Design of S&T Heat Exchanger with COG away from Shell side 1

[Vessel_Eng]

Hi,

I have a situation where girth flanges & A cover of AEU Tema type Exchanger are very heavy (Tube Side Design pressure is very high, 140 barg).
As a result, Exchanger COG (Center of Gravity) is away from Shell side (COG is in the Channel side area).
Normally, Saddles are located on Shell side & COG is somewhere between Saddles means COG is on Shell side, which is not the case here.
We arrived at following 2 Possible solutions:
1. Design Exchanger for 3 Saddles (2 on shell side, while one saddle on Channel Girth Flange adjacent to A Cover, to support Channel Side weight)
2. Locate one saddle on Channel Cylinder & one Saddle on Shell Cylinder, such that COG is between two Saddles.

Solution 1 has got its own challenges such as design calculations for 3 saddle case, uneven distribution of loads etc.
Solution 2 will result in Operating/maintenance/handling issues, since it is a removable bundle design.

Request advice from Forum members about above 2 solutions.
Especially, if you could advise any other solution, that would be highly appreciated.

Regards

 

[r6155]

Never use 3 saddles. No problem with 2 saddles on shell side. Do calculation with uneven distribution of loads.

Regards

 

[Vessel_Eng]

Thanks r6155 for your feedback, much appreciated.

By calculations with uneven distribution of loads, Do you mean to have both Saddles on Shell side while Exchanger COG is away from the 2 Saddles (i.e. Exchanger COG is in Channel side & hence not between 2 Saddles on shell side)?
Have you seen this kind of configuration?
Is it practical from handling/operation/maintenance perspective?
Can Zick Calculations be carried out for this case (i.e. COG is away from the 2 Saddles, not between 2 Saddled)?

Regards

 

[r6155]

See TEMA. You need to define the load on each saddle. COG is not the problem.

Regards

 

[EdStainless]

This situation isn't that uncommon. You may need to worry about how the head will be removed and handled. And piping stresses can be a big issue (make sure that they are not taking load from the HX).

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[SnTMan]

Vessel_Eng, I have run into this at time or two. Not a lot you can do about it.

Go with two saddles. Make sure eveybody knows one is subject to uplift and that the uplift is reversed at maintenance. Forget Zick.

Regards,

Mike

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

 

[Vessel_Eng]

SnTMan/Mike, I guess you are implying to the solution of having 2 Saddles on Shell side. If Zick is not applicable, then what alternate calculation method we should consider in this case?

Noted on your point of one saddle design for uplift & reverse uplift during maintenance. I guess we'd have to ask Vendor to provide a temporary support/skid base for maintenance case.

But do you see any problem during normal operation in this case?
With COG away from Saddles (COG in Channel side) & Channel side Nozzles are piped with 16" Piping from Top & Bottom side (Tube side is 2-pass), will Exchanger, in your opinion, be stable/self standing during normal operation?
How did you resolve these kind of situations in your experience?

Regards,

 

[SnTMan]

Vessel_Eng, I'm not aware of any "canned" alternative to Zick meeting this case. Do you really think the shell is so thin that loads imposed by the saddles are important? If so, make the shell a little thicker or go FEA.

Re maintance, all you have to do is bolt the exchanger down. Bolts will be in tension normally but zero load on reversal of reaction. More of a problem for the foundation designers really.

If the exchanger is bolted down, I see no need to consider that meaningful loads are imposed on the piping.

The imporant thing IMO is to make sure the owners are aware of the pecularities of the design. Communicated thru the outline drawing, showing CG and perhaps reactions dry & flooded, and with bundle removed. Perhaps also channel removed, maybe even cover removed. That way they can plan the foundations or the steel for all cases.

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

 

[r6155]

Vessel_Eng, can you send us general dimensions and weights?
I don't believe the uplift of support.

Regards

 

[SnTMan]

You can always counterweight it

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

 

[Vessel_Eng]

Mike, r6155, I'm attaching conceptual sketches for 2 Solutions (Note: Solution 1 with 3 Saddles has been dropped).Exchanger material is CS.

Which Solution do you suggest in attachment, 2 or 3?

Would look forward to your valuable feedback.

Regards,

 

[r6155]

Vessel_Eng.
1)Why not type BEU ?

2)Can you send us weight of channel and weight of shell side without bundle?

Regards

 

[SnTMan]

Given the tube length I'm surprised the CG is as shown, I'd guessed a much shorter tube length. Still, I'd go with plan 3). The back support will need to be bolted down.

For plan 2) you will need some means, a skyhook, a third saddle or something, to allow bundle removal.

If a third saddle is placed on the shell it could be designed to be clear of the foundation during normal operation, and shimmed to it for maintenance. Avoids the problems associated with more than two saddles.

RE the counterweight idea: Could the shell be a flat head rather than a formed head? It'd be much heavier I guarantee


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

 

[r6155]

Vessel_Eng, see the difference
A-Type front header
This type of header is easy to repair and replace. It also gives access to the tubes for cleaning or repair without having to disturb the pipe work. It does however have two seals (one between the tube sheet and header and the other between the header and the end plate). This increases the risk of leakage and the cost of the header over a B-Type Front Header.

B-Type front header
This is the cheapest type of front header. It also is more suitable than the A-Type Front Header for high pressure duties because the header has only one seal. A disadvantage is that to gain access to the tubes requires disturbance to the pipe work in order to remove the header.

Regards

 

[Vessel_Eng]

Mike, Thanks for your feedback.

r6155, We had initially proposed BEU, But Client insisted for AEU.
Channel Side Weight = 38 Ton
Shell Side Weight (without Bundle weight) = 15 Ton
Removable Bundle Weight = 8 Ton

Regards

 

[SnTMan]

Vessel_Eng, might be useful to discuss w/ the client. No use engineering arrangements that won't work for them.

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

 

[r6155]

Vessel_Eng
1) As an engineer you must understand what the customer needs and why.
2) Send the budget as the client wants and your alternative with the BEU type.

Regards

 

[r6155]

I did a simple calculation with 2 saddles support in shell side and no uplift occur.

Regards.

 

[Vessel_Eng]

Client wants A type, due to ease in maintenance activities (avoid disturbance to attached piping). Nevertheless, we'll take up this issue again with Client along with supporting data.

r6155, possible to share the uplift calculations that you carried out?

Regards,

 

[r6155]

Vessel_Eng. Try to do a minimum effort with a simple calculation.
if you can't do this simple calculation then you can't calculate this equipment.

Regards