PWHT of Entire heat exchanger-NEN 3

[deepu78]

For NEN type Heat exchanger as per client specification Shell side PWHT is Mandatory-Due to service requirements.
Shell SA516-GA70 Thk 38mm, I.5 meter dia,10 meter long
Tubes SA-179 Dia 19.05, 2.11mm thk,9150 mm long(approx 3000 tubes)
Tube sheets(fixed): SA765 Gr.II

Which is the preferred PWHT option in the above case? and also why?

A.PWHT of the entire heat exchanger( with tubes assembled in tube sheet)after fabrication.
B.PWHT of shell welds completed first.Then after all the fabrication is completed, local PWHT at the circumferential seam closing welds .
C.IS there any other options?

Is there a industry practice which is recommended.Is PWHT of the entire vessel after fabrication recommended?

Thx,
DV

 

[r6155]

One more data: weld detail in channel side.

Regards

 

[EdStainless]

I have seen all flavors of this.
I hate full HX PWHT but some plants do it.
Local PWHT of closure is common.
More common is making the closure in an alloy combination that does not require PWHT.
Then you can fabricate, weld, and PWHT all of the parts and then simply slide it together and weld it closed.

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

 

[r6155]

1) RT or UT 100% before and after PWHT.
2) PWHT the entire HX without expanding the tubes (tubes with free thermal elongation during PWHT).
3) Expannd the tubes in the tubesheet
4) Hydrostatic pressure test

Regards

 

[deepu78]

Attached the weld detail in Channel /shell side & TS detail.
TTS joint is expanded and strength welded: so the TTS weld and expansion done after after PWHT if entire vessel is PWHT

IS there any risk with closure local PWHT?

DV

 

[r6155]

More info: material of channel, thickness, .....

Regards

 

[Christine74]

If you have a sufficient number of baffles to guide the installation of the tubes all the way to the opposite tubesheet you should be able to do it this way:

1) Weld the tubesheets, nozzles, supports, nameplate bracket, etc. to the shell with the tube bundle skeleton (baffles/tie rods/spacers) installed, but do not install the tubes. Pay close attention to the alignment of the tubesheets and baffles.
2) PWHT shell assembly.
3) Install tubes, strength weld, and expand (no PWHT required).
4) Weld the channels to the tubesheets (no PWHT required).

But no risk to using local PWHT of the closure weld if you decide to you that route.

-Christine

 

[deepu78]

Thanks All.

 

[r6155]

Always assemble the complete tube bundle and then push it into the shell.
But if you install the baffles, braces, ..etc. without tubes . after that in the shell and PWHT, how do you know if any deformation/distortion occurs ? It may be impossible to insert the tubes.
Example:What to do if one tube can not insert?
"No welding after PWHT". Only orbital weld in tube to tubesheet with a special pattern secuence.

Regards

 

[Christine74]

You can temporarily install a few pieces of 3/4" diameter bar stock through the bundle around the periphery of the tube field during the PWHT in order to maintain alignment.


-Christine

 

[0707]

your tubes, being strait tubes don´t need to be PWHT.
PWHT OF the welds of tubesheets to shell can be done separately in a HEATER, after this last PWHT you can weld or expand your tubes and finish the work with all required final ndt.

 

[EdStainless]

Either PWHT the shell with no internals or with all of them.
I am not a fan of seeing entire HX PWHT.
I presume that you have checked how it will change the properties of all of the materials in the HX?

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

 

[r6155]

@ Christine74
“You can temporarily install a few ….............”
“temporally”??- How desinstall if tubesheets are welded?

You think in theory, but in practice (in shop manufacturing) the cylindrical shell is not perfect and round. The distance between the holes is not the same. The holes are not the same. The tubes are not perfectly straight. Everything has tolerance.

Please tell us, what to do if a tube cannot be inserted?

 

[Christine74]

r6155 you extract the rods by pulling them out of the shell. It's not complicated. What I said above is perfectly clear to anyone with any knowledge of heat exchangers. Please stop pretending that you are some sort of expert here when at most you've worked as a low-level inspector decades ago.

Your point about the roundness of the cylindrical shell is of course totally irrelevant to the question at hand. And as I explained above that the option of PWHTing the shell without the tubes installed was conditional based on the baffle spacing. There are multiple other viable options.


-Christine

 

[r6155]

Again. Christine74
Please tell us, what to do if a tube cannot be inserted?
Why not answer?
Why do you respond with offense and aggression?
Why do you respond to confuse?

Regards

 

[SnTMan]

I'd very much agree with Ed's last post. If furnace heat treated with tubesheets attached but less tubes I'd be concerned about scaling of the tubeholes. Yeah, I know they mask threads with some compound, but still. Undesirable.

I think the way I have seen this done most often is to PHWT the shell, less tubesheets, tubes and bundle. Local when assembly is complete.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[deepu78]

Tube holes tolerance is for special close fit (nominal dia 19.25 +-.05)
Scales(if scales are formed due to PWHT) can be detrimental to the tube insertion after PWHT.
DV

 

[r6155]

Yes
The proposal of SnTman is more comfortable for orbital welding of TTS and expand when channel is not installed. But when welding the channel the preheat and welding temperature may damage the joint of TTS, basically those close to the weld.

Regards

 

[Christine74]

Another potential complicating factor: A 10 m long fixed-tubesheet heat exchanger would typically have a bellows expansion joint installed in the shell unless you're operating at a very small MTD and there's no possibility of steam-out or loss of flow on one side or the other. I've never seen a bellows expansion joint get PWHTed.


-Christine

 

[SnTMan]

Yeah, I would not do that

The problem with sloppy work is that the supply FAR EXCEEDS the demand