Constant Gas Flow Using Pressure Regulator

You can't obtain constant flowrate by maintaining the pressure constant, as you have seen.

What is wanted is oxygen trim control. Modulate the gas flowrate to obtain minimum excess oxygen in the stack. That way, smoke, an indication of too much fuel, will be minimised. If in doubt, consult the furnace manufacturer for more information.

the static pressure remains constant. thats what the regulator do, it maintains the static pressure, but the dynamics pressure increases with the increase in main line pressure, upstream, and hence the gas flow rate increases. I am using a pitot tube type concept to measure the total pressure the the outlet of regulator, and will feed that back to diaphragm of the regulator. that will maintain a constant total pressure. and hence a constant volume flow rate.

we dont want to use oxygen trim its way to expensive.

Then you haven't got constant flow. The starting point for this thread is false.

I NEED CONSTANT FLOW RATE! That was the starting point of this thread. Currently i am facing the problem of variable flow rate, i need to have constant flow rate, by only using a gas pressure regulator

So you said earlier. I did read it. You're not going to get what you want with the installation that you have.

we live in a rational world. Reply with a REASONING!!!

See #58⇓.

Any particular reason you're posting on a thread from 2016?

I apologize, if you feel that way, but i didn't mean anything like that.

You can see the data sheet you have attached yourself which clearly says, that if inlet pressure increases, the flow rate would eventually increase. You are right, i must try my idea, and check how it goes.

thanks

Typically, here in the United States, there are always at least two regulators, one on the down let pressure from the main gas line, and a second near the burner controls (of a boiler for example).

But if the system adjusts things to keep flow constant, the dynamic pressure is constant, so no need to adjust for it.

In any case, at a typical velocity 5 m/s, and gas density 0.8 kg/m³, pressure 0.5*ρ*V2 ~ 10 Pa = 1 mm wg. 2.75 psi is ~ 1900 mm wg, so dynamic pressure likely to be negligible.

Sounds like you have a sophisticated regulator if it can accept control inputs, e.g. 4 - 20 ma. Usually these things are self-actuating, with an adjustment knob.

the flow is around 12000 to 13000 cubic ft per hour in a 2 Inch 40 SCH pipe. Any regulator data sheets says that increase in inlet pressure will cause increase in flow rate, and if the static pressure downstream remains the same, obviously there is an increase in the dynamic pressure, which we dont want to increase.

Do you have a gauge on the downstream pressure? If so, does it remain steady?

Yeah there is a gauge downstream. it remains steady. when the gas burner jets are open the the gauge shows 2.75 psig. and when the the valves are closed the gauge shows 5 psig. there is no intermediate position of the valves.

With natural gas, one wants to not have any venting due to pressure changes, for the simple requirements of equipment safety.

Yes, I suppose it would have to bleed to atmosphere. I haven't dealt with these things for some years. But if there's a pressure relief (I think there is on some of the regulator models in the link on your #26, but I haven't time to reopen it) that would discharge to somewhere appropriate, and bleed-off could go into same line.

"when the gas burner jets are open the the gauge shows 2.75 psig. and when the the valves are closed the gauge shows 5 psig. "

It should be obvious what your problem is related to right there.

Your regulator is set for 5 PSI not 2.75 and you are only getting the 2.75 PSI output pressure due to excessive internal restriction of the regulator due to it being undersized or the gas flow you require.

More than likely when running below the 40 PSI inputs the regulator can not pass enough volume through its internal metering valve and thusly you get the 2.75 PSI output but once the input pressure is high enough it can push more gas though thus raising the output pressure above the 2.75 PSI value eventually reaching someplace near its 5 PSI setpoint.

Good point, I'd noticed that, velocity is very high for gas supply. I meant to ask OP if the pipework is run in same size as the 2" regulator, but forgot. I'd be surprised if the gas inlet flange on the burner is 2", more likely 4". Perhaps he'll come back.

"when the gas burner jets are open the the gauge shows 2.75 psig. and when the the valves are closed the gauge shows 5 psig."

That is normal operation for direct acting pressure regulators. The pressure above the setpoint is called Lock-up, as shown below in a characteristic pressure vs flow performance chart:

Standard adjustment procedure is to adjust the operating pressure at some nominal flow rate, not at no-flow conditions, because the no-flow lock-up pressure is higher than flowing conditions.

WE won't know until YOU try it.

They are your client, you are being g paid to make it work.

