20 replies to this topic

[buck2233]

 CP plant.pdf   631.9KB   122 downloads25MM/D natural gas processing plant.  Not sure of exact design but it has inlet gas of around 1100 btu.  We have had problems getting the demethanizer tower to function normal.  It is overpressuring.  We have tried many things and cannot find what is causing this.  When the expander is brought online the tower I think freezes off in a matter of minutes.  If we run only on JT we don't pressure up on the tower.  We have changed our mole sieve in the Dehy beds and are still having the issue.  I suspect the dehydration beds have leaking valves.  Could this be the problem?  The regen pressure is running higher than the inlet pressure and the regen gas is residue gas so it is dry.  Could the rich inlet gas be leaking by even with the regen being higher pressure?  The regen flow is about 2MM/D and the inlet is 25MM/D.  We have tried warming the tower for many hours and when we come online it does good for a few minutes but still overpressures and we have to open the JT and cut the expander back.  We started having this issues after changing the center section of the expander out.  The expander is a Roto flow.   We have changed the mole sieve out and it is new also.  We have to regen and heat for much longer than we should to get a dew point of -112 to -120.  This is the best we can get on the new beds.  As stated above we plan to change all dehy valves and see if this helps. 

 

Can anyone offer any suggestions on what else to look at and rule out.  I think we are getting moisture in the Demethanizer tower and this is freezing off.  How is the moisture getting in?

 

Thanks for any help  I will try and get some drawings posted also.

Edited by John Cambern, 11 February 2013 - 11:45 AM.

[thorium90]

Yes, it would help to have a simple flow diagram with temperatures and pressures otherwise there wont be much replies...

[benabed]

John cambern

You should also provide actual and design data of your dehydration section such as dry gas dew point.

[narendrasony]

Buck2233:

 

Do you mean to say that this problem was not there prior to changing the center section of the expander out?  This should not have any effect in my opinion.  Please provide more details?

 

A detailed PFD would help to under stand dryer regeration line up - is there a separate Regen Gas compressor.  Why are you using regen pressure higher than inlet pressure?   whether Regen gas is getting recirculated limits its regeneration capacity.

 

Regards

Narendra

[Aranga]

We have a system little similar  as yours (I guess).  This is an Ethylene recovery plant from Off gas.. What I think is that this is a problem with the expander.  If you have moisture passing out from Dehydration beds, you might have pressure control probems in the Upstream system of the Demethaniser.  What is the JT here?   I assume there is a Bypass valve around the expander.

 

Need more details.

[buck2233]

Yes there is a bypass on the JT valve.  If the plant inlet pressure is higher than design the JT will open so the expander doesnt go down on overspeed.

[buck2233]

Yes we started having this problem after the change - but I don't think this is an issue. 

[buck2233]

I have provided some drawings in the first post for viewing to help with this.

 

Thanks

[thorium90]

Wow, I didnt quite expect your P&ID to be as sparse as it is. Perhaps you could offer a pfd with temperature and pressures too?

I cant find the JT valve that you say is the expander bypass. How is the speed controlled? By IGV, by flow to the expander? Frankly, there is no indication of anything on the expander. I suppose this drawing is just a draft?

Edited by thorium90, 12 February 2013 - 12:02 AM.

[Aranga]

For ensuring moisture is not carried over, You can take a sample at Downstream of Dehydration beds for analysing Dewpoint. (Since you dont have inline Dewpoint analysers). I am bit confident that, this is something issue with Expander and the compressor associated with it. But not sure what is the problem.

[buck2233]

We do have the online dewpoint analyser and we can only get -112 to -120.  We have pulled a tube downstream at the dust filter and have been getting 3 pounds of water on the tube.  This tells me our beds are not doing the job and we are freezing up when the expander is online.  This also makes me think the process valves on the beds must be leaking by allowing moisture in and it carries into the process.

 

 

For ensuring moisture is not carried over, You can take a sample at Downstream of Dehydration beds for analysing Dewpoint. (Since you dont have inline Dewpoint analysers).  I am bit confident that, this is something issue with Expander and the compressor associated with it.  But not sure what is the problem.

[narendrasony]

Knowing operating conditions would have given a better idea of the problem.

 

What is your regen gas temperature, it should be 250-260 DegC for 4-A type mol sieve. (480-500 DegF), are you able to achieve vendor specified  temperature  at dehydrator top (~450-460 DegF for 4-A) at the end of cycle. What is the vendor presecribed regeneration cycle time and actual cycle time?

 

You must also check dehydrator feed gas for any possible water /  liquid caarry over. Any liquid will detorirate the performance of molecular sieve and will definately lead to pre-mature moisture breakthrough. Refer to following link:

http://www.hydrocarb...ntaminants.html

 

What was the change carried out with the expander, can you provide some details? Where exactly is the pressure rising up in the system ? Is there methanol injection facility to alleviate hydrate foramtion problem ?

 

Probably, you can reduce the  throughput for 1-2 cycles and see the effect.

 

Valve passing should not be a major problem in my opinion, other forum members can throw some light. During regeneration passing is not possible,since regen gas pressure is higher than process gas pressure. It may occur during stand-by period at lower pressures and this should not be an issue.

 

I would have preferred, 3 X 50% configuration instead of 2 X 100%.

 

Regards

Narendra

[buck2233]

We have been reaching a temperature of 475 F on our beds and currently are running 20 hours to get it.  This is much higher than manufactured recommended time.  This is to get -110 to -120.  They recommended like 10 hours top.  We do have methanol injection at the cold gas seperator prior to expander.  We had to change the center section becuase they had been having moisture issues becuase the glycol system upstream wasnt working properly and the expander trashed from thrust issues.  We have since fixed the glycol issues and think we are doing very good removing moisture prior to the beds.  The pressure hits the Demeth tower very quick.  It goes up to 250 psig in a couple of minutes.  This is happening when our inlet volumes are very high 25MMCFD.

