Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Compressor Pressure Control


This topic has been archived. This means that you cannot reply to this topic.
11 replies to this topic
Share this topic:
| More

#1

  • guestGuests
  • 0 posts

Posted 05 October 2008 - 06:01 AM

Can anyone tell me the which would be the better control strategy for a simple single stage variable Turbo Compressor package. The Turbo Compressor is variable speed (AND NO IGV's) and we intend to control it's discharge pressure at 86 bar, by varying compressor speed. To do this, we obviously need discharge Pressure Transmitter connected to PID Controller (onboard UCP PLC).

Quite recently, someone suggested to me that using the Suction Pressure is preferable but did not qualify his position why. I cannot see how by using suction pressure to control the speed would work in gaining us a fixed discharge of 86 bars. Am I going crazy or missing something??? What is wrong with using discharge pressure to control speed and hence discharge pressure? I think maybe the only way to control thru put of compressor using suction pressure would be to apply controller output to suction Choke valve (in the absence of IGV's in the compressor

#2 JoeWong

JoeWong

    Gold Member

  • ChE Plus Subscriber
  • 1,223 posts

Posted 05 October 2008 - 05:35 PM

Sorry for a very simple question. What exactly is "Turbo Compressor package" ?
Centrifugal or reciprocating compressor ? Is it gas turbine driven ?
Both compressor type has very different control strategy.

Discharge or suction pressure control subject to the target selection and overall control scheme.
Both have been implemented...

Your target is compressor discharge pressure at a fix pressure, i don't see why and how you can choose you suction pressure as control parameter and how your discharge pressure can be maintained in the event of flow fluctuation (if centrifugal type) ?

#3 djack77494

djack77494

    Gold Member

  • ChE Plus Subscriber
  • 1,282 posts

Posted 07 October 2008 - 02:14 PM

I believe you are refering to a (gas) turbine driven centrifugal compressor package. You will need various control systems including anti-surge controls and possibly load sharing controls if multiple compressors are in use. I'd be very hesitant about using speed control if that's on your mind. Throttling the suction pressure could work. Or spillback gas from the outlet, through a cooler, back to the inlet. Talk to your compressor vendor about what controls he/she would recommend for the situation. They know best.

#4 Zauberberg

Zauberberg

    Gold Member

  • ChE Plus Subscriber
  • 2,718 posts

Posted 24 October 2008 - 03:11 AM


I think the question should be re-phrased. You are asking what is the bast way to control the compressor at fixed discharge pressure of 86bar?

Suction throttling is an old-fashioned way of controlling (mostly) fixed-speed centrifugal compressors which are designed for the lowest molecular weight of suction gas. Everytime when feed gas is heavier than design, suction is throttled in order to increase the polytropic head which has to be developed by the compressor - by droping the pressure at compressor suction.

So the answer on your question is - you cannot control the discharge pressure by controlling suction pressure: the discharge pressure is already fixed (that's the usual case), and back-pressure on the compressor is maintained by the PCV located somewhere downstream of compressor discharge. Since you already have the turbine-driven (variable speed) machine, I cannot see what other efficient and energy-effective method you can use for controlling compressor operation. Am I missing something?

#5

  • guestGuests
  • 0 posts

Posted 27 October 2008 - 09:34 AM

QUOTE (Zauberberg @ Oct 24 2008, 04:11 AM) <{POST_SNAPBACK}>
I think the question should be re-phrased. You are asking what is the bast way to control the compressor at fixed discharge pressure of 86bar?

Suction throttling is an old-fashioned way of controlling (mostly) fixed-speed centrifugal compressors which are designed for the lowest molecular weight of suction gas. Everytime when feed gas is heavier than design, suction is throttled in order to increase the polytropic head which has to be developed by the compressor - by droping the pressure at compressor suction.

So the answer on your question is - you cannot control the discharge pressure by controlling suction pressure: the discharge pressure is already fixed (that's the usual case), and back-pressure on the compressor is maintained by the PCV located somewhere downstream of compressor discharge. Since you already have the turbine-driven (variable speed) machine, I cannot see what other efficient and energy-effective method you can use for controlling compressor operation. Am I missing something?



OK. Thanks for your input Guys. To continue.