Call the manufacturer!

I was seeking a suggestion on the proposed solution i have thought of. Installing a pitot tube to determine the total pressure (static +dynamic) right after the regulator, and feed it back to the regulator diaphragm.

Try it, and report back, but make sure the feedback is negative, so that increasing the signal dynamic pressure will reduce the static pressure at the regulator, or something to that effect. There are more elegant solutions to flow regulation, and it usually requires a direct measurement and a control valve.

Yeah, I will definitely try it. Yes you are right. Adding the dynamic pressure signal will reduce the static pressure at the regulator, but we will re-adjust from the spring to get the required flow rate of the gas, the good part is that if the main line pressure rises from 12 psig to 40 psig the total pressure at the outlet of regulator will remain constant. Hence i am quite positive that we maintain a constant flow with this technique. Alternatively i have used a Pressure Transmitter upstream which give a 4-20 mA control signal, which i have used to control a motorized butterfly valve downstream. The problem with this system is Fine Tuning, primarily.

Why not use a doggone flow transmitter?

If we use flow transmitter, we will have to use an orifice, nozzel, or venturi to attach it to...adding more complexity to the system. And rather than doing this, we would simply buy a mass flow controller and attach it to the gas line, which will have all the things included in it. I am going to try that dynamic pressure feedback thing, and lets hope it works..and if it does, nothing would make me more happy, and I am sure its not a bogus idea to implement.

If you had ever solved this problem, you would know the answer. But you have not, nor have I, and the OP doesn't need either of us to tell him that we don't know. Someone may have knowledge that you and I don't posses, and the fact that your ignorance on the subject compels you to be an asshole over and over and over is so tiresome. Where do you get the stamina?

The supply pressure fluctuation is really not client's fault. Its Natural Gas Supply Company's fault. Sometimes the supply pressure is around 12 psig, and some times its 40 psig. We cannot change.

Installing double regulators, was also an option on the table, but the problem will remain there, because once the main supply pressure increases from 12 psig to say 40 psig, the flow rate will eventually increase.

My point is, what if we using a pitot tube downstream right after the regulator, use that total pressure (static +dynamic) to control the diaphragm of the regulator. Pressure regulators are always designed for static pressure, so why dont we try to utillize the total pressure upstream

You do understand that flow usually reduces the dynamic pressure on a system?

Well, what I have learned in Fluid Dynamics Course at engineering, the flow reduces the static pressure. For example, if we have a tank full of compressed air, 40 psig, which is the static pressure = total pressure with no flow. now if we connect a hose pipe to the tank and open the valve, the tank pressure will remain 40 psig, but in the pipe where the area is less and velocity is high so the static pressure is less tan 40 psig, and the part of total pressure is converted into dynamic pressure. Higher the flow, higher the dynamic pressure at any given point.

Not necessarily true, since the Bernoulli effect can even produce negative pressure.

So you are saying that we must have one regulator on the mainline, that provides 15 psig or 1 bar and then we must have another regulator that reduces 15 psig to 3 psig at the point of use. But say even if we have two regulators, the problem still remains the same, the mainline pressure will increase upto 40 psig, which will increase the dynamic pressure, although the static pressure remains the same after each regulator, but the flow has increased due to increase on dynamics pressure, which is coming from increase in mainline gas pressure rising to 40 psig.

People usually use multiple regulator concept because its costly to install a high pressure regulator for each appliance, what they do is that they install a high pressure regulator on the mainline, and use low pressure regulators for appliances or furnaces installed after that for individual control, But here its just the mainline, and the furnace, so we can have only One Regulator working for us.

So, why is the furnace manufacturer not involved so far?

The regulator is not malfunctioning, if you come across a data sheet of a regulator it will tell you that for a constant downstream pressure, the flow rate increases with the increase in upstream pressure.

I don't know what data sheet you've looked at, but that isn't right. A simple throttle would give that characteristic. The function of a pressure regulator is to maintain constant downstream pressure, irrespective of changes in upstream pressure and downstream resistance, within the regulator design range. (that would be ideal, in practice downstream pressure unlikely to be exactly constant). Also explained in Iris's #51.

As I said in #43, at a given inlet pressure, reducing downstream resistance gives increase in flow, at constant pressure (ideally), up to a point, after which the pressure falls rapidly (the internals being wide open), and the flow reaches a maximum when the outlet is open to atmosphere. If that is repeated at a higher inlet pressure (same set pressure) the pressure fall-off occurs at a higher flow. Maybe that's what you're thinking of. But it's irrelevant if the regulator has been correctly selected for the proposed duty.

