Compressor vent gas injected to lower pressure network

[CostasV]

Hi,

We have a compressor station where the process vent gas is 1000 normal m3 for each stop. This gas quantity is now released into the atmosphere. The total number of venting actions is about 20-60 per year.
If we inject this quantity into a lower pressure network (city distribution network) that passes near the compressor station, we could reduce the methane emissions.
The typical pressure of the process vent gas is 40-45 bar, and the typical city distribution network is 15 bar. The city distribution network pipe diameter is 4 in, and can easily accept a flow of 1000 Nm3/h at 15 bar.

I have googled and I have asked some of my colleagues, but I haven't found any example of a similar facility.
This facility would include pressure regulators, and possibly a valve controlled by the city distribution operator that would block the gas in case they do maintenance to the network. And a custody flow meter, if agreed upon.

Does anyone know of any similar facility?
Is there any fundamental reason that this facility is not feasible, or safe? I think not, but maybe I miss something.
This solution seems to be a low-cost, high-reliability solution. Any comments?

Thanks!

 

[LittleInch]

I'm assuming what you mean is that when the compressor station stops, you blow down the locked in gas inside the compressor?

The reason no one does this is basically the cost involved (regulators, pressure reliefs, meters (really?) etc, far out weigh the value of the gas released. I would not describe it as "low cost" myself. Even then you are only saving 2/3 of the gas as once you get to 15 bar what happens? You still need to blowdown your compressors to <7bar in 15 minutes so how do you get rid of the rest? - you then need to stop sending gas to the local network and send the rest up the vent stack.

So your flowrate that the local system needs to accept is closer to 4-5,000 nm3/hr, not 1000.

You need to put in a quite high level of reliability and safety to avoid over pressuring your local system.

Now environmentally you might want to do it, but I think there are far better things to spend your money on.

Are you now seeing a few of the other issues?

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

 

[1503-44]

1000m3 about 30,000 cu ft. Even today's price its doubtful you will save the cost of your yearly salary. Your boss is going to get upset.

A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[MortenA]

@1503-44, since he is speaking in Nm3 its probably in Europe where co2 equivalent emissions matters - so he's boss might be upset if he dosnt do anything...

But i agree with Littleinch, it will result in some safety measures that will probably be quite complex.

Best regards, Morten

--- Best regards, Morten Andersen

 

[LittleInch]

Also what happens when your city system is at low flow - say 2am on a hot summer night.... Can your city system suddenly accept 4,000 nm3/hr? without warning.

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

 

[1503-44]

Maybe he's in UAE, or India, or... like 187 countries.
Add the carbon tax then.
Still not saving his yearly salary.

Generally city gates buy their gas deliveries on contract. Who says they want to even buy it. You will have yo renegotiate the sales ontract as well. What will be their offer for gas they don't necessarily want?

Giving them a high instantaneous flow rate may exceed their meter flow capacity, if not their pipe pressure. You will probably need a storage tank to enable a slower release over longer time.

If your heart is set on this, why not just build that huge blow down tank next to your compressor station, inside may violate your design code,[pre][/pre] and then send it out from the tank through the regular connection, released at the same delivery pressure and flow rate as always. Nobody will know or care where it actually came from.



A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[CostasV]

Hi,

thanks for your replies.

LittleInch, correct, only 2/3 of gas is "saved". The rest 1/3 may be blown-down.

So your flowrate that the local system needs to accept is closer to 4-5,000 nm3/hr, not 1000.

The time needed to inject the gas into the city distribution network is not 15 minutes, it is 1 hour. So the flow rate is closely to 1000 Nm3/h.
If the compressor needs to be blown down for safety reason (and not just to get rid of the remaining gas inside the compressor) then, as you correctly say, the pressure must drop to <7 bar in less than 15 min. In this case the normal blow-down valve will open.

1000m3 about 30,000 cu ft. Even today's price its doubtful you will save the cost of your yearly salary. Your boss is going to get upset.

My boss asked me to investigate how to minimise the vented gas.

Also what happens when your city system is at low flow - say 2am on a hot summer night.... Can your city system suddenly accept 4,000 nm3/hr? without warning.

