13 replies to this topic

[kaidlut]

Dear forummembers：

 

Our hydrogen plant uses the SMR （Methane steam reforming）process。

Recently，Three temperature（the reforming steam and the nature gas/steam preheater Inlet and outlet temperature）experienced a significant drop once ，We suspect that it is steam carrying liquid from the sour water stripping tower。Liquid water entering the furnace tube can cause tube rupture and catalyst pulverization

 

Here are some more information:

 

1. The opening of the stripping steam control valve is significantly larger than the design.

(design max flow opening:61%, Actual opening: 87%, Equal Percentage, △P is larger than design);

 

1. The steam Flowmeter  overhead of stripping tower display 21ton /h，but the simulation result according to downstream process is 43ton /h。Technip Specifies  22.6ton /h = Maximum HPS into Stripper

 

Here's my question：

1. Whether steam carrying liquid from the sour water stripping tower is common，What are the general reasons？
2. in my case，Whether excessive stripping steam is the main cause？
3. How can I tell that the steam orifice flowmeter is inaccurate？

Edited by kaidlut, 12 September 2024 - 03:25 AM.

[breizh]

Hi,

for point 3, you need inspection of the element. Should be done during turn around!

What about the quality of the steam? compare data, design vs actual. Is the steam overheated versus saturated? 

Note: Orifice is not the best technology.

You need to review all the parameters (design Vs Actual). 

Let us know the gaps found and we may be able to support.

Note 2: You may have internal of the column damaged, take the opportunity of the next turn around to inspect the column. I've experience with such damages on several columns. Sometimes operators increase the steam flowrate too much leading to equipment failure (trays, packing).

Good luck.

Breizh 

[PingPong]

Without a process flow scheme it is not clear where the mentioned sour water stripper overhead enters the SMR.

 

I have never seen an SMR design that receives a feed from a sour water stripper overhead.

 

Normally an SMR has a process water stripper to remove acid (and other dissolved) gases from the condensate, which is recycled as BFW. The acid gases are vented to atmosphere, either directly or via the furnace stack.

Recycling the acid gases back to the reformer makes no sense.

[kaidlut]

This process is from technip, Our plant has been in operation for ten years。

 

The saturated steam from the top of the sour water stripping tower is mixed with superheated steam and then mixed with natural gas. It is heated by the convection coil and enters the reformer tube containing the catalyst for reaction。

Without a process flow scheme it is not clear where the mentioned sour water stripper overhead enters the SMR.

 

I have never seen an SMR design that receives a feed from a sour water stripper overhead.

 

Normally an SMR has a process water stripper to remove acid (and other dissolved) gases from the condensate, which is recycled as BFW. The acid gases are vented to atmosphere, either directly or via the furnace stack.

Recycling the acid gases back to the reformer makes no sense.

[PingPong]

 

This process is from technip, Our plant has been in operation for ten years。

 

The saturated steam from the top of the sour water stripping tower is mixed with superheated steam and then mixed with natural gas. It is heated by the convection coil and enters the reformer tube containing the catalyst for reaction
 

Through the years I have seen Steam Reformer designs from Technip as well as other reputable licensors but never a design that recycles acid gas from the stripper overhead back to the furnace. Without a flow scheme it is not clear.

 

If you suspect that too much stripping steam is used then why not gradually reduce the stripping steam flowrate while frequently sampling and analysing the stripper bottoms to verify that it is still on spec?

[kaidlut]

give me your email ，I'll send you the flow scheme, please.

 

 

This process is from technip, Our plant has been in operation for ten years。

 

The saturated steam from the top of the sour water stripping tower is mixed with superheated steam and then mixed with natural gas. It is heated by the convection coil and enters the reformer tube containing the catalyst for reaction
 

Through the years I have seen Steam Reformer designs from Technip as well as other reputable licensors but never a design that recycles acid gas from the stripper overhead back to the furnace. Without a flow scheme it is not clear.

 

If you suspect that too much stripping steam is used then why not gradually reduce the stripping steam flowrate while frequently sampling and analysing the stripper bottoms to verify that it is still on spec?

