Shell and Tube Heat Exchanger, Tube Failure, Water in Gas 1

[KBO_BC]

Dear all,
I have a Shell and Tube heat exchanger. Process gas on Tube 0.5barg, cooling water 5barg on shell side. Temperature of Gas inlet 170°C outlet 35°C.

How can we avoid that water entering the following Process equipment (compressor) in case of a Tube failure?

I can not reduce the water pressure due to back pressure of the water reservoir.

My only solution so fare is to install a separator or Siphon.

Has anybody experience with such a case?
Thank you for your help
Kind regards
Kathrin

 

You have not described the process gas. Why are you concerned about possible failure - perhaps corrosion?

In the event of a tube leak you can't prevent water entering the tubes if it is at higher pressure, so the approach must be preventing occurrence of a leak. You accomplish this through design (primarily selection of materials), process control, and adequate non-destructive examinations through the service life. If a tube fails it is vital to understand how and why (and you cannot do that by crowdsourcing, as many folks try to do online).



"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[EdStainless]

After you do everything that IM suggests ...
You also install a detector on the outlet end you tell you the humidity of the gas.
The issue will be that you will detect moisture, shut the process down, and when they inspect they they will say that they didn't see any leaks.
The trick is when moisture is detected to inject a fluorescent dye into the water, and then after a short run time shut it down and inspect with a black light.
Often you cannot find these leaks with water when cold.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[KBO_BC]

Thank you @ironic metallurgist.
The concern is that we create a water hammer in the compressor.
Of course we design the Cooler as best as possible, this issue was coming up from a HAZOP.
In cooler with gas pressure higher than water pressure the scenario of a tube failure has to be taken into account and a rupture Disc must be installed.
So as consequence during HAZOP we also have to check what happen if a tube failure occurs in the cooler with low gas pressure.

Thank you @EdStainless this is helpful.
I will look into the detector of gas humidity

 

[GD2]

KBO BC,

There is no way you can prevent tube failures of water gas mixing if the tube gets a hole or there is leak through the tube and tubesheet joint. One way to prevent the tube and tubesheet leakage path is to flare the tube ends.

I am sure the compressor have condensate trap. But that is for removing the condensate from cooling off the gas.

Another way to avoid water entering the gas compressor is to install a sump at the lowest point of the inlet piping, with a drain valve installed at the bottom of the sump actuated by a level switch. Interlock the drain valve to the compressor motor so that the compressor can't start if the water level is high or the drain valve is open. Include a position valve in the drain valve so that the control room operator can see whether it is closed or open.

GDD
Canada

 

[r6155]

To reduce risks, you can do the following:
1) Use seamless tubes
2) Pneumatic test after forming the U-tubes and before installing them.
3) Check the thickness reduction in the bending sector of the U-tubes.
4) Tube expansion and seal welding.

Regards

 

[EdStainless]

Sorry r6155,close
1)SMLS tubes assure you of nothing other than them being seamless. I have seen them fail an air underwater test also. Requiring AUW on tubes is worth the cost.
2) Ubends get a hydro after bending and any HT. Air test of bends in not possible in most shops because of how wide they are.
3) YES, In many bend sets it not unusual for the inner 4-5 rows to be 1ga heavier than the rest because of this.
4) YES, Monitoring the control of expansion is critical, both location and amount. AWR is not a measure of the amount of expansion unless they have precise measurements of the actual hole size (per hole) and actual tube wall thickness (per end, and accurate post expansion ID sizes. No one does that, the good shops do it on a small sample and then control rolling by measuring the torque.
If you are welding then they must do a light contact only exception, weld, and then do the hard expansion.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[LittleInch]

I would go with the gerneal flow here - you cant stop it happening.

As you say water hammer then I'm assuming this is a PD piston type compressor, especially as you're at 0.5 barg.

Going from 175 to 35 C you must be pretty confident about the dewpoint though - that's a big temp drop so any water vapour in the gas will start to condense.

If you're concerned about it then you will need to introduce some sort of water trap / knock out drum before the compressor inlet. No other option.

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

 

[Brunel777]

Seconding litteInch. Since the tubeside pressure is more and not knowing the process gas if it is a critical issue due to tube failure of the cooling water ( tubeside) leaking across the only way to abate this risk is to have a knock out vessel with demister pad downstream so that any water that does go into the Process gas side is knocked out and any droplets are removed by the demister pad.