Sandblasting

Date 02-22-2009 06:02

I know several companies who do blast before welding, mostly for bridge work. (Do remove the sand) If I were to do a PT I would not recommend it, that would smear the surface hiding cracks.
Maybe someone else will have better insight on this subject.

By gndchuck (**)

Date 02-22-2009 13:05

When we have to do a underwater weld, we prefer to grit (sand) blast the area. But before we do we MAG the area and any indications we either drill stop holes, cut out, or mark with a low distortion punch. If we mark it then we take pictures before and after and let the engineers decide what to do. For us anyway I know that grit blasting is faster and more fun. But then again we only SMAW for wet welding.

Charles

By 803056

Date 02-23-2009 14:43

I don't see how grit blasting with silicon oxide (sand), crushed slag, or steel shot should be detrimental to welding carbon steel. If anything, it should be beneficial to welding if it removes the mill scale. The minimal residue left on the steel should be countered by the deoxidizers employed by the filler metal, be it GMAW, FCAW, SMAW, ect. A light brushing with a wire brush should easily remove any residue that remains if there is a concern.

As for penetrant and magnetic particle testing, anything that can fill the discontinuity will interfere with PT. However, it should have little effect on the MT. Steel is not soft (as is the case with aluminum or some copper alloys), so there is little danger of smearing the metal. MT can be performed even when the discontinuity is "filled" with contaminants such as paint, carbon, etc.

Best regards - Al

By James Corbin (**)

Date 02-24-2009 22:05

A warning about sand blast smearing was part of my PT training class years ago. Course grit such as "Black Beauty" will dig deep into steel as much as 3 to 4 mil when SP10 or SP5 is specified. We were told to chemical clean & PT first, then blast if needed to remove surface coatings for weld repairs. If it's already blasted when you get called to do a PT note the blasted surface on your report.
Here is a link & quote.

http://www.ndt-ed.org/EducationResources/CommunityCollege/PenetrantTest/MethodsTech/preparation.htm

"It is very important that the material being inspected has not been smeared across its own surface during machining or cleaning operations. It is well recognized that machining, honing, lapping, hand sanding, hand scraping, grit blasting, tumble deburring, and peening operations can cause some materials to smear. It is perhaps less recognized that some cleaning operations, such as steam cleaning, can also cause metal smearing in the softer materials."

By 803056

Date 02-25-2009 03:45

The key words in your response are "soft metals". Any operation that could "clog" or interfere with the discontinuity being open to the surface is going to have a detrimental effect on the outcome of the penetrant test. In addition to grit blasting it could include any operation that uses water, acids, alkalies, solvents, etc. that could fill the open discontinuity.

In the laboratory it is common practice to place the test piece in a drying oven to ensure there is no liquids in the discontinuity. The same procedure is difficult at best when the test is performed in the field, but it is no less important.

Best regards - Al

By James Corbin (**)

Date 02-25-2009 04:02

Agreed

By 803056

Date 02-25-2009 04:05

You and I are usually on the same wavelength. You started the conversation, I simply tried to expand on the kernal you planted.

Best regards - Al

By James Corbin (**)

Date 02-25-2009 04:24

After watching our blaster dude leave his leaking valve nozzle pointed at an a588 beam and then go to break which in 10 minutes wore a hole through the Bridge beam he was cleaning I never doubted the abrasiveness of Black Beauty or the blasts ability to smear steel no matter how "hard" I think it might be. I just follow the PT class instructors warning about PT on any blasted material and the possibility missing a potential crack or a cracks volume being hidden.
Is that what you had in mind?

By 803056

Date 02-25-2009 04:37 Edited 02-25-2009 04:43

Amen to that, but I believe it wasn't so much a case of smearing the metal as it was a case of "cutting the metal" by abrasion.

"Black Beauty" is crushed slag if I remember correctly. It is abrasive in contrast with steel bead, which would have a greater tendency to smear and peen the steel surface.

In any case, if PT is the NDT to be performed, dirt and debris from grit blasting, shot blasting, etc., can definitely "clog" the openings to the discontinuity. That alone is a good argument for using MT on ferromagnetic materials.

Al

By Mikeqc1 (****)

Date 02-25-2009 19:57 Edited 02-25-2009 20:00

I have had concerns on this, i have been told that sand blasting could cause aluminum (or whatever the medium is) to be inprgnated in the steel , so wirebrushing it after blasting may reduce contamination..
the parts failed rt. so after sand blasting i had the weld prep machine to bright metal and that seemed to do the trick.

MDK

By CHGuilford (****)

Date 02-26-2009 14:41

I have heard the concerns that blasting steel could close up surface indications so that penetrant will not work well. And I suppose that must be based on actual experience because NDE folks are typically very conservative in what they do. So I wil not say it can't happen.

I can say though that we often experience just the opposite. Blasted welds typically will open up any porosity or other discontinuities that are not obvious visually before blasting. I have been embarrased on some welds where I have gone over every fraction of an inch of the welds, using a strong flashlight, only to have some horrendous defects show up after blasting. It's happened to all of us here enough so that we inspect welds after blast as well as before. This is not simply a matter of the dark surfaces becoming lighter so that it's easier to see the defects - much of what we find was not there before blasting or was small enough to accept.

Back to the original question: I have not seen that blasting interferes with the welding overall. Spatter sticks on more, smokes and smut doesn't brush off as easily, but I haven't seen that weld quality suffers. I have seen deep blast profiles make "wetting out" at the weld toes a little more difficult, but that is mostly when very large blast media is used so that you get about a 4 or 5 mil blast profile (not common practice). However even in that case, the problem is not as bad as it would be to try to weld over heavy mill scale or if dealing with inconsistent joint fit-ups.

By 803056

Date 02-26-2009 23:10

Chet, I've had the same result as you with regards to discontinuities being revealed after grit blasting. In my case the media was steel grit and all welded surfaces were blasted before welding and all surfaces before painting. Leave it to the painter to see problems in areas that were already accepted by the shop inspectors.

I usually use MT when checking welded steel structures, but I have used PT on occasion. I haven't had any real problems where the MT picked up something PT missed, but in general I don't PT grit blasted surfaces if for no other reason that the clean up is time consuming due to the roughness. It's like PTing a sand casting!

Best regards - Al

By Jim Hughes (***)

Date 02-26-2009 15:52 Edited 03-01-2009 17:45

Sloppyjoe,
I know first hand that silicon (sand) can be detrimental to the weld. We were not using GMAW or FCAW but the GTAW process on P22 material we kept having rejects. (Radiography) We were getting rounded indication and could not figure out why. I did every thing I knew to do including checking TIG rigs to see if they were sucking air, to checking flow coming out of the gas lens, to checking the high pressure argon bottles, to watching the welders during welding to see if I could see anything. As I would watch the welding I could see little sparklies coming off the weld puddle. When I would ask the welders if they saw anything they would say no. I finley sent a piece out for failure analysis. The report came back that there was silicon or sand imbedded in the pipe. When we changed our base metal prep from flapper wheel to cone rock it cleaned right up and we quiet busting film. We called this the gland steam massacre because we busted 27 welds before we figured out the problem. The cost of failure analysis $1500. Money well spent. This was a Hitachi boiler and the pipe material came from oversees.

Jim Hughes