turboexpanders 1

[tomato5]

could you please assist in the following. I was hoping for some reliable experienced information rather than "I think "

1. A turboexpander driving a centrifugal compressor on the same shaft. Does the inlet gas to the turboexpander have to be fairly low in temperature, or can the temp be high say 250 F?

2. Is the Joule-Thompson effect the basis of all turboexpander design? If so, is the exhaust gas from the expander always at a very cold temperature whether the plant is PTA, Air Separation, Nitric Acid, etc?

3. Is there a difference in the expander construction compared to say a European mfr versus an American one. I was referring to things like radial flow and axial flow, reaction type versus impulse type (if there is an impulse type)?

4. Is it possible to have an arrangement where say a three stage expander, integrally-geared, drives a three or four stage compressor integrally-geared. If so, is it right to say that there will no longer be a single integral shaft?

5. I read in an article that in an expander-compressor combination there is usually no need for an anti-surge system. Is this correct? If so, why?

6. In a train schematic I have seen an expander compressor steam turbine gear box electric motor what is the purpose of the steam turbine if the motive force is from the expander what is the electric motor doing? I may be misleading you here because I did not have the time to read the article in more detail

Several questions I would appreciate some answers I do understand that this may take more than five minutes

thank you

 

[dcasto]

Inlet temperature can be most anything you want, look at a turbocharger on a car, its a high temperature inlet to an expander with a coupled air compressor. As a matter of fact, the higher the temperature, the more work you can get out of the expander for the same mass flow and differential pressure.

An expander is not a joules Thompson process as the JT is a pressure reduction with no work removed.

Expanders are the same everywhere on earth or in the universe. The mechanical design is chosen to get the most cost effective capture of the available work.


Such a machine with multi stages could be built but boot strapping it up wouldn't be fun.

There are combinations of flows that can get an expander compressor package into surge so everyone I've installed includes an anti surge on the compressor.

The system you described is possible. If the system needs more energy than the expander can produce, then you can add more work via any other input. Again I'd question making such a device because of the complicated controls and gearing to get all those items to the single output speed, a 20,000 rpm expander with a 4000 rpm steam turbine and a 1800 rpm electric motor going to a 12000 rpm compressor?

 

[turb1]

For note number 5, what you describe is called a compander. They are typically used when the motive force from the expander is not great enough to drive the compression stages, so an additional driver is used (steam turbine, gas turbine, electric motor). They are not very common, and the stages are mounted on a large multi-pinion gearbox to accommodate the different speeds required for each process connected to the compander. Some companies advertise as many as 10 stages on a single gearbox (mounted front to back on the same pinions, so a maximum of 5 total high speed pinions).

I think dcasto hit the nail on the head on the other points.

 

[tomato5]

I learnt something, also have some comments

A turboexpander I feel is strongly a joule-thomson phenomena joule thomson does not have to reach refrigeration, but there will be an appreciable reduction in outlet temperature as compared to inlet temperature

Multistage integrally-geared compressors and multistage integrally-geraed expanders are an easy item now, widely used and accepted. There is no bootstrapping

In a PTA plant, like in a nitric acid one, or an air separation on, high-speed expanders, compressors, gear-boxes, electric motors(not high speed,to provide initial torque to start the train) are accepted components. Obviously, determination of appropriate speeds of the different components to give the needed process output is a matter of detailed design and consideration.

Thank you and Regards

 

[25362]

This is not a J-T process. See, for example:

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

 

[rotw]

Tentatively:

In a gas expander, I believe this happen:
1/- Aerodynamic/Friction Losses (as the gas flow through the expander, etc.) which causes by the way some temperature increase.
2/- Total/overall Pressure drop due to losses. Seems to me J-T process co-exist and contribute to some temperature decrease.
3/- Gas 'isentropic' expansion largely predominant resulting in power generated at shaft. (overall combination of 1/ and 3/ makes the expansion process polytropic instead of isentropic)

I do not think that pressure drop in 2/ lead to generation of work/power this is why It should be associated to J-T effect which I believe to be very significant although J-T expansion is normally defined as a throttling/expansion through a valve/orifice. I would imagine that to figure out this effect, we may keep the rotor at stand still and pass the gas through the expander and measure the resulting pressure drop from inlet to the outlet.

Any corrections or thoughts are welcome.

 

[25362]

Philosophically assuming such a simultaneous J-T effect takes place, how can one be sure, in every case, that the expansion is not outside the inversion curve (i.e., heating up) or near it, i.e., with almost no change in temperature?

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720020315_1972020315.pdf

 

[rotw]

25362,

I assumed that in real applications, you don't expand hydrogen or helium and I had in mind some any exhaust process gas.

But you are right I should not presume in which way the temperature will change when supposedly such J-T effect takes place.

 

[MortenA]

Guys - get a grip: An expander IS NOT a JT process - a valve is. A JT process is an _enthalpathic_ process - that is all the "work" performed when letting down the pressures is "absorbed" by the fluid. Any fluid going through a valve heats up for this reason. A liquid, however, does not expand so it get a little warmer. The expander removes (some) of the work - and is therefor in the ideal world an isentropic process. In reality there is an efficiency - so its not truen isentropic - but somewhere in between.

 

[MortenA]

spotted an error in my rant (always a bad sign) - it says_enthalpathic_ BUT it should say isenthalpic

 

[rotw]

MortenA,

I think you did not read my post completely. Let me put it again:
00
Consider a turbo machine, say a radial expander impeller.
Keep the shaft fix, so that it cannot rotate and no power can be generated.

Let the gas flow through the compressor. Do you have a pressure drop yes or not ? yes you have.
On that pressure drop, do you have an expansion of the gas. Yes.

When a gas expand the temperature should change and why should this be different than a J-T effect through a valve ? I am guessing this to be probably case and I wanted that to be possibly debated here, but your arguments are not addressing this point on that terms.

Now, release the shaft. Gas is expanded and that expansion is not J-T. It is a polytopic expansion.
Overall I am saying that most of the expansion is polytropic and co-exist a negligible fraction where J-T effect takes place.

By definition, J-T is expansion through a valve, orifice, yes clear but that is not the point.
The point is the physics behind the phenomena much more than terminology/definition issues.

I might be wrong of course, So I would be glad to have some insights here.

 

[MortenA]

well there you have it - no i didnt study your post sufficiently good (thanks for putting me straight ):

When the shaft is fixed there is no Work removed, you have an isenthalpic process like a valve or an orifice and the process is thus a JT process. Release the shaft and you get (in principle) a isentropic axpansion (with an efficiency factor). You could say that the in-efficieny is similar to JT - at least the part that is not Associated to mechanical losses in the expander.

 

[dcasto]

Morton, don't give up.

Rotarw, I assume you meant "let the gas flow through the expander" not compressor. By locking the rotor, you just made it a VALVE, a JT vale is ANY restriction that causes a pressure drop and the UNIVERSAL meaning of a JT process is just that. You can install a 1/16" plate in a 10" line and open the outlet to atm with a 100 psig upstream and that GAS will get cold. OH, but not always, yes at some point on the methane Mollier diagram wayyyy up high, a pressure drop can actually make the gas get warmer, BUT it's still a JT effect. The jt effect, THERE is NO delta H, that is what it all boils down to.

I guess you could argue with your example of the locked rotor that an 80% efficient expander is actual a 100% efficient expander on 80% of the gas and the other 20% of the gas is actually going through the JT effect. Naw, lets stick with the rules describe by Joules and Thompson, an 80% efficient expander has an 80% change in delta H as compared to a constant s

I ain't picking, but that's engineering its all about the terminology and practice.