Dear all,

has someone already experienced exchanger tubes damage due to stress corrosion cracking mechanism which originates and propagates from the inner surfaces of tubes??

It seems that this phenomenon is due to the concomitant presence of 2 factors:

1) RESIDUAL EXPANDING STRESSES ON THE TUBES ;
2) PRESENCE OF AGGRESSIVE ELEMENTS SUCH AS CHLORIDES.

The additional data are:

- we are talking about of a shell & tube heat exchangers with fixed tubesheet and an expansions joint on the shell;

- tube material is SS 304
- shell side material is carbon steel

- tubeside fluid is REFORMED GAS

- tubeside working temperatures are Inlet 125°C / Outlet 125°C

- shell side fluid is PROCESS WATER

- shellside working temperatures are
Inlet 212/197°C
Outlet 138/136°C

- shellside working pressure 1,31 barg

- tubeside working pressure 18,49 barg

Any kind of help and suggestion will be highly appreciated.

many thanks

gengi

One of possible answers may be: transients. Does the heat exchanger experience rapid fluctuations of flow/temperature or thermal shocks?

Second possibility is alternating tension in the tubes. Is the shell provided with an expansion below? Are the tubesheets fixed? Is there any simple way to increase flexibility of shell-tube bundle setup? Do you have the IP calculations of the HX? Is there differential thermal expansion between tubes and shell calculated? Are these calculations based on right assumptions? How many thermal cycles a day does the HX see?

There may be more. Is the corrosion localised at the tube ends, or does it occur in the middle of their length? Are there baffles provided? How dense?
Regards

_________________
Konrad Anikiel

Dear akonrad
thanks for your reply.

SOME MORE INFO....

- the exchanger is horizontal with fixed tubesheet and an expansions joint on the shell
- the chlorides may be in the tubeside
- the tubesheets are in carbon steel with SS weld deposit on the tubeside

the cracks have developed, randomly in the tubes thickness of the tubes part near the shell side face of the tubesheet.
I mean, where the tubes are in contact with the back face (shell side) of the tubesheet.

I have some doubts about the correct work of the flanged & flued expansion joint installed on the shell........

There has been a uncorrect use of the working temperatures ( mean metal temperature different from those considered during the design)???
or....
The expansion joint does not work correctly (wrong design)???
Now I would check the expansion joint calculation....

What do you suggent to do?????

hope this could help your evaluation...

thanks

regards
gengi

Stress corrosion in the point of contact between tube and the tubesheet at its back face is quite often seen, but it occurs usually at the outside surface of the tube, where the contact takes place. Inside corrosion might be seen earlier due to chlorides inside, but this might indicate localised bending of tubes at that point. So maybe the bellows are not flexible enough or the tubes are not supported by baffles (if they are long).
But I would also investigate whether thermal ratcheting is not taking place. This might be due to sudden changes in tempertures/flow rates, when temperature of inner surface of the tube is higher than outer surface. From your data it seems that it is not the case, but the question lies in the transient states, how fast can any change in parameters occur and in which direction?
One more thing: is the corrosion seen at one end of the tubes only, or at both ends?
Regards

_________________
Konrad Anikiel

Dear Konrad,

the corrosion has seen at one end of the tubes only, at the TUBESIDE INLET where the temperatore is higher ( HOT SIDE ).
It also seems that there is pretty temperature difference between the two sides of the INLET TUBESHEET ( between shell side and tubeside-inlet ).

The tubesheet is TEMA Type "N" - welded at both sides to the shell and channel inlet.

rgds

gengi,
This indicates that the problem is caused either by the temperture absolute value or by temperature difference between tubeside and shellside. In the first case, there's about nothing you can do except replacing the tubes with duplex or similar material. In the second case I would try to estimate whether change of local flow in shellside wouldn't help. Maybe adding an impingement plate could lower the difference?

_________________
Konrad Anikiel

Dear Konrad,

thanks a lot for your suggestion and precious help.

now we will see what happen.....I will keep you informed

rgds

gengi