By -
Date 05-04-2005 16:45
Can anyone out there provide me with some advice on fabbing some simple welds with relatively controlled defects (meaning that before I weld it I would actually have a reasonably good idea of how large, what shape, location, etc) for use in NDT practical testing for MT, PT, UT, etc?
Thanks,
Mankenberg
prep your part and place some copper wire in the grove I know it works very good for UT.
Mankenberg:
I have lot's of sample welds I use for training and demonstrations.
Lot's of ways to induce discontinuities into welds:
*inserting chunks of slag during welding
*applying weld spatter paste in appropriate areas of the weldment
*welding over water
*inserting disimmilar metallic materials into the weld pool
*drill holes to simulate pores
* depleting the weld of shielding gas
The list can become very extensive.
I also have the radiographs for these same welds to locate the discontinuities along the weld axis. UT reports to determine specific depth and orientation, etc. For some reason I've had several UT tech's offer to puchase these (with all the NDT data of course). Must be like gold to these guys.
Us UT nuts like to have practice plates! :)
John Wright
I've looked into buying this type of plate with built in defects and I can tell you, they are expensive. Creating your own sounds like a reasonable solution.
John Wright
By -
Date 05-04-2005 18:25
Thanks for the replies gentlemen. Some good ideas here. I'm hoping to minimize how much trial and error I have to go through to get suitable plates. I need to make test plates for Lev I and II practicals for MT, PT, and UT, and I do not want the discontinuities so gross that my eight year old son could find them. I want discontinuities that can be found, but at the same time are of such size, orientation, location, etc that the test candidates will have to refer to the procedure/acceptance criteria in order to determine the proper disposition.
I'll post again on this topic when I've made a few up and let you know how they came out.
Mankenberg
Hi Kip. We made some test samples in the past. You can do many things.
But if you want UT reflectors in a weld, get a very small side grinding disk and start welding the joint. You can place notches in the weld at the fusion line in any pass you prefer and just weld over them. Being careful not to comnsume them and fuse them.
You will pretty much know the length but the thru wall dimension would have to be verifed and accepted by UT. The after completion, you can place these notches in the root area or even in the cap.
You can also make some of them acceptable or rejectable.
By -
Date 05-05-2005 15:35
Yeah Ed,
Thanks for the suggestion. I've done something similar on TKY mockups, and it is actually easier due to the curved surfaces as you can use a hacksaw for the same purpose.
Kip
Very fine cracks can be induced in SS with zinc . Experiment and check with liquid penetrant.
To make longitudinal cracks, try a VERY high depth to width ratio notch done with a cutoff blade. Make the notch deeper than you will penetrate. Then weld over the notch using GTAW, High Heat, and NO Filler metal. You shoud see a crack upon cooling or sooner. You can then backgouge the other side to get rid of the notch left by the grinding wheel and weld the backside up. The crack MAY follow.
Have a good day
Gerald
Gerald,
One plate that I saw with cracks built in was actually bent and then straightend. The cracks were in no way visible to the eye and NDT was needed to verify that they were actually there.
John Wright
By -
Date 05-05-2005 15:36
Thanks for the excellent suggestions gentlemen. I'll let you know how they turn out.
Mankenberg
And I resonded to your posts? What was I thinking of?
By -
Date 03-03-2006 14:02
vonash,
Please feel free to not respond to my posts.
Best regards,
Mankenberg
a light sand blast will drop the Db rating of a ut indication. this can be done repeatedly to change the class of an indication
hogan,
I'd like to hear more about your sandblasting technique. I'm guessing that the sandblasting smooths the defect in order to change it's reflective signal?
John Wright
John,
I think he is talking about blasting the scanning face. We have several ASME blocks with different finishes of surface roughness and also some painted and coated ones.
~thirdeye~
Hello Mankenberg;
I've used copper wire to produce cracks and I've used drilled holes to represent porosity, but tungsten electrodes placed in the weld work very nicely. I've made test pieces for UT by inserting tungsten electrode of various diameters and lengths into the weld. To simulate a large reflective surface that can represent a crack, I simply grind the tungsten flat along the length. By orienting the flat, I can control the amplitude of the signal reflected off the tungsten. The high melting temperature and limited solubility of the tungsten results in good control over the size of the reflector.
Best regards - Al
Several days ago I created LOF in a plate and it worked out pretty good. I beveled one plate at a 45 and left a square section that would not get fused when welded. I welded the joint up and back gouged the second side leaving the nonfused section and rewelded the second side. After I ground the plate back smooth on both sides, it all blended nicely to where you could not find the joint by eye. It turned out great and I've been using it to train another UT operator in our shop. I'm hoping to be able to lean on him when I get swamped with UT. He's been scanning base plates and column flanges with me and has done well with straight beam, now on to some angle beam training :).
John Wright
This might be my favorite post.
Finally, we are training the forum.
Even if only in a small way, this thread might enhance the welding/joining profession.
I have no idea the impact this one idea (post) may have on the participants; this has to be good.
Training is the key. We who know must give it over to those who are coming up; that is our function, to teach.
Best regards,
Vonash
the sandblasting will cause the surface roughness in the scanning area to increase, this will drop the amount of sound that is sent and recieved, compared to your cal standard