I realise that doesn't solve your problem. From the link in Iris's #51, depending which orifice is fitted it seems possible the regulator is borderline sized, but that still wouldn't explain why the flow has increased if, as you say, the downstream pressure has not increased. The pressure/flow characteristic of the burner doesn't change, unless there's something you haven't told us. Are you sure your pressure gauge is accurate? If in doubt I would rig up a water manometer, easy to do and can't be argued with!

As you said earlier, you could use a control valve, flow meter and controller (with PID to eliminate droop) but that would be more expense, and isn't usually necessary for boiler supply.

Let us know how you get on.

I'm suspecting the being undersized since he says in post 5 that when the valve to the burner is shut off the output pressure goes up to 5 PSI.

Also given the high volume gas flow through the line if the gauge is connected at a 90 degree tap onto the line (like any typical pipefitter would do) it's not likely to be all that accurate either being the gauge port into the main line could very well be experiencing high degree of venturi effects that are giving it a false reading well below what the actual pressure is inside the main line.

Output pressure could go up yet the increased flow past the gauge port and a venturi effect in play could counteract the gauge reading from going up so even though there was a considerable volume increase (as his burner flame shows to be happening) the gauge itself could very well show no change as he is saying.

If you do the numbers 12 - 13,000 CFH through a 2" line is a working velocity of ~150 - 170 feet per second which in my books would be capable of a substantial venturi effects given any hole in the wall of the pipe!

That's what I have suspicions of.

...and if one sticks a thumb across the downstream port of the <...regulator...>, thereby blocking it, then it doesn't matter what the upstream pressure is, the downstream pressure remains constant, and the flowrate drops to zero.

Which model Elster regulator?

its Elster American Meter Co 1803 with internal static connection. Size DN50

Have you considered that the black smoke may be from the quality or mix ratios of their fuel supply changing?

Natural Gas suppliers add air to their fuel delivery to change its effective thermal content per unit of volume in order to adjust for different amounts of other combustible or noncombustible gases that may be present in the supply and in many cases some do it to simply cheat the customer into having to buy more gas than they need as well.

Whats in utility supply gas.

I would have less confidence in natural gas supplies in Pakistan, but the OP never stated this was natural gas, or unnatural gas. It could be something really, really bad, of varying BTU content, or even of widely varying ratios being saturated gases as methane, and unsaturated gases such as ethene, acetylene, etc. It might even be refinery off-gas that could include BTEX as vapors. Who knows what they really are burning?

I understand completely but he doesn't 'so it has to be the regulators fault' being that's adjustable on his end and appears to correct the problem.

He said natural gas in #10. What that means in Pakistan I've no idea!

dont worry, nothing like that. Fixed BTU per Cubic Ft. and its 95% Methane

By far the easiest and cheapest solution is to go to a stack gas oxygen sensor and throttle control the fuel supply to maintain oxygen concentration in the exhaust gas.

A second, bulletproof method would be to regulate into a buffer tank at 12.5 psi. Take the buffer tank output and feed it to the existing regulator, which becomes a static setting throttle valve.

The problem you describe is one I have seen in pressure reliefs and regulators in that the existing valve has hysteresis due to internal component friction and a feature called accumulation due to dynamic gas flow and internal surface area configuration. This is built into the regulator and the only solution is to go to a different valve with a different internal design and geometry. Sensing line geometry and location has been mentioned by others already. It is not reasonable to expect a single stage, garden variety regulator to cope well with a 300% swing in inlet pressure.

Why?

The LPG regulator on my house, and most BBQ grills, puts out 10" - 11"s WC with a input range of 2 - 315 PSI as single stage units.

I would be willing to bet real money (about $0.25 in my pocket), that your source pressure is really steady. Especially since it runs off old tires.

The house furnace uses regular propane from a tank like everyone else's. It's the backup heating system.

The tires go in the shop boiler for fuel for the hot water heating system that does the primary heating.

MFC- mass flow controller. this seems like a simple fix. Perhaps you should look this up.

