Control valve at the suction of centrifugal compressor 1

[Jack Nicholson]

Hi. I have a question. In our plant, sales gas at the top of column compress and throttle and goes back to column as a recycle stream. At the suction of compressor we have a control valve named PV-149. Is there anyone explain to me how this valve can be used?
P.S.: Compressor is centrifugal with fixed rotation speed.

 

[bimr]

That is what I stated above and you disagreed with:

"It is controlling the inlet flow, not the pressure. The pressure will vary slightly as the flow changes. As the flow decreases, the pressure will change as shown on the performance curve."

The reason to do this on the inlet side is that it is possible to save energy. You will not save the energy if you do it on the discharge side.

 

[rotw]

Lets reverse the problem as follows;

Suppose there is no throttle valve. The machine has fixed speed.

Process is working at its nominal capacity. Compressor is running normally.

Then Operator wants to increase methanol production capacity by 20%. What happens?

The gas is circulated into a big loop. If you increase the flow, pressure drop increases through the process loop.
Machine has constant speed. So with flow increase, discharge pressure reduces along constant speed curve. So this will not work.

What can therefore be envisaged, is to throttle the suction valve PV-149 at normal operation (and live with the wasted energy costs); say you throttle by 10% on normal case. Now comes the increase of capacity case, say the 20% extra methanol capacity. In this case, the valve is gradually opened, and so compressor generates higher discharge pressure ratio. In this way, discharge pressure corresponds to a known flow capacity level by knowing in advance system resistance and behavior. By manipulating the discharge pressure set point, the plant control can increase capacity to desired level as a bonus suction throttle valve is less closed so more energy efficient. We have to be clear PID is pressure control although indirectly the ultimate goal is flow capacity change.

Its also possible to develop similar thinking when sales gas composition is off spec etc. There would be a min and max case and throttle valve offers some flexibility. flexibility is limited, however and is at energy cost. The most flexible solution and energy efficient would be variable speed machine but that would be at higher capital cost.

 

[rotw]

Addition, there is no risk of vacuum whatsoever (sealing system, explosive mixture etc.) as the suction pressure is at 30 bar.
Pressurized re-start up when applicable may also take advantage of the same throttle valve to relax starting torque requirement which I already mentionned.

 

[shahyar]

Hi Hosein,
My idea is that two control valves of PV-149 and FV-139 works together to control the compressor. They may work on override concept, the normal loop can be through PV-149 (which is more economical control option) and the override loop is handled by FV-139 when there is not enough flow back to the tower.

 

[Jack Nicholson]

Mr Shahryar Thanks. Great explanation ...

 

[Jack Nicholson]

Dear ROTW.
Quoted: /The gas is circulated into a big loop. If you increase the flow, pressure drop increases through the process loop / Unquoted.
When we increase methanol flow, the recycle flow through compressor decreases.
And in constant speed, if flow decrease, discharge pressure will increase.

 

[LittleInch]

We can see the outlet, but where is the inlet of gas??

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[rotw]

hosein elbaghli, no disrespect but I think you know more than letting know.

 

[georgeverghese]

For FV 139 to act as (a) low suction pressure recycle (as mentioned by @shahyar), which is required when there is low net fresh feed appearing at 310-T-1, and (b) compressor min flow recycle.

On closer inspection, this arrangement seems a little odd :
a) the location of FV 139 discharge seems to be wrong; it should be tied in downstream of PV 149, and not upstream, to allow for times when FV139 is acting as min flow recycle (and also when running on low pressure recycle mode).
b)though I'm more used to seeing the excess suction pressure blowoff PCV (in this case PV 135) upstream of the station inlet block valve (in this case UV300/346), I dont see anything wrong with the way it is set up here.

 

[Jack Nicholson]

Dear Mr. ROTW.
Frankly no, I'm really curios about this and I'm learning right here too.
I would not put you out, if I was fluent on this subject.
Really appreciate dear all. I think I get what I want. Really thanks for your coporate.

 

[LittleInch]

Every time I look at this drawing I find different / odd things with it.

1) Is it me or how does 244 ton/hr of gas get heated from minus 93 to + 30, apparently only by 75 tons/hr of the same gas going from 40C to minus 88???
2) It's pretty clear that the re-cycle line has a compressor in order to go from 30 bar to 32 bar, plus pressure losses through 4 HXs. Why the designer thinks it takes up to 30 bar to do this I don't know, but that's what they seem to be aiming for.
3) What's the flow coming into presumably 301-T-1 or does the gas just appear out of thin air?

To do this with a fixed speed compressor and to ensure that the inlet into the compressor is always a certain amount, the control valve PV 149 drops the pressure to some unknown value which is then compressed to 60+ barg. The valve then opens and closes in response to outlet pressure which will vary with flow requirement for the re-cycle loop in general - not clear what sets flow in this system - but flow is massaged by opening the compressor re-cycle line. Keeping the compressor re-cycle line at a reasonable temperature, there are 2 Hxs, one air, one water for some reason - ambient temp issues?? who knows.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[rotw]

Hosein Eghbali,

What is sketch coming from, FEED, concept study phase?
existing machine? Don't you have any other accompanying document? emails? Ok not emails.

Thanks

georgeoverghese,

a) the location of FV 139 discharge seems to be wrong; it should be tied in downstream of PV 149, and not upstream, to allow for times when FV139 is acting as min flow recycle (and also when running on low pressure recycle mode).

Guess if you bring gas back to compressor suction you need to pass through the knock out drum to avoid risk of condensate / water carry over penetrating the machine. Anti-surge/recycle loop is a fast circuit and the gas here is taken downstream after cooler.

 

[georgeverghese]

@rotw, Agreed, the PCV 149 and the return recycle stream from FV139 should both be upstream of the suction scrubber in standard design practice.
Does seem like this recycle gas compressor is part of the column top reflux - compression - refrigeration loop for the natural gas de ethaniser column 310-T-1. Net flow of 160tons/hr of de ethanised gas (mainly methane) flows out to the natural gas reformer at the front end of the methanol plant for syngas production.

 

[georgeverghese]

This suction side PCV149 would be throttling for all part load refrigeration duties, which is a waste of the energy in the feedstream to the compressor. Better operating flexibility woud be obtained with a VSD on this compressor. The PCV149 would then be operating only when required refrigeration flow is below the turndown limit of the VSD.
More OPEX / CAPEX savings can be obtained if we reduced the cold end temp approach on the last HX 310-E-4 from the current 5degC (93 - 88 degC) to say 2-3degC, which should be possible for compact plate type exchangers. This will help to decrease the discharge pressure on the refrigeration compressor.

 

[georgeverghese]

@LittleInch,

Agreed, the enthalphy balance across the 3 exchangers doesnt seem right, given that the Cp value for 2 streams would be about the same, and there is probably no phase change for the stream being cooled. Suspect the error here is on the column top exit temp of -93degC, which could actually be higher for a lower C2 recovery ratio at the column.

The 30bar drop across FV111 would possibly be what is required to get the exit stream to a temp adequate to enable the desired column top temp / create enough reflux and to meet the column spec on C2 content in the column over head vapor exit stream.

Another energy saving option that would reduce the compressor discharge pressure is to enable an isentropic drop with an expander rather than an isenthalpic drop with the JT valve FV111. A mechanically coupled booster recompressor on the suction or discharge of 310-C-1 is also included as a speed brake for the expander typically. This increases mechanical complexity however.