Good time of the day,

I want to ask questions about SIFs for butt weld and Tapered Transition. Both of these SIFs include mismatch. The maximum value of SIF in both cases is 1.9. If mismatch =0 then both SIFs take small values. But even a small mismatch value leads to a large increase in SIF. However, in the theory of pipe SIFs, butt weld is already taken into account, which means the standard mismatch value. The mismatch in any case will be present after the installation of pipes.

1. Practically how do you use these SIFs? Those. when should one decide to use additional SIF on butt weld and Tapered Transition?

2. Should I use Tapered Transition for type B fittings? After all, there the thickness of the fittings is greater than that of a straight pipe. And since the assortment may not match, the difference may be noticeable.

Best regards Pavel

For ASME B31.3 SIF's (you can refer to Appendix D, B31.J and FEA).

FEA can be used for obtaining the SIF's of non-standard fittings or physical testing as per B31.J.

The standard fittings SIF is based on physical testing of fittings connected to buttweld pipe. So in theory the buttweld SIF=1 since the basis of testing is a buttwelded pipe.

But, just need to pay attention to strain concentration especially for high temperature systems (you can refer to good article in "Mechanical Engineering News - August 1992" entitled "Piping Failure Caused by Elastic Follow-Up").

Cheers!!!

Thanks for your reply,

1. Do I understand correctly. The Butt weld SIF as a mismatch function should only be considered if there are effects such as creep at elevated temperatures. And don't specifically specify this SIF if creep isn't expected in the pipeline. And it is not necessary to set SIF even if Type B components are used (components with increased thickness) and it is not necessary to set SIF if we have seamless pipes (extruded pipes).

2. I additionally set the Tapered Transition SIF in those cases if I put reducer type B one step higher. In cases where I have heavy valves with an offset center of gravity in my pipe line. And in case of the lateral seismic, there is increased stress in nearby small reducer diameter (since SIF of reducer are essential). However, without taking into account the mismatch SIF for the Tapered Transition is not so big (1.41-1.48).

Best regards Pavel

Hello,

Actually I haven't seen in the Appendix D the SIF for reducer. But I think your doing a good practice of giving enough attention to this component since it's a source of strain concentration (energy of large pipe being absorb by smaller pipe).

How about our fellow stressers, what's your opinion regarding this subject? What is your approach on this?

For high temperature systems at creep range, there is this "CRP" load case in Caesar II, which essentially considers some portion of expansion stress as unrelieved stress and considered in sustained case. But I think this is for EN code and not in B31.3.

This is where sound engineering judgment may come into play.

Cheers!!!

Appendix D in B31.1 includes the SIF details for a reducer component.

Thanks Richard for that information. Not so much involved in Power Piping that's why. It's a good guidance to follow especially for critical systems.


Cheers!!!

Hello Borzki,

In truth, I must do my job in accordance with EN13480-3. I did not specify this at the beginning, since the SIF for the butt weld looks the same in both codes. If I understand correctly in the latest B31.3 revision, appendix D is no longer available. And instead you have to use B31J and there is a SIF for reducers: 3.1 Concentric or Eccentric Reducer Meeting ASME B16.9 In general, with regard to the reducer, here, in my opinion, everything is obvious on a small diameter, increased stresses clearly arise, which must be taken into account in the SIF. Type B reducer has a transition to a cylindrical section SQRT(Dt), so I set the thickness of the reducer at this point to be different from a straight pipe. Otherwise, it is impossible to appreciate the benefits of increased thickness for stress reduction. But I set the SIF in the reducer unambiguously. In this regard, in CAESAR II there is a conflict situation for EN13480, the fact is that there SIFs are included into two tables H1 and H3. The first table does not have a weldolet SIF, and the second does not have a reducer SIF. If you switch CII to the second table, then the reducer will be left without SIF. Therefore, working with EN13480, you have to check the SIF and sometimes set them manually.

As for the butt weld, both codes say to use the SIF for the butt weld for thin thicknesses. Those, if strictly comply with the code, then for thicknesses less than 5 mm, according to EN13480, you must always set SIF=1.8 Which looks a bit conservative. In EN13480 this SIF is not a mismatch function as in B31. But even if you calculate SIF according to B31, it is still difficult to evaluate mismatch, since mismatch will always occur during welding, and even more so for seamless pipes. Therefore, if we strictly follow the code, not taking into account that the butt weld is already incorporated into the theory, then we must apply the SIF for the butt weld. However, we do not. Taking in to account that SIF for butt weld are still present in the codes, I would like to clarify for myself in which cases we are obliged to apply it.

Best regards Pavel

Hello,

That's nice that reducer is include in ASME B31J. For critical systems, this would increase the reliability of the piping system in cases where fatigue or elastic follow up is present. But for general piping systems where there is no severe cyclic loads and temperature is not so high, I think this is not so much of a concern.

Cheers!!!

Pavel,

In EN-13480, the basic equations (in paragraphs 12.3.2 to 12.3.6) take the SRSS of all three moments and apply a single SIF - which is to be taken from Appendix H.1. This matches the B31.1 (pre-2020) approach.

EN-13480 paragraph 12.3.1 permits an alternate approach, where a more detailed determination of the stresses can be performed by separating moments (in-plane, out-of-plane, and torsion). Here the SIFs should be taken from Appendix H.3. This matches the B31.3 approach.

The CAESAR II configuration switch changes which set of stress equations (and SIFs) are used to determine the stresses in the piping system. If the switch is set to use separate moments, then the SIFs from Appendix H.3 are used. If the switch is set to SRSS the moments, then the single SIFs from Appendix H.1 are used.

What you could do is set the switch to use the SRSS moments and the SIFs from Appendix H.1. Take the reported SIF value for the reducer and manually specify this value on the reducer element in the input. Then change the switch back to use separate moments. The setting of the reducer SIF and its origin should be documented in the "title page" of the input for documentation and review purposes.

I hope this helps.

Thanks Richard for that elaborate explanation and also to Pavel for bringing up this topic.


Warm Regards,