What you want has an interesting parallel with anaesthetic machines. This are supplied by a regulator at 4 bar, although the sorts of pressure a human lung tolerates are measured in cm of H2O. The needle valve at the bottom of the flowmeter acts as a very high resistance, so as soon as there is flow, the pressure at the outlet is dropped to safe levels. There is a second effect, though. There is a high pressure applied to the top of a high resistance. If you do this with volts and high ohms, varying any further resistance has little effect on the flow, i.e. it becomes a constant-current source. If you do the same thing with a high pressure regulator and a length of thin tubing to act as a restrictor, you will end up with a constant flow.

Looking at this again I notice you say set at at 2.75 psig. But what is the actual downstream pressure when the inlet pressure varies?

The regulator curve, plotting flow horizontal and outlet pressure vertical, at a given inlet pressure should show constant pressure (horizontal line) as the downstream resistance falls (on a test setup) up to a point, after which the pressure starts to fall, going to near zero if the outlet is opened wide. At a higher inlet pressure it's similar, initial pressure as before, but the pressure drop-off occurs at a higher flow. And higher still for higher pressures.

If you're not sure about downstream pressure, it's possible that with 12 psig inlet you're on the falling part of the curve, but higher inlet pressure takes you back to the horizontal bit, at higher flow. Specially if the regulator size is borderline for the flow required.

Thank you for a serious feed back. The downstream pressure remains steady at 2.75 psig. The PROBLEM is that the INLET PRESSURE DOES NOT REMAINS CONSTANT. We need a constant flow at 12 psig inlet pressure and 40 psig inlet pressure, and whats in between. I hope you get it. My concept was to utilize the TOTAL PRESSURE (dynamic +static) downstream and feed it back to the regulator.

Elster 1803, 2" (DN50) regulator

1) Be aware that any mechanical pressure gauge is subject to damage when over pressured. Is the gauge responding in a linear fashion (the pointer is not stuck at some fixed position due to having been overpressured?) at a reasonable accuracy?

2) A pressure regulator maintains sensed downstream pressure and attempts to hold it constant, regardless of upstream pressure, within its design operating limits. It does that by regulating the flow through the orifice. It is a regulating valve.

It does so in a proportional-only fashion, meaning that there is some droop (drop off in pressure regulation) as the flow increases. Droop is always in a negative direction (lower maintained pressure), never in a positive direction.

3) Flow rate is determined by the downstream demand (how many devices consume volume). If the demand is constant, the flow at a constant pressure is constant. If the demand varies, the regulator adjusts flow to maintains the outlet pressure as closely to its setpoint as it can.

The implication is that a pressure regulator can maintain a constant pressure (minus droop) but it does not maintain a constant flow rate, it is not a differential relay (a constant flow regulator). This is physics.

4) Regulators need to be sized for the expected flow rate and required pressure. There are a variety of orifices and spring ranges used to accomplish matching the required flow rate and pressure.

The manual at this link
http://www.elster-americanmeter.com/assets/products/products_elster_files/SB8540.pdf

shows tables with the variety of combined orifices and springs.

The 1803 DN50 (2 inch) with a spring for 2-3 psi outlet pressure sizing table (page 9, SB 8540.3) is shown below:

If the valve orifice is 7/8" (depending on holder), or 1" or 1 1/4", then the 45 psi supply pressure exceeds the operating limit as shown by the solid black line. A 2nd upstream 'knockdown' regulator with sufficient flow capacity would be needed.

5) If the pressure gauge indicates a steady pressure, the regulator is doing its job, maintaining down stream pressure. It does that by regulating the flow through the orifice valve.

6) Given the conditions above, either
- the regulator is not functioning correctly. It is not holding a constant downstream pressure with varying upstream pressures due to
* incorrect sizing
* internal malfunction
- or the regulator is working properly but something is varying downstream of the regulator to create a periodic gas rich/combustion air starvation condition.

7a) How do you know this regulator is sized correctly for the flow rate at the pressure required?

7b) Is the combustion air adequate? Is the combustion air being choked?

7c) The internals determine the pressure regulation ability. Have you checked the ID tag on the unit to confirm the correct spring and orifice for this application (assuming you know the required flow rate for the burners)?

Springs and orifices are replaceable. Could the internals have been replaced and not identified properly? (used regulator?)

http://www.kinggageengineering.com/Portals/0/Products/Manuals/Elster%20American%201800-2000%20IOM.pdf

The IOM's (Installation Operating Manual) parts list shows that the springs are color coded.

Are you thinking of letting the forum know if and how the problem has been solved?