 

When our volumes are down in the 16 MMCFD we have not seen this issue.  One thing we are wondering is it possible that the side reboiler is flooding the chimney tray and the liquids cant flow down the tower fast enough thus flooding the entire tower.  We also are wondering if at higher volume is the mole sieve not doing the job and we are getting to much moisture to the tower.  When the tower pressure spikes it goes on JT and we shut the expander off.  We see the pressure spike go away as fast as it showed up.  The other thing is the pressure is legit becuase the compressor side of the expander starves for residue gas and the residue units go down on low suction pressure.  We run the tower around 180 psig normal.  I was at another one of our facilities and witnessed a scenario like this and we changed mole sieve and all process valves and our problem went away.  We know we had a few valves leaking by but I cant confirm for sure what fixed it changing valves or changing mole sieve.  We have  a DP gauge installed on the demister pad at the top of the tower and one across the entire tower.  We have seen high DP on the demister but not acrosse the entire tower.  We are trying to install a DP gauge so we can see if we are flooding the tower.  Currently the high level indicator at the top of the tower is not functional and will require complete shutdown to fix.  Thanks for any help folks

 

Knowing operating conditions would have given a better idea of the problem.

[buck2233]

One more thing when we were having issues with the expander we lost seal gas and lost all of our lube oil into the process.  We are questioning could the synthetic lube oil for the expander have got into the tower and messed up the packing in the tower thus the liquids cant flow to the bottom fast enough when we are at higher rates.  Also would this oil plug off the demister pad in the top section of the tower.  I may try and post more pics.  We scoped the tower once and looked at the demister.  I think it looked fine.

[Bobby Strain]

If you take 20 hours to regen vs 10 hours design, maybe your regen flow is too low. Check it out, then gradually increase rates from a comfort zone and observe column DP.

 

Bobby

[narendrasony]

As Bobby Strain has pointed out, you may be running on low regen gas flow.

Another possibility is you are over-doing the regeneration part. The temperature curve becomes flatter and flatter with time. You may be acheiving 450 DegF in say first 8-10 Hrs and then 480 DegF in next 5-6 Hrs. Is -110 to  -120 DegC dew point really required ? Usually vendors prescribe 8-12 Hrs of totally regeneration cycle including.

 

You say, 25 MMSCFD is very high. What is the design capacity ?

 

Flooding in demethanizer doesn't seem to be the problem as indicated by your dP gauge across the column , Are both dP indications avaialble at DCS ?

 

Regarding the mol-sieve replacement in other plant, check if  vendor prescribed life was over or not and were there any upsets in past upstream of those beds damaging and shortening bed life. Passing valves should not be a concern and you can live with it.

 

 

Regards

Narendra

Edited by narendrasony, 13 February 2013 - 11:52 PM.

[buck2233]

Orginal design was for 20 MMSCFD.  We only have local DP gauges this is not on the DCS.  Thanks

As Bobby Strain has pointed out, you may be running on low regen gas flow.

 

Another possibility is you are over-doing the regeneration part. The temperature curve becomes flatter and flatter with time.  You may be acheiving 450 DegF in say first 8-10 Hrs and then 480 DegF in next 5-6 Hrs. Is -110 to  -120 DegC dew point really required ?  Usually vendors prescribe 8-12 Hrs of totally regeneration cycle including.

[narendrasony]

Are you able to achieve 20 MMSCFD with expander? May be there is not enough margin left in mol sieve. As you increase feed gas, you should be able to reduce regeneration cycle time (e.g. reduce standby time), if you can not do it , you can not increase capacity. You need to reduce regeneration time.

Other possible options are: increase regen gas temperature by say 10-20 DegC if possible. Increasing demethanizer operating pressure will also help, it will increase column top temperature and hence relax dew point requirement. I've seen these columns being operated @ 220 Psig and one column @ 290 Psig also.
What are the operating conditions of dryer feed gas? Is it possible to reduce moisture content in feed gas ?
Having these differential pressure readings at DCS will be helpful for troubleshooting .
Regards
Narendra

Edited by narendrasony, 16 February 2013 - 03:00 AM.

[buck2233]

We cannot run at 20MMSCFD it will pressure up.  We can and are changing time in the beds and regen times.  We also use Residue gas for regen and it is very dry no moisture present.  We have not tried raising the pressure on the tower overall to reduce the dew point issues.

[buck2233]

Some current operating data for review.

 

Plant Inlet                    24.9

Tower Top DP                  8

Tower Column DP            1

Tower Pressure SP       210

Top Temp                     -98

Expander RPM            33000

BTU                             1050

Dew Point                    -111

CO2                             0.66

JT                                  17%

Side Reboiler In           -36

Side Reboiler Out         25

 

If we close the JT and ramp up the expander we would see our pressure spike again.

[narendrasony]

Hope you are moving ahead to resolve your problem.

Find attached an excel sheet for calculating dehydrator loading (Wt. of moisture removed / Wt. of Mol. sieve charge).

You may compare it with the design loading value provided by supplier.

Just a 2-3 DegC reduction in feed gas (to Mol. sieve) temperature, reduces dehydrator loading significantly. Don't know how far is it feasible in your system.

 

Best luck

Narendra

 

 

 Dehydrator loading Calculation.xlsx   13.37KB   52 downloads