National Average 1

[geauxinspect]

I'm in search of some information and I can not find ANYTHING! I'm looking for the National Average for weld reject rates. Does anyone have any information on this? I'd like a URL where I can direct someone to this info if requested. I have been told many times by many different people that it's in the neighborhood of 1.5 to 2.5 percent. Thanks in advance!

 

[HgTX]

That doesn't sound like an answerable question. There's no national clearinghouse of all weld inspections ever performed in every industry in every product fabricated to ever standard. There are no doubt plenty of companies who couldn't answer that question about themselves, never mind any kind of national statistic.

Hg

Eng-Tips policies: faq731-376

 

[geauxinspect]

This is no secret, but in today's world its all about documentation and records. I would think that as far as contractors and client companies, their should be some type of statistics put together for some type of comparison and information. BUUUTTT I guess ya never know...

 

[PSE]

geauxinspect,

Some companies may indeed maintain documentation and quality records on welding, they likely will not make it public. Perhaps with Non Disclosure Agreements, you could obtain the information from a company of interest.

It would be a challenge for gathering weld quality information on a national or even regional or local basis. You would need to capture information on materials, methods, environment, and training or skill level I would think at a minimum. That information would have to be compared against an appropriate standard for the weld type in order to generate any meaningful data.

Regards,

 

[geauxinspect]

I felt like I would not be able to find any information on this... I was just hoping that I would find a surprise somewheres! Thanks for your responses...

 

[unclesyd]

As stated above I don't think that you will find a national average on defective welds or even a consensus on what is a an acceptable reject rate for welds. To collect this data is extremely tedious and time consuming. I've collected data on a small number of projects to track RT costs and ended up tracking weld reject rates by default. The redeeming factor while I was doing this was that the project called for 100% RT and VT. Any less percentage inspections would require much more effort and time. You also have qualify whether would want all rejects, repairable rejects or cutout rejects.

The information you want would be generated by "Weld Mapping" a process is just starting to take off because of requirements of various agencies for traceability.

There is some software available now that will greatly simplify the collection and analysis of welding data.
Here is one from C-Spec. I haven't used this one so I don't know at what level or how it sets the flags on weld rejects. I would assume that the owner sets the level.
Look at the reports and analysis section of the software description.

http://weldmap.com/

 

[unclesyd]

While talking with my NDT buddy one of them brain things came along about checking with any large inspection company that does weld inspection on your request based on their reader sheets.
They may have a number for rejects based on the number of welds they have inspected but not the total number of welds overall. This may only be RT but could include all of inspection disciplines. One would have to know the percentage for RT inspection on each job to back out a number for the expected rejects in the overall number of welds assuming the whole population is represented by inspected sample.

Again this would only be for small sample in one segment of your total area of concern. As stated above a lot of companies get quite upset when you start wanting to discuss this. Construction companies consider this proprietary and and industry tends to treat this as confidential information. In my previous post and posts in other threads I have discussed the problems with welding socket weld fittings particularly one project of over 3000 inspections. I was deposed twice about collecting this information, the whys and wherefores.

 

[JoeTank]

From my own experience and observation...
5% or less would not upset me.
5% to 10% would catch my attention.
Greater than 10% would cause me to jump up and down and yell.


Joe Tank

 

[JoeTank]

Sorry, I hit send too quickly...
I'm speaking from experience with tanks and vessels to API and ASME Codes.

Joe Tank

 

[geauxinspect]

Yes, Same here... We average somewhere between 1% & 5% sometimes we may jump up between 5 & 10, when so my boss starts asking questions! As a company we are less than 5%.

 

[Pud]

You might try these:

http://tinyurl.com/3h6ja9

Generally acknowledged as one of (if not the) World's best...

Cheers


Harry