Reinforcing Pad vs Gussets 4

[Packermanfan]

Hi all, the COMPRESS software we use for validating ASME Code designs does not support gussets. This becomes a problem with higher temperatures and pressures, as the nozzles are usually the limiting factor. Forming metal into a "dome" to fit the ends of elliptical heads to use as repad is much harder than adding gussets welded to the head and nozzle.

My question is, how does one validate a design when gussets need to be used to reinforce inlet/outlet nozzles? Thickness of material and repad equations are laid out in the code book, but there is almost no mention of gussets.

 

[SnTMan]

Different horses for different courses. The Code rules for reinforcemnt are to replace area removed for the nozzle opening. Gussets add essentially nothing to this.

I suppose gussets can be used to reinforce for external loads, but I don't know of any recognized treatment of this. Personnally, I would avoid gussets on high temp, high pressure vessels.

Regards,

Mike

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

 

[r6155]

PIP VESV1002 COMPLETE REVISION
Design and Fabrication Specification for Vessels January 2019
ASME Code Section VIII, Divisions 1 and 2

Gussets shall not be used to strengthen, stiffen, or reinforce nozzles,
unless demonstrated by calculations to be suitable for the specified
cyclic life or thermal condition, and the dimensional requirements
(e.g., tolerances) of the device as furnished by the device
manufacturer are considered

Regards

 

[dik]

It's not my area, but I would think gussets would reinforce nozzles. Is there a reason for this exclusion? Are there readily available calculations that show this? either for or against?

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[TGS4]

I’m with r6155 on this one. Gussets are a really bad idea in vessels hotter than 300F, due to the differential thermal expansion between the nozzle and the gusset. In my experience, I would avoid at all costs.

Instead of a repad, use a self-reinforced nozzle.

 

[jt1234]

I have been through old time and 21th century. In the old time, most vessels will be designed at 650F as the allowable stress was set the same up to 650F. Does it mean no re-pad, no gusset. etc. ? That is not practical and not industry common practice for the past 100 years. Our common practice is for design temp 800F and above, avoid using re-pad and if gusset is used, make sure full penetration weld is used, and perform FEA as needed. Other oil company may set 750F. We are just a little bit aggressive for hundreds vessels we have designed. As for why gusset is not a good idea for reinforcement, I will post next from technical point of view because it will be quite length.

 

[jt1234]

Nothing left to do in office in this Friday afternoon, so here we are. You can disagree if you have better technical insights.

Code already addresses this issue, but most people can not reason it from the code.
See Figure UG-37, Figure UG-37.1, and the definition of “F”.
Using the most simple concept “Full Area Replacement” that has been used in early days: how much area removed, put back the same area by re-pad within the boundary per Fig. UG-37.1.
The other simple concept: Fig.-37, when angle is 0, maximum hoop stress PR/t occurs, normal to the areas in Fig UG-37.1 , such that you need Full Area Replacement (F =1) to resist the maximum stress
When angle is 90, the minimum longitudinal stress PR/2t occurs, also normal to the areas in Fig UG-37.1. However, this longitudinal stress is 50% of the hoop stress, so you only need 50% of the Full Area Replacement (F=0.5) to resist the minimum stress.
Any angle between 0 to 90, will have stress between PR/t and PR/2t, also normal to the areas in Fig-37.1. The required area is adjusted by F between 1 and 0.5.

Code is way to conservative by saying using full area replacement (F=1) in all direction unless integral reinforcement is used. I do not agree with Code from stress point of view as explained above that stress will actually reduce from angle 0 to 90, such that the required area can be reduced accordingly. But can’t against it since it is Code, stipulated by human just like law.

If you read so far and understand what I am talking about, you will see using gusset does not make sense and can not qualify full area replacement from angle 0 to 90, unless you install hundreds or infinite gussets to meet area requirements in every cross section from angle 0 to 90. It is purely the stress issue around the opening that make using gusset as reinforcement become impossible.
What you missed is, you only look at angle 0, but you forgot the rest as well as what code says in the definition of "F". Code already covers all, very smart.

 

[r6155]

1) Some specifications use two gusset bars of 5 x 38 mm for nozzles up to NPS 2, fillet welded to the flange.
2) For agitators use compact flange according to ASME VIII Div 1 Figure UW-16.1 (p-2).
Regards

 

[GD2]

Gussets are typically used in small branch lines where Code doesn't call out repad and these lines experiences vibration due to fluid direction and velocity change.
In fact, small branch lines(instrument connections etc) are almost always a problem in terms of vibration. Many Owners even has made this as a standard practice.