City system has a total geometric volume of more than 2000 m3 (steel pipe diameter of 10", 6" and 4"). Consumption flow rate does not drop below 1000 Nm3/h, while maximum has reached 150,000 Nm3/h.


I admit that I am little worried that no-one else has seen this kind of solution. If this solution (gas injected to distribution network) is rejected, an alternative solution is to recompress the gas into the inlet header. Recompression is more expensive as a project, less fast (it takes several hours to drop the pressure down to 10 bar) in operation, with higher operation and maintenance cost.

Thanks again!

 

[LittleInch]

OK. It can be done alright, but you will need flow and pressure control which is not too hard. The biggest issue is getting paid. Given that you are currently letting it go to vent then if I was the utility I would say well give it to me for free or max 10% of the value.

But with the required safety system's, design, supply and installation you're probably looking at $100 to 200,000.

Now no one has done this before because it was too expensive and no one cared about the odd few thousand cubic feet of gas going up the stack.

Now things have changed.

So go do the outline design, cost it up, see what the utility say and then let us know what happened. Maybe everyone will do this in the future and you can say you were the first?

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

 

[1503-44]

"If the compressor needs to be blown down for safety reason (and not just to get rid of the remaining gas inside the compressor) then, as you correctly say, the pressure must drop to <7 bar in less than 15 min. In this case the normal blow-down valve will open."

So you are only talking about restart gas. ESD will always go to vent.

"The time needed to inject the gas into the city distribution network is not 15 minutes, it is 1 hour. So the flow rate is closely to 1000 Nm3/h."

Where will this gas be stored (at 7bar ??? Stored at what pressure) during that 1 hr you are slowly sending it to the city at 1000m3/h?

There is no reason to think that it cannot be done. You just have to give them gas as close to contract provisions as possible. You apparently do that already. See how this gas could fit into the usual delivery system.



A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[LittleInch]

Or look more closely at the volume you are releasing and why.

20-60 times a year is up to once per week. Why is is being vented / released? Could you instead isolate more gas to reduce the volume?

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

 

[CostasV]

But with the required safety system's, design, supply and installation you're probably looking at $100 to 200,000.

A first cost estimation is 350 k euro. One significant cost is the 3.5 km (between compressor station and distribution network) pipeline of 4".

"If the compressor needs to be blown down for safety reason (and not just to get rid of the remaining gas inside the compressor) then, as you correctly say, the pressure must drop to <7 bar in less than 15 min. In this case the normal blow-down valve will open."

So you are only talking about restart gas. ESD will always go to vent.
"The time needed to inject the gas into the city distribution network is not 15 minutes, it is 1 hour. So the flow rate is closely to 1000 Nm3/h."

Where will this gas be stored (at 7bar ??? Stored at what pressure) during that 1 hr you are slowly sending it to the city at 1000m3/h?

I am not sure what you mean as restart gas. ESD, correct, will always go to vent.

The requirement "must drop to 7 bar in 15 minutes" is the design requirement for our vent system.
When gas goes to vent stack, compressor loop piping (including the compressor) pressure decreases from 44 bar to 7 bar in 15 minutes, and then pressure continues to drop from 7 bar to atmospheric for the next minutes.

If gas goes to city distribution network, compressor loop piping pressure will drop from 44 to 15 bar in 1 hour, and then the rest of gas will go to vent stack.
One thought was to send this last quantity of gas (from 15 bar to 4 bar) to the city 4 bar network but the 4 bar network geometric volume is too small to receive this quantity.

20-60 times a year is up to once per week. Why is is being vented / released? Could you instead isolate more gas to reduce the volume?

It is vented, according the design of compressor manufacturer. One other solution could be to supply continuously seal gas to the seal system, so that the gas inside the compressor doesn't need to be vented. But that would also produce methane emissions continuously.
What do you mean "could you instead isolate more gas to reduce the volume"?

 

[1503-44]

I use 15.55C as standard, or normal temp and 14.696 psi (1.013 bar) std pressure when calculating following volumes. Your std p&T may be different.