In general, what does Suction Pressure Control mean in terms of Compressor Control?

Does it mean "To control Compressor thru-put by varying speed of compressor or does it mean to control compressor thru-put by controlling suction pressure.


Option 1
i.e If controlling Compressor speed then possible scenario is that if suction pressure falls, then for any fixed speed point, then discharge pressure will fall accordingly. To make up desired thru-put, speed increase required to raise discharge pressure. Almost like feedforward control (no feedback from discharge pressure)

CONVERSELY option 2

If controlling Compressor suction pressure then possible scenario is that when suction pressure falls off, control valve in suction of machine is opened more therefore maintaining thru-put by increasing suction pressure

WHICH IS the right assumption for what Suction Pressure control means?

I am leaning to the assumption that varying suction pressure by modulating a control valve is for a fixed speed machine with inlet guide vanes. In my application, I have a variable speed Turbine so I can discount option 2.

Does anyone agree / disagree with this?

Therefore I can stand my ground and implement Discharge Press control by speed control or consider suction press control varying speed to make up for lost suction pressure. I think I have answered my question.

It has to be discharge pressure control????

Any more thoughts guys? Big Thank You to all


#6 Zauberberg

Zauberberg

    Gold Member

  • ChE Plus Subscriber
  • 2,718 posts

Posted 27 October 2008 - 11:21 PM

QUOTE (glennmitch @ Oct 27 2008, 07:34 PM) <{POST_SNAPBACK}>
In general, what does Suction Pressure Control mean in terms of Compressor Control?

Does it mean "To control Compressor thru-put by varying speed of compressor or does it mean to control compressor thru-put by controlling suction pressure.


Neither of two. If discharge pressure is fixed, you use suction throttling in order to accomodate for increase in gas molecular weight. Since the compressor will develop higher differential pressure by pumping heavier gas, suction throttling is the only way to run the fixed-speed compressor at fixed discharge pressure. When suction valve is fully open, you are running out of compressor capacity.

As you have variable speed machine (turbine driven or VFD motor), there's no point of suction throttling because you are able to run the compressor at different speeds and accomodate for changes in gas flow/molecular weight. Compressor speed controller is connected directly to the suction pressure controller, and it changes the compressor speed accordingly to the changes in gas flowrate/properties.

In case of variable-speed machine, what would be the purpose of having suction throttling valve?


#7 JoeWong

JoeWong

    Gold Member

  • ChE Plus Subscriber
  • 1,223 posts

Posted 28 October 2008 - 02:19 AM

Compressor is just a piece of bulky equipment pushing fluid passing through.
Higher the rotational speed, higher the flow...

For a variable speed drive compressor, you control the compressor speed. So you are controlling the throughput of compressor. The speed is set by control signal.

The control signal can come from suction pressure transmitter (if you want to maintain the suction pressure) OR discharge pressure transmitter (if you ant to maintain the discharge pressure) OR...

Above just from compressor control aspect.

Now the question is what is your target ? Is suction or discharge pressure better for your system ? This is hard to say without the detail background of the process, operation, overall control scheme, etc.

If you have discharge pressure control, do you have any control element to maintain suction pressure, Do you happy with suction pressure floating? How much is the suction pressure floating ? how this pressure floating affect your process performance ?.

Similarly if you have suction pressure control, do you have any control element to maintain discharge pressure, Do you happy with discharge pressure floating? How much is the discharge pressure floating ? how this pressure floating affect your process performance ?.

I guess only yourself can answer...



#8 djack77494

djack77494

    Gold Member

  • ChE Plus Subscriber
  • 1,282 posts

Posted 30 October 2008 - 06:57 PM

I find the direction this thread has gone to be very confusing. What do you want to control? What is your independent variable? Initially I was thinking that your discharge pressure was externally maintained. Keep in mind that a centrif compressor is a kinetic machine. It generates head, not pressure. The outlet pressure is related to the head, suction conditions, and gas properties.