GDD
Canada

 

[r6155]

if gusset is used, make sure full penetration weld is used,

Never full penetration weld the gusset plate, use a continuous fillet weld with a smooth contour radius.
For low temperature service reinforcing pads are not allowed.

Regards

 

[0707]

I am with GD2

 

[EngineerEPC]

Thanks for your detailed explanation. I understand the variation of stress from 0 to 90.
However I am having trouble imagining figure UG-37.1. Can you please explain it further.

(I understand that for tangential nozzles, area requirement along the oblong dimension of the opening will be more if the opening is in circumferential direction and there where the factor F will be used)
(If a tangential nozzle is located along the long axis of a vertical vessel, stress along the long axis will be PR/t which will cause more area requirement)


Thanks and regards
EngineerEPC

 

[RadiateurFou]

For low temperature service reinforcing pads are not allowed

Could you elaborate? We use reinforcing pads at low temperature around -200°C

 

[r6155]

@ RadiateurFou

Reinforcing pads shall not be used under conditions where severe corrosion/oxidation is possible or where there is the possibility of severe temperature gradients occurring, in service, across the thickness of the shell.

The gap between the pad and the shell acts as a crack; Brittle fracture may occur, even during hydrostatic pressure testing.
Nozzle with reinforcing pad is not an integral design.

Regards

 

[XL83NL]

I’m not experienced in cryo applications. But the delta for -200 is t that far for from say +230 deg C. The latter, although situations and configurations differ and there is no 1 limit, seems like a reasonable threshold above which repays should not be used anymore.
EN 13445 has hard limits and I think those limits are above 230 deg C.
Of course, codes are no cook books but the fact such limits exist (not only in codes, but also seen em in client standards at the same approximate level) gives the impression repads can be used nader certain safe conditions.

Huub
- You never get what you expect, you only get what you inspect.

 

[RadiateurFou]

@r6155 I understand thanks for the answer.

@XL83NL good remarks, also i should say that our temperature designs start at +65°C so the delta is closer to 300°C. Maybe we should be careful when using repads, i'll try to see if there is any recommandation in the code

 

[XL83NL]

I believe one of the issues with repads is that the pad doesn't thermally expand with the same rate as the shell (since the pad's temperature will have some lack on that of the shell). At a certain point, the differences are so great that the additional stresses are too much. So the issue is with the delta T, either negative or positive.

Huub
- You never get what you expect, you only get what you inspect.

 

[Rob VE]

@XL83NL I think AD2000 Merkblatt has a hard limit on temperature for the use of repads.
EN13445 says you need to avoid it, but in the same sentence they say if you need to use repads, you need to use the same material and need to avoid thermal stress concentrations. What you can interpret as: you need to have decent insulation to minimize thermal difference in the shell/repad.
Additionally they speak of mean wall temperature, so you could argue that this is not design temperature.

in words of the code:9.4.5.2 "In case of high mean wall temperature for the shell (more than 250 °C) or in the presence of severe
temperature gradients through the shell, the use of reinforcing plates shall be avoided; if it is necessary then
the material of the reinforcing plate shall be of the same quality of shell material, and special measures and
warnings shall be taken to avoid thermal stress concentrations."

Edit: We avoid gussets in high temperature or cyclic load cases because gussets are prone to cause cracks at the tip of the gusset.
In div1 you are bound to the code rules for reinforcement. In div2 you could use FEA (part5) to overrule code rules. However I do not expect you will end up with a better or cheaper solution.

 

[Krausen]

Just use a LWN nozzle (self reinforced). Never seen a vessel nozzle gusseted before. As GD2 stated, you typically only see gusset braces on small bore piping branches (under B31.3), and often it's designed & installed poorly.

Nozzle repads at high design temps also have drawbacks with differential thermal expansion.

 

[Packermanfan]

Thanks everyone for the responses, I've learned even more than I asked for with this thread.

From what I understand gussets should not be used to reinforce the nozzle for internal pressures, only for supporting external loads on the flange/nozzle. Even then there should be great care to make sure temp gradients don't cause the gussets to fail prematurely. Same goes for the pads as well, as they are still separate from the shell/head.

So when a nozzle is through the end of an elliptical head the only choice to reinforce the nozzle for a higher internal pressure is to either add a special "dome" reinforcement pad, or increase the head thickness? I've increased the nozzle wall thickness before but I've only gotten results that are a few PSI higher than with a thinner nozzle. If the MAWP of the nozzle is something like 90 psig but the design pressure is 100 psig, it would be nice having options that don't involve buying thicker heads.