"If gas goes to city distribution network, compressor loop piping pressure will drop from 44 to 15 bar in 1 hour, and then the rest of gas will go to vent stack. One thought was to send this last quantity of gas (from 15 bar to 4 bar) to the city 4 bar network but the 4 bar network geometric volume is too small to receive this quantity."

So, 1000 (we say) "standard" std m3 gas is contained in the compressor loop piping at 44 barg.
The volume of pipe is 23.03 m3

Case1 - Flow to vent.
Pressure drops from 44 to 7barg in 15 minutes, (841 std m3 is removed, 159 std m3 remains in pipe) after which the pressure drops from 7 barg to atmospheric. (136 std me is removed, 23 std m3 remains in the pipe.)

Note that 841 std m3 removed in 15m is an equivalent hourly flow rate of 3364 std m3/h.

Case 2 - Flow to City
"The pressure drops from 44 to 15 barg in 1h (659 std m3 is removed, 341 std m3 remains in pipe) , then the rest of gas will go to vent stack."
What is the rest of the gas? Is it 341m3 - 23m3 = 318 std m3

Case 3 - Thought
"Send this last quantity of gas (from 15 to 4barg) to the city 4barg network".
At 15 barg there is 341 std m3 of gas in the pipe.
At 4 barg there is 91 std m3 of gas in the pipe.
250 std m3 must be removed to reach 4 barg pressure.

In your first post you said,
"the typical city distribution network is 15 bar. The city distribution network pipe diameter is 4 in, and can easily accept a flow of 1000 Nm3/h at 15 bar."

Now you talk of 4 barg. What's up with that?
Is the city system operating at 15barg, or 4 barg?






A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[CostasV]

1503-44

I use 0 degC and 1.013 bar as normal conditions. This makes our calculations be different by 5,6% which is ok for me.

Case 1 is as you described.

Case 2 is also as you described.
The rest of gas is, as you said, 318 std m3.

Case 3.
Sorry, I should have mentioned that there are two gas networks. The one is a steel pipe 16 bar network, operating between 11 (at the far end locations with maximum flow) and 16 bar. The other is HDPE pipe 4 bar network. The HDPE network is supplied from the 16 bar network, by district stations.

 

[1503-44]

OK.

Case 2: You will send 659 std m3 to the city during 1 hr, at a delivery pressure of 15 barg. When your pressure drops lower than the city line, you can no longer deliver to the city, so you open the vent and blow down the last 318m3.

Do you have the 15min requirement to reach 15 barg in this case? That would mean delivering to the city at a maximum flow rate of 2636 m3/h. If that is too high a rate, then you will have to store that 659 m3 in a tank at a pressure high enough to drive your stored gas into the 15barg network at whatever their acceptable flow rate is. Once you can no longer flow into the 15barg city network, because your stored gas pressure equals the same 15 barg city pressure, you could conceivably switch and continue to deliver into the city 4 bar network, again until the storage tank reaches that 4barg equalisation point. When you reach 4 barg, conceivably, the rest of the stored gas could be saved in the storage tank until the next time you do this.

With proper distribution pressure regulator and a pressure safety relief installed downstream of the regulator in each of the lines connecting into each city pipeline, you will have the beginnings of a typical "city gate" delivery point. You'd probably just have to add a pipeline connection inlet block valve, check valve, sales meter, backpressure regulator for the meter, final outlet block valve, oderant system, and do the tie in with the telemetry and control scheme.

Sounds like a lot of work for minimal result, but lots of stuff is like that now.

A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[CostasV]

1503-44

Case 2: You will send 659 std m3 to the city during 1 hr, at a delivery pressure of 15 barg. When your pressure drops lower than the city line, you can no longer deliver to the city, so you open the vent and blow down the last 318m3.
That is correct.

Do you have the 15min requirement to reach 15 barg in this case?
No, the 15 min requirement is only when venting is done for safety. The usual venting is done because of the operational instructions from the compressor manufacturer.

