Fillet Weld Break Test Question

Date 11-02-2005 14:57

Is this slag line actually in the thin gap between the two plates...and is it past the efective throat and the fillet root profile...or is inside the required fillet cross section? Depending on the tightness of the joint fit up I have seen some slag flow under and between the two plates, past the normal weld volume and would consider that acceptable. Any slag inside the required weld volume would not be acceptable. Saw cut and etch one before bending to see the weld. you should see the entire fillet profile and where the slag really is. I would be curious if you welded two plates that had both surfaces milled to a nice, really tight clamped together fit-up -would you get the same slag?

By rsx-s-02 (*)

Date 11-02-2005 15:17

The slag line is right at the root of the weld, where the two plates come together. I have cut cross sections and you can see where the slag is. I have pictures if you would like to see, just send me your email and i can send them. If slag being trapped there is acceptable, is there a guide that defines how much is acceptable?

By swnorris (****)

Date 11-02-2005 16:26

A fillet weld where the weld metal does not penetrate to the root could be caused by insufficient root gap, current too low or incorrect angle of electrode. Have you considered any of these?

By jon20013 (*****)

Date 11-02-2005 16:39

In my humble opinion, if the slag in past the point of the "root" i.e., the 90 degree area of the plates coming together, then it would be acceptable.

I've experienced this problem in the past when doing groove welds with backing that were then subjected to RT, they were rejected numberous times but the slag line was always between the backing plate and the base metal, not in the weld itself... took quite a while to convince the radiographic interpreter of the situation but eventually proved the point and welds were subsequently accepted. I believe the slag in these cases probably got ahead of the puddle and flowed into the easiest path of escape.

The joint configuration was one where backing was necessary otherwise we would have simply done open butt and back-gouged; it wasn't simply a matter of trying to be a pain to the NDE Tech!

By rsliker (*)

Date 11-02-2005 16:42

From your description, the slag, as you describe it, is between the two plates and extends in a direction running from the joint root and between the two test plates.
D1.1-- " showing complete fusion to the root of the joint". I seem to recall, in earlier versions of D1.1 the phrase, "but not necessarily beyond" but I may be mistaken.
If your macroetch show this line extending continiously from the root and between the two plates, then I wouldn't think you have not passed the acceptance criteria.
It may be, as Bill described, that you do not have a good fit between the test plates. I 've sometimes seen what you are decribing occur when plates with rolled edges are used without having been machined.

By rsx-s-02 (*)

Date 11-02-2005 17:19

I am currently a CWI and we have tried numerous things. Position of the electrode, higher heat, sawing the edges of the plates for a better fit. Getting a better fit did reduce the amount of slag and i wouldn't reject it, but it seems that the weld is not penetrationg the actual root. Milling the edges of every plate is not practical. There is good penetration on both plates. We were using a .045 wire and have changed to a .052. The slag seems to be getting ahead of the pool and flowing into the joint. Perhaps this is normal. I would be interested n knowing if any other structural fabricators coiuld try a test like this and let me know of their results.

Thanks for all the feedback!!!!

By CHGuilford (****)

Date 11-02-2005 17:58

FCAW is prone to trapping slag at the fillet roots. As was already mentioned, all a fillet has to do is penetrate to the root, but not necessarily beyond. Of course it's nicer if it does. If your slag line is beyond the fillet triangle, you don't really have a problem. If the slag is within the triangle you do.

A change in parameters and technique will cure the problem. Basically you need to turn up your voltage and wire enough so that you can stay on the leading edge of the puddle and still build up your fillet size (for those who don't know this - don't get carried away and go nuts with the dials. small changes ar all that is needed).
Holding the puddle back to build up size will only allow the slag to roll ahead and be trapped along the root.
Pushing the puddle is not good unless you have small fillets to make. A drag angle of of about 10 to 15 degrees should work well. If the fillets are horizontal, I would try "laying" the gun to about 35 - 40 degrees from horizontal and "hold" the wiire about 1 -2 wire diameters out from the vertical member. That should allow the weld to penetrate and wash in very nicely with no undercut. If you do get overlap or undercut simply adjust your angle and move the wire in or out until you get the profile right.
But again, don't hold back to build size; instead turn up your wire feed speed and voltage and move faster. No fancy whip or weave motion needed. I've generated a small truck load of cut-up tee tests for welder training and qualification with this method.

Chet Guilford

By Lawrence (*****)

Date 11-02-2005 18:38

Chet.

This responce ought to be published in a textbook.

Thanks.

By jon20013 (*****)

Date 11-02-2005 18:50

I agree, good one Chet!

By Shane Feder (****)

Date 11-02-2005 23:10

rsx-s-02,
Just a couple of small points to add to Chets posting.
If you can't mill your test plates try putting them in a vice and using a 7 or 9" grinder and laying the new grinding disc flat on the edge.Take 1/16 to 1/8" (See what this forum is doing to me, I am now talking imperial) off the edge keeping grinder perfectly level and you should have a nice square edge on plate.This will help with fit up and not allow any voids to be filled with slag.
However, you cannot do this on all your production welds so you have to do as Chet has said and play around with your parameters.
Do you need to try welding smaller, hotter, faster fillets?
Regards,
Shane

By rsx-s-02 (*)

Date 11-03-2005 15:55

The majority of welds that we do are 1/4" or 5/16". We do these in a single pass using a .045 wire. They need to travel a bit slower to build the size. I had the wire changed to .052 but the amount of smoke it was creating was causing a health issue. So we are going back to the .045. Additional ventilation at this time is very costly and our welders health is more important. Basically from all my research and from what I’ve heard here is that the fillets are acceptable. It's simply a case of the slag filling in a tiny space at the root. The steel plates are not perfectly square so it makes it easy for the slag to get trapped there. I really opened a can of worms at work with my concerns but hey, that's my job! This forum is definitely an asset to the welding community. Thank you everyone for all your feedback. I can sleep again.

By CHGuilford (****)

Date 11-04-2005 17:30

Gentlemen, thanks for the kind comments.