good)
Elevate liquid pressure to boiling point 90C evaportor, heat source 100C.

Not trying to be funny or anything.

What I'm reading, water/refrigerant enters evaporator @ 40 C from condensor/receiver and absorbs 50 C of air temperature to boiling point 90 C, when its probably at more than 1 atmosphere. Then rejects 60 C of heat through the expander/compressor subcooling down to 30 C. I'm sorry but that doesn't make sense.

Probably better off with a standard system but with a water cooled condensor. And heat/superheat the exhaust water to make hot water for a domestic hot water tank. Or steam for a steam turbine generator. Similar to a cooling tower setup at a power station for water reclamation and boiler input water temperature efficiency. The losses are much less and are worth considering.

Anyway I think I'll just move on from the thread, its getting too sticky for me.