About the odorant system, I hadn't thought of it. Is it really necessary, for such quantity? I guess the city distribution steel pipe is saturated with odorant, and that the steel pipe could release some the absorbed odorant back to the gas that comes from the compressor. But may be I am wrong.

About the sizing of the pressure regulator, I find it difficult because the regulator inlet pressure will vary from 44 to 16 barg. So, I am thinking of having a flow control valve which will be regulated to keep the rate of compressor loop pressure drop constant (at about 0,75 bar/min , equal to 0,75*23*60 = 1035 Nm3/h).

 

[1503-44]

Yes use a control valve. I wrote CV first, then thinking it was a low pressure, I changed it to a regulator.

You should be able to find a CV with a 44 to 16barg pressure range. I would configure it as a discharge pressure control valve. Then you can just set it to hold whatever discharge pressure you want, 16barg as its downstream pressure, and it will open or close as needed to hold that discharge pressure. Try to size it for that 1035Nm3/h flow rate such that at a 44barg inlet pressure it is around 20% open discharging at 15 barg and not more than 80% open when it's inlet pressure is in the 20 to 16 barg zone. Do not let anybody convince you that it should be configured as a flow control valve.

I think I forgot to mention that you will need a check valve near the meter to prevent any backflow and metering it more than once. It will also keep oderant and whatever else happens to be in the city gas from entering your piping.

You should probably plan on injecting the oderant. In any case I believe somebody has to ensure that the downstream mix is safe.

A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[1503-44]

These may help...

https://pgjonline.com/magazine/2009...undamentals-of-gas-pipeline-metering-stations

That may not be the best design.
Flow is from left to right.

https://epcmholdings.com/gas-metering-stations/

http://www.icggroup.com/Pages/Gas-Pressure-Reducing-Stations.aspx

http://www.hep-co.com/city-gates-stations-natural-gas/

http://www.en-fabinc.com/city-gate-stations.html

A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[1503-44]

What I struggle with is how come a storage tank and a small compressor to reinject this gas into your own pipeline, or selling it to some CNG dealer is apparently not being considered. Why is that? Its similar to dealing with condensates in a gathering system, seperating liquids and gas to meter separately and reinjecting the liquids back into the gathering line. Much the same process, except its liquids coming from a well rather than gas from compressor blowdown. Is the tank too expensive?

Maybe a pipe built slug catcher type "pressure vessel" could work.
You'd need 62.5 m3 volume at 16 barg, 62.5m of 1100mm diameter pipe.

http://staging.weldfit.ulcomm.com/products-and-services/slug-catchers

http://www.cngstoragetanks.com

A black swan to a turkey is a white swan to the butcher ... and to Boeing.

 

[LittleInch]

Ok, by the time you finish this with time costs, design, materials, construction etc you're looking at 500,000 euros once you have a 4.5kn pipeline thrown into it.

What I meant by volume was look at the volume which is being blown down between isolation valves and think, can I reduce the volume by say moving an isolation valve closer to the compressor??

Why not buy a pressure vessel or even say a l few lengths of 36" pipe and then compress the gas into that when you need to blow down then feed it back into the system using the same compressor in reverse when you start up again?


But if you want to go ahead, then a control valve is just that - it opens and closes to create or reduce a pressure difference across it when there is flow. How you control it is up to you. Flow, pressure, temperature or all of those things - you only need one valve. just run the control elements through a low selector block and your valve then controls on flowrate, pressure, temperature , you name it depending on which one is driving the valve more closed than any others. you do need a decent PLC to run it though and not purely mechanical or pilot driven.

I've done this many times and it works a treat. There is no such thing as a "flow control valve". That's just lazy terminology and thinking.

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

 

[1503-44]

62.5m of 40" as above.

Only discharge pressure control is needed at 15 or 4 bars, depending on which pipeline is selected. As above. For example, putting it on too high of a flow control setting could set off the downstream relief valve.

iMO there are flow control valves, pressure control valves, temperature control valves, motion control valves, speed control valves, composition control valves, and many others, the valves all possibly being exact duplicates, differing only in what signal activates them.



A black swan to a turkey is a white swan to the butcher ... and to Boeing.