#9

  • guestGuests
  • 0 posts

Posted 26 November 2008 - 06:01 AM

Big Thanks for all your replies and help Guys. To finish the subject I will describe the Process. Sorry not responded earlier but been away from internet whilst on site

Suction side of Machine is connected to Feed Gas from Offshore (Through a Suction drum). Sea Line pressure is let down and maintained by a PCV from 88 barg to 55 barg. The Turbine Driven Compressor has no IGV and is variable speed. Discharge of Compressor goes to dehydration package and after that into National Grid. 86 barg is deemed necessary discharge pressure to overcome Pressure drops across Discharge Drum, dehydration Unit, Flow control Valve and Metering Skid. Gas enters national grid which is at approx. 72 barg pressure.

Reading your responses, pretty sure now that discharge pressure is the only way to go. Need 82 barg discharge to enable gas to flow into National grid, so Compressor speed will be controlled to maintain 82 barg discharge. The guy who keeps telling me that suction pressure control is much better obviously not much idea

Thanks again guys

#10 Zauberberg

Zauberberg

    Gold Member

  • ChE Plus Subscriber
  • 2,718 posts

Posted 27 November 2008 - 12:36 AM


Exactly. You don't need to control suction pressure in the system with a variable speed machine, when discharge pressure is controlled by means of SC (speed controller). Or, better to say - when the machine performance is adjusted based on different suction conditions and a fixed discharge pressure.


#11

  • guestGuests
  • 0 posts

Posted 23 December 2009 - 07:07 AM

I think the question should be re-phrased. You are asking what is the bast way to control the compressor at fixed discharge pressure of 86bar?

Suction throttling is an old-fashioned way of controlling (mostly) fixed-speed centrifugal compressors which are designed for the lowest molecular weight of suction gas. Everytime when feed gas is heavier than design, suction is throttled in order to increase the polytropic head which has to be developed by the compressor - by droping the pressure at compressor suction.

So the answer on your question is - you cannot control the discharge pressure by controlling suction pressure: the discharge pressure is already fixed (that's the usual case), and back-pressure on the compressor is maintained by the PCV located somewhere downstream of compressor discharge. Since you already have the turbine-driven (variable speed) machine, I cannot see what other efficient and energy-effective method you can use for controlling compressor operation. Am I missing something?

Suction throttling is not old-fashioned at all, and not just for compressors with fixed speed drivers. Particularly turbo-drivers have only limited range of speed variation. If the operator requires a wider turn-down range than the chosen driver can provide, suction throtling is the most efficient way.
Without suction throttling, the compressor control in the low capacity range (close to the surge line)would be provided by the anti-surge system. The A/S system moderates a bypass valve that spills back to the suction line enough gas to keep the discharging gas from flowing backwards internally in the compressor (surge). The spill-back control is very inefficient because the gas stream being spilled back had to be compressed first, and a large portion of the invested energy is lost as heat.
Although the compressor needs to develop higher pressure ratio when throttled (in order to deliver gas to the required discharge level) the mass flow is significantly lower because gas has lower density at lower pressure. The high efficiency of suction throttling is in the fact that the energy saved by reducinh the mass flow far outweighs the energy wasted by delivering higher pressure ratio.
In the case of a variable speed turbine drive, one would use a split-range control to lower the compressor speed to its permissible minimum and then begin throttling.

#12 Zauberberg

Zauberberg

    Gold Member

  • ChE Plus Subscriber
  • 2,718 posts

Posted 24 December 2009 - 09:50 AM

I agree with what majority of what you're saying - and if you look at our postings the difference is only in how the compressor operation is controlled once when/if we move towards the minimum speed, which is not very likely for majority of applications and if the equipment has been sized properly in the first place.

If suction pressure has to be maintained at a certain set point, and discharge pressure is fixed either by means of a PCV or fluid condensation temperature (e.g. refrigerant compressors), then speed variation is definitely the best way to handle the operation. Throttling the suction causes polytropic head (and speed) requirement to increase even further, and that would call for more ACFM at compressor suction for the same flow of process gas. The conclusion - as I see it - is that we'll get more suction flow either by spillback, or by suction throttling plus spillback, but at a higher pressure ratio when suction throttling is used. Moreover, from control point of view, the machine could never achieve required polytropic head if it is close to its minimum speed - that's why spillback + antisurge is probably the only way to control the compressor in cases of ~fixed polytropic head requirements.




Similar Topics