Using 2 regulators in a parallel setup simultaneously

[jari001]

Hi everyone,

I'm trying to confirm my understanding of how to use spring loaded pressure regulators in parallel. My understanding of how to use regulators in parallel simultaneously, to deliver a flow rate that can't be achieved by one regulator alone, is shown in configuration B. My understanding as to why configuration A is not recommended is that having both regulators at the same set point would cause the regulators to interfere with each other. Configuration B would be setup such that the regulator at 30 psig would provide, say 1/3 of the total flow, and the regulator at 28 psig would provide the remaining 2/3s when the 30 psig regulator started to operate further into the droop as the flow demand increased. For configuration B, the two different setpoints would still meet the downstream pressure requirements of the bioreactors.

Edit: The flow demands I am trying to meet break down to typical and peak flow demands. Normally one regulator would be sufficient for the typical load, but a peak load scenario is now introduced where the flow demand is much higher. To resolve this, configuration A proposes two identical regulators working simultaneously to meet the flow demand downstream whether it is the typical or peak load. Configuration B, in my mind, basically dedicates the higher setpoint regulator to the typical load and the lower setpoint regulator to the peak load. In both cases, the droop from the higher flow scenarios would be looked at to ensure it was acceptable.

Do you agree with my assessment or have I missed something?

 

[LittleInch]

You seem to show two self contained flow control valves.

If you instead have two valves and one controller you can send identical position signals to both valves so that they truly act in parallel.

Then option A will work.

Option B looks to close and as the lower set valve opens it will eventually go full open and have difficulty controlling before the second one opens. Not sure this is a good move.

Why not one bigger valve?

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

 

[jari001]

@LittleInch
I made some post updates to clarify my situation, possibly changing your advice for me.

My intuition to favor B comes from the fact that these are spring loaded pressure regulators and therefore there can be differences in how either of the regulators perform relative to each other. For option B, I don't think the higher setpoint regulator would ever close, I think as the flow demand increases and the downstream pressure falls to the lower setpoint, then the other regulator opens and adds another inlet stream at the lower pressure to satisfy the overall demand. I'm trying to understand this particular scenario in particular because configuration A was proposed by my client's PM that got the idea from another engineer that I don't know and I'm trying to see if my reasoning is faulty or not.

To date, my design was based on different client requirements where I did favor one bigger regulator, but this scenario analysis is part of some requirement changes.

 

[LittleInch]

Can you give us some details of the regulators?

Are these like a pressure relief valve or something else?

Is the 28 / 30 psi upstream or downstream.

What's the upstream pressure?

If you can dampen the response of the valvess A night work.



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

 

[Compositepro]

I would start with "A" and if there is any issue then you can change one setpoint. Spring loaded pressure regulators have significant droop (outlet pressure dropping with increase in flow rate). It is possible for oscillations to occur but that depends on a number of factors being just right. It is also possible for a single regulator to oscillate, particularly if it is over-sized.

 

[jari001]

@LittleInch
I don't have particular models chosen, I was approaching it from a more theoretical standpoint. A typical example of regulators I would use are something like this or this. For the purposes of this discussion, the upstream pressure in all cases can be 50 psig and the setpoint (28 psig or 30 psig) is for the downstream environment. These are not pressure relief valves, they are in-line devices that reduce pressure as the upstream gas flows through the device. When the dP between the inlet and outlet is 0 psi, the flow through the valve is nominally 0 scfm/slpm. For my case, the gases that will flow through are air or CO2 (separate piping systems for each gas). I'm not sure I can dampen the response since the spring in the regulator is set and imposes a constant force on the gas flow, but I will talk to a vendor about that possibility.

@Compositepro
Agreed on the oversized regulator creating issues, since it becomes more likely to operate in the lockup region. I was missing that option of being able to adjust configuration A to become configuration B or vice versa.

 

[georgeverghese]

B is the better configuration to avoid regulator destabilising control interaction( since the higher set regulator would already be wide open when the lower set regulator kicks in). But would be better if the SP for the lower set regulator were 25psig or less; 28psig seems too close to 30psig.

 

[LittleInch]

The only thing missing from option B is that the peak load ( which from a re-read of your OP seems to be three times the flow of "typical" load?) would end up at a delivery pressure of max 26-27 psi and more likely 25psi. Given that your pressure drops increase as a square of the flow rate you need to work out what min pressure is required at max flow at your end point and work backwards from there. Of course it then means at low flow the end point pressure will be a lot higher.

The risk of option B is that if the demand fluctuates, the pressure at the regulators could drop for a short while below the set point of the second regulator which then cuts in with a brief burst of gas before shutting off again, Regulators tend not to like doing that.

That 2 psi will need to be within the droop of the higher valve at full flow. Seems quite a small gap to me.

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

 

[jari001]

Hi everyone,

Agreed that the line pressure will be lower during the peak demand vs. typical demand, and I will have to make sure that those extremes are suitable for the downstream equipment. I take the point that my proposed 2 psig gap between the setpoints seems too tight, so I will scrutinize that more.

My current understanding of what the client is trying to do that forces the peak load scenario to occur is a gas sampling activity that requires much more flow than the bioreactors. The sampling activity usually takes around 30 minutes, so I don't think the lower setpoint regulator would be forced to cycle open/close very quickly to meet an ever fluctuating load. The higher setpoint regulator would manage the load variation within the typical load scenario and be sized for that.

 

[georgeverghese]

With pilot operated regulators, droop is less and turndown is better too (relative to regular spring op regulators).

 

[jari001]

Follow up:

I spoke with a couple senior engineers at my company and one person said they agree that configuration B is how they implement dual, parallel regulators and the other person hadn't seen such an application in his time. I also spoke with an engineer from the client side at a different site and that person said he has used configuration A without problem many times. His only stipulation was that the regulators he used were non-relieving.

So like most things I have to decide how to pick between these different bits of guidance. I will also be talking with a vendor soon about this and hopefully they can school me on the pros/cons - perhaps my understanding of how config. A raises stability issues is only true for certain situations.

 

[LittleInch]

That's good feedback.

Each system is different so you need to have your specifics available when talking to others or seeing if your system is similar to yours.

So you need your typical flowrate and your peak flow rate so everyone can see what the multiplier is.

Then you normal pressure required d/s the regulators and what the min pressure d/s the regulators can be at max flow

So e.g. if typical to peak flow is only a multiplier of 2 or less, then dual identically set valves can work. There is a risk of some instability, but often might just need some longer lengths of pipe to reduce pulses affecting each valve.

But if you go to three or more then you really need different sized regulators and set at different pressures.

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