reaction time for stand by gas compressor

[DerekLJ]

This question is really a series of smaller questions, all realating to a system of 2- 100% gas compressors.

1. What is the possibility of having 2- 100% gas compressor, and when one goes down, the other reacts immediately.

2. What operating conditions would the stand-by need to have to react in this form, or the quickest possible...

3. What type of control system would i need to manage this- (the immediate reaction)

4. What type of lag time can i expect from the loss of power of compressor 1- to the complete functioning of compressor 2.

5. Regarding question 4, what is the shortest possible lag time and what do i need to do to achieve it.

 

[gepman]

To give the most accurate answer you should give the compressor type and size. Suction and discharge pressures would help. Are you trying to maintain suction or discharge pressure?

 

[davidbeach]

Also, what does "reacts immediately" mean?

(I don't have any idea, but immediately can mean different time frames to different people and that may mean very different solutions.)

 

[DerekLJ]

very good points.

I forgot to put that it is screw type natural gas compressor with capacity of 4800 lbs/hr. It has 507 psi suction pressure and a 940 psi discharge @ 80F

Regarding my deffinition of immediately, i would say, 1 ms. I know that 1ms is completly unrealistic, so i'm looking for a solution that would put me as close as possible to that number while being practical.

I guess in other words, the time isn't the MOST important factor. What IS the most important factor, is that there is no pressure drop in the gas between the shut-down of the first compressor and the start-up of the second compressor.

Is this type of system common or not?

 

[gepman]

I am not an expert in natural gas compression, most of my experience has been with refrigeration compressors but they are similar and many times the same brand names (Howden, Mycom, etc.).

Nothing can react immediately and for compressors of that size you will still have a rotating mass to contend with. What I would do is have both units operating at 50% capacity and then if one fails open the slide valve on the remaining unit as fast as possible (which even then probably won't be one millisecond). Any kind of feedback control system always controls to an error so your controlled variable will have to change in order for the feedback system to make the change. Therefore you will need a feedforward control system with feedback trim in order to make your transition as smooth as possible. I have utilized feedforward control systems on refrigeration systems where I have had to maintain plus or minus .5 deg. F with 100% instantaneous load changes.

Unfortunately at 50% slide valve position the compressors will not be very efficient. You could run these on VFD's at 50% speed and be close to 100% slide valve position efficiency but then you will not be able to accelerate from 50% speed to 100% speed in one millisecond.

Another possibility would be to have one running and loaded and the other running and unloaded but there will be some issues that will come up running a screw compressor with no gas flow (I have never done this). Again you will not be efficient since there is no work being done.

You could also investigate putting accumulators (receivers) on the suction or discharge side of the system (you never said which side is your controlled variable) to miniminize any pressure transients in your system during switchover.

 

[DerekLJ]

Sorry, i want to control the discharge pressure

 

[DerekLJ]

Thank you gepman...
For the sake of learning, are there any other opinions?

maybe any experts in natural gas compression?