 

[latexman]

If you want, you can attach the flow scheme to your post.  Click "More Reply Options" in a new post, or "Use Full Editor" while editing an old post.  The file attaching functions are under the new text window that comes up.

 

Members do this all the time.  No worries.

[Pilesar]

I used to design SMRs for a different engineering company than Technip. PingPong has more recent experience. Feed saturators are to improve plant efficiency. To address your questions:

1) liquid entrainment from a stripper column is possible. Column pressure drop is an indicator. After a certain point, more steam does not usually help strippers -- enough is enough. I agree with PingPong's suggestion to back off the steam. Reduce the steam till you see a quality change, then increase the steam to a reasonable margin above that flowrate point. Compare plant data with historical and expected performance to judge whether the stripper column internals are damaged. Consider a column scan and/or a crawl-through inspection. Column mist eliminators can saturate and send slugs of water overhead in certain conditions.

2) Excessive stripping might cause the problem. You can rate the column performance using computer software. It might suggest where descrepancies lie. If steam flow data does not agree, determine where the preponderance of evidence leads.

3) I assume all steam flow meters are inaccurate until proven correct. Have the instrument tech check calibration. Trash in the line at the orifice or a worn orifice plate or partially plugged pressure taps might give misleading readings. A missing square root sign in the DCS calcs can also mislead.

4) Technip should have a field group which might help your troubleshooting.

[kaidlut]

Here are the process flow diagram, thank you.

 

 

This process is from technip, Our plant has been in operation for ten years。

 

The saturated steam from the top of the sour water stripping tower is mixed with superheated steam and then mixed with natural gas. It is heated by the convection coil and enters the reformer tube containing the catalyst for reaction
 

Through the years I have seen Steam Reformer designs from Technip as well as other reputable licensors but never a design that recycles acid gas from the stripper overhead back to the furnace. Without a flow scheme it is not clear.

 

If you suspect that too much stripping steam is used then why not gradually reduce the stripping steam flowrate while frequently sampling and analysing the stripper bottoms to verify that it is still on spec?

 

 

Attached Files

•  热区 PFD.pdf   1.05MB   16 downloads

[kaidlut]

Sir, I checked the Calculate parameter settings，As shown in the attachment：

 

Measured DP is 26.2649Kpa，but DP—FS （flow full scale）is 25Kpa，Is this the problem？I am not familiar with instrument 。

 

Hi,

for point 3, you need inspection of the element. Should be done during turn around!

What about the quality of the steam? compare data, design vs actual. Is the steam overheated versus saturated? 

Note: Orifice is not the best technology.

You need to review all the parameters (design Vs Actual). 

Let us know the gaps found and we may be able to support.

Note 2: You may have internal of the column damaged, take the opportunity of the next turn around to inspect the column. I've experience with such damages on several columns. Sometimes operators increase the steam flowrate too much leading to equipment failure (trays, packing).

Good luck.

Breizh 

 

Attached Files

•  压差.jpg   325.52KB   0 downloads

[breizh]

Hi,

Check with your instrument tech or engineer. The reading seems to be out of the normal range. 

How often your meters are calibrated?

My understanding is that teh volumetric flow rate is converted to mass flow rate using P,T to calculate the density of the steam, thus the mass =qv*Ro.

Attached document to read. 

As others said, gamma ray scan of the column could reveal damage inside. Always good to have a gamma scan at the start up and be able to compare later on with new scan. This makes easy the investigation.

Good luck

Breizh 

Attached Files

•  ISO-5167-p-1-2003.pdf   602.29KB   1 downloads
•  DIFFERENTIAL PRESSURE FLOW METERS.pptx   1.2MB   2 downloads

[PingPong]

I did not have time to study the PFD until now.

 

My interpretation of the PFD is that:

 

- Process Condensate Stripper C352 has packing with one (or more) spray nozzle(s) above it, and without any demisting device. Spray nozzles are likely to create liquid entrainment, even at normal vapor flowrate. At a much higher than design vapor flow that entrainment could indeed be excessive.

 

- The HPS stripping steam to C352 is controlled by a valve so as to maintain a specific ratio between stripper overhead flow (stream 385) and stripper feed flow (stream 271). Reducing stripping steam flowrate would then require changing this ratio set point.

