The reducer thickness we model normally as the average of line thicknesses on eithr side of reducer. There are some companies modeling reducer as - OD= average of either side pipe ods and thickness as thickness of the higher bore. Is it OK? please clarify.
regards,
sunil

As of Version 4.50, you can model the reducer directly. This element has a diameter and thickness specification for each end. Why not use your real data, from the actual reducer?

Dear Richard,
I do not think I got the answer. In versions prior to 4.5 if someone models the reducer as described, do you think he is right?
Regards,
Sunil

hi Sunil

Prior to 4.50 we tended to model our reducers as a mixture of the two methods recommended in the Applications Guide. We would model the length of the reducer, with an SIF (typically 2), at each end. However we tend to use the larger bore diameter and wall thickness.

We generally have a large number of different sizes of reducers and defining each with the average wall thickness and diameter meant when we were looking at the graphic plots showing wall thickness/diameter there were lots of non-standard wall thickness and diameter values, which appeared confusing as we add the graphic plots into our stress reports.

Sunil,

I doubt that the specific modeling of the reducer would make that much of a difference to your results. The system would have to be very sensitive to the flexibility of the reducer.

Agreed your sweating the small stuff while the big stuff may be getting away... spend more time on boundary conditions and SIFS....

A couple years ago, there was a very nice database floating around that was chock full of good data including dimensions, thickness and SIFs for reducers. But with the new feature in v.4.50, I haven't touched it in many months.

Speaking of SIFs and reducers....

Itchy mentioned she typically used an SIF of 2.0 on reducers. ASME B31.1 is the only code I know that implements SIFs on reducers. Does the new reducer modeler in v.4.50 include SIFs for other codes, say B31.3?

Sunil,

I was somewhat confused by your reponse to Mr. Ay's first reply, demanding correct or not correct for a modeling technique where you gave no information whatsoever as to the situation that it is being used.

The correct technique is that that will give you accurate analysis results for the system that you have modeled.

I agree with the others, it's noramlly not a big deal how you model the reducer, because it normally won't effect the accuracy of your results in most systems.

Here's where it does make a differance:
Where a the small end of the reducer or swage terminates at a flange, valve or other fitting, using the average wall and OD for the entire length of the fitting won't give you accurate stresses on the small end where the moments are usaully the highest. In this case you need to model a small piece of pipe at the small end matching the properties of the small end of the reducer.

Regards,

That database was one I created (thanks)....

I also agree with Mr. Monroes comment (currently I disagree with the information on modeling the reducer in the CAESAR help file, the SIF should be applied to the smaller NPS x Wall schedule end)...

Now my personal opinion (rant)... I feel that B31.1 is "wrong" for requiring that a SIF be applied to Concentric reducers.

Why?

First..
In order to calculate the SIF you have to know the tangent lengths of the reducer to calculate the angle. THIS IS NOT A STANDARD DIMENSION AND VARIES WIDELY FROM MANUFACTURER TO MANUFACTURER, (B31.1 could impose a "standard dimension for these lengths but they remain silent on the matter, if the damn thing is important enough to them to require a SIF they should clarify what they want for the tangent length, or get rid of the formula and impose the overly conservative value of 2.0)

Second...
There are no recorded instances to my knowledge of B31.3 reducers failing!!!! B31.3 does not require the use of a SIF.

Third....
Why does B31.1 only require the SIF on Concentric reducers surely the geometry of an eccentric reducer is more extreme!

Fourth...
A review of the original WRC paper by Dr. Rodabaugh discusses in length all these concerns and more yet here we all are years later yapping about something that is in the real world of very little significance!

John,

E-mail me with the location in the documentation of what you disagree with and I'll take a look.

Rich

I completely agree with John and Kevins comments. I have come across more than one calc that would have failed by a good margin had correct small end properties been used.

It would be an interesting challenge to someone to find and prove a more accurate method of modelling real reducers that is measurably better than a single step change with 50% big end properties, 50% small end properties. I would guess an FE calc or field test of the reducer and a bit of pipe each end could be used. However, there are the small questions of what shape is used for the transition as John highlights above. Secondly the direction of load applied to Ecc Reds might be a factor. Hmmm.

Thanks to all of you for the replies. It all started with one system on which I am working presently. The problem is that when you handle engineers / designers who are doing layout and have no concept of stress , they are not ready to co-operate with stress engineer. Rather stress engineer is viewed as one who is the last person to think about schedules. I had a case where no routing change was possible( not even a small run of pipe could be added near the reducer) and static equivalent PSV analysis was showing failure at reducer( 103% in OCC). The force of PSv pop up was 4 tons on 12"x6" reducer. One of my colegue told me that one consultancy uses higher thkness of reducers to overcome these problems !
Thanks again.
Sunil

Sunil,

To actually install heavier wall fittings and pipe is a legitimate way to overcome your stress problems. The pipe, fitting or flange attached to the small end of your reducer also needs to have this same heavier wall or bore.

From your original post, I understood that your colleague suggested you model your reducer as a heavier wall than you actually had (on the small end). Now that you have disclosed that this was suggested to purposely hide a stress problem, I can say that NOT ONLY IS THIS NOT CORRECT, IT IS DISHONEST AND A SEVERE BREACH OF ETHICS! You were right to question such tactics. At all of the places that I have worked, someone pulling this kind of stunt would be fired on the spot and booted right out the door with a swift kick from a size twelve steel toe.

To all,
Mr. Monroe may be correct in raising the question of ethics, however , I feel that he could have expressed his anguish in a less harsh way as this is a forum for technical discussion and not street fighting.

Regards

anindya et. al.,

If your sensitivities have been offended oh well. Mr. Monroe is just going on the record for his experience at the firms he has worked for in the past.

I am very sensitive to the question of ethics surronding this work. My experience has been there is a great deal of questionable ethical behaviour involved by a great many people involved in this industry.


Perhaps Mr. Monroes choice of metaphors offends you but here in the U.S.A. we are all allowed free speech (at least today)so long as it does not defame or cause harm. Hopefully this freedom will be extended and honored by COADE the owners of this bulletin board.

Mr. Luf,

I appreciate your view on ethics as I and so does everybody participating in the form strongly believe in the same.There cannot be any question on compromising on the same.My reply was not on the question of ethics , but on the language used. Language does not harm or defame anybody, it only hurts and we are not here to hurt anybody's feeling, but to correct him if he is wrong and this I think can be done in a polite way .

I am not specifically saying anything about Mr. Monroe's reply,but of late I have found many users using hurtful and derogatory expressions, which certainly is not the objective of this forum.This is a forum where everybody participates to share and gain knowledge.Freedom of speech and expression should not be used in an abusive way,at least not in a forum like this.

Regards