 

If the stripper overhead steam is not dry but contains significant water entrainment then the stripper overhead flow measurement will be wrong even if there is nothing wrong with the instrument.
Note also that the mentioned 43 t/h (that you calculated using Hysys) is not correct if there is liquid water entrainment as that calculation is based on dry stripper overhead steam.

 

It is strange that there is no flow meter shown in the stripping steam line. Maybe there is one shown on the P&ID and/or DCS screen?

If there actually is one then it should be possible to change the stripping steam control into a ratio with the stripper feed flow (stream 271). Or simply have it on flow control without any ratio with any other stream.

 

If there really is no flow measurement in the stripping steam line then that was a rather penny-wise-pound-foolish decision.
An instrument specialist within your company should however be able to estimate the stripping steam flowrate simply from the % opening of the valve, the valve characteristic, and the inlet temperature, inlet pressure and outlet pressure of the valve.

Edited by PingPong, 15 September 2024 - 08:03 AM.

[kaidlut]

Mr PingPong,

 

Thank you for taking the time out of your busy work to answer my question.

 

1. For the demisting device, i'll check it with our equipment engineer，Perhaps the details were not displayed on the PFD.

 

2. Yes,there is no flow meter  in the stripping steam line.

 

3. As you mentioned, There is a  ratio  control between stripper overhead flow (stream 385) and stripper feed flow (stream 271). the design ratio  [Design Ratio (Steam/Water) = 0.35]. operators adjust stripping steam flow according to this ratio.

 

4.I calculated stripper overhead steam flow by the downstream process parameters，before the problem“

Steam Carrying Liquid” happened, not by the stripper side.

 

5. I have attached a screenshot of DCS，and the SPECIFICATION of the  stripping steam control valve PV3527, we can see the  % opening of the valve(81%) is significantly larger than normal(61%). also, the △P is largeer than design.

 

 

 

 

I did not have time to study the PFD until now.

 

My interpretation of the PFD is that:

 

- Process Condensate Stripper C352 has packing with one (or more) spray nozzle(s) above it, and without any demisting device. Spray nozzles are likely to create liquid entrainment, even at normal vapor flowrate. At a much higher than design vapor flow that entrainment could indeed be excessive.

 

- The HPS stripping steam to C352 is controlled by a valve so as to maintain a specific ratio between stripper overhead flow (stream 385) and stripper feed flow (stream 271). Reducing stripping steam flowrate would then require changing this ratio set point.

 

If the stripper overhead steam is not dry but contains significant water entrainment then the stripper overhead flow measurement will be wrong even if there is nothing wrong with the instrument.
Note also that the mentioned 43 t/h (that you calculated using Hysys) is not correct if there is liquid water entrainment as that calculation is based on dry stripper overhead steam.

 

It is strange that there is no flow meter shown in the stripping steam line. Maybe there is one shown on the P&ID and/or DCS screen?

If there actually is one then it should be possible to change the stripping steam control into a ratio with the stripper feed flow (stream 271). Or simply have it on flow control without any ratio with any other stream.

 

If there really is no flow measurement in the stripping steam line then that was a rather penny-wise-pound-foolish decision.
An instrument specialist within your company should however be able to estimate the stripping steam flowrate simply from the % opening of the valve, the valve characteristic, and the inlet temperature, inlet pressure and outlet pressure of the valve.

 

Attached Files

•  微信截图_20240916102105.png   275.08KB   0 downloads
•  ba6a1f61cf2655cd9e510a7cc676ed3.jpg   275.69KB   0 downloads

Edited by kaidlut, 15 September 2024 - 09:16 PM.

[PingPong]

 

3. As you mentioned, There is a  ratio  control between stripper overhead flow (stream 385) and stripper feed flow (stream 271). the design ratio  [Design Ratio (Steam/Water) = 0.35]. operators adjust stripping steam flow according to this ratio.

 

5. I have attached a screenshot of DCS，and the SPECIFICATION of the  stripping steam control valve PV3527, we can see the  % opening of the valve(81%) is significantly larger than normal(61%). also, the △P is largeer than design.

 

So it seems that the operators still use a too high ratio set point.