Full blown newbie question

[e57]

Fresh meat here - looking for free information (Happy to reciprocate in any electrical question you guys might have as I know more than I really need to there.)

So anyway - I'm building a work bench/out-feed/tool storage/assembly table/cross-cut sled/GP all-use table for my little garage shop. It really need to be as flat as possible - and WAS right until I messed it up yesterday.

It is a 2X2X.125" Sq Tube framed cart on V-grooved casters for placement from stored to it's working area - where it will lock in place. It was as near perfect as could be yesterday - right before I welded the brackets on the underside of the upper rails. A MDF, and poplar framed torsion box table top will go between the two upper rails flush at the top - so it really needs to be flat.


I wasn't naive enough to think that it would not flex when I added the brackets - But I did think that it would have gone the other way - up. (Which would have been beneficial as it will have some significant weight on it to counter act that)

What I did was put 3 beads on the inside edges (ends - then the middle) of some 1 1/2" angle to make the brackets - and the beads weren't bad considering it was the first day I opened the tank...


I should also mention I just changed processes, from flux cored to MIG (C25). With flux cored my welds always seemed to expand - then contract just a little tighter. (Which is why I thought the 2x2 tube would have flexed up.) The MIG seems to have for a few reasons acted very differently - please correct me if I'm wrong here:

1. Heated the base material much faster, and much more due to the increased voltage, and wire feed.
2. Once heated, LOCKED the expansion into the base and attachment materials.

So anyway, the main questions are: 'How do I correct it?' (Most important at the moment) and 'How do I avoid it?'

Thanks in advance for your patience....

[Jonesy70]

I think I get what your saying...when you weldede your brackets...you should have stong backed the other side (Tacked a flatbar on the opposite side to help keep it straight...To correct it, cut the bars apart at the corners and put them in a press to straighten them...then weld the corners back together...heat works in wonderous ways...for the thin material you have...you have alot of weld on the brackets...which caused the bowing

[DSW]

Nice explanation and I love the diagrams. Depending on what use you plan to put the table to, you could alwasy just shim the top to flat. This should be the simpleest thing to do. I would take a good staright edge, say a good 6' level and determine how thick the shims will need to be. The shims could be tacked in place or just held by the bolts attaching the top. It is very hard to get a thin steel frame perfectly flat. The stock often has a bend from transport, heat from welding warps it ( you found this out), uneven gaps due to welding the far end and so on.

Good luck.

[Craig in Denver]

e57:
I agree with your thoughts. Since all welds shrink, the bottom should have pulled in and bowed the top surface UP. My only consolation to you is: now both of us know that 1/8" may be too light for the needed rigidity.

My biggest aggitation in welding is trying to 'out guess' the shrinkage.

As DSW said, awesome drawings to convey the idea.

[dave powelson]

So anyway, the main questions are: 'How do I correct it?' (Most important at the moment) and 'How do I avoid it?'

1-You don't mention 'how much' it bowed. This would help. Presumably you do have a decent straight edge, at least 6' long.


2-Correction? You'll need the straight edge to gauge what's happening. You can either camber by welding (easier) or by heating. This cambering will not provide a dead flat surface--due to the welded angles, but it can raise the basic contour uP. At this point, you also may have some twist in the tube tops-relative to each other. This cannot be readily corrected.

a-Quick and dirty is to camber the beams, by stitch welding
-mark off 4" segments on tube bottoms the entire beam length. You will not be stitching over the angles, just up to them.
-This stitch welding is in the center of the beam-mark a centerline
-Set the welder for a small, not hot, not super penetration, high,ropy kind of bead--
one that's just wetting into the edges. (you can always do more passes--but, once the bead is down--that's it-as far as the amount of camber it's going to pull.)

-Beginning at one end--'Backstep'-that is, make direction of this weld--towards that closest end. Repeat on the other end, starting away from the close end, moving towards it. Back and forth from one end to the other, stitch, stitch, stitch........until you finish at the center of the beam.

-During this stitching, pay attention to what's happening with your straightedge. Finish this first ropy pass, even if it doesn't appear to be correcting the unknown amount desired.....then let it air cool and do again, if you think you need to.

b--Heating the topside to shrink it, is the 'other way'.

3-Avoidance?

For starters, see my post linked below. It gives one some clues on distortion control,
which is summarily ignored by most.

safe temperature for shrinkage

http://weldingweb.com/showthread.php...445#post190445

a-do not weld across a beam's length--just along it. This is simply sane practice--unless one's creating a shear type application, in heavy stock and doesn't mind distortion.

b-Pre-stress the beam--mechanically with one or two temporary tubes to bow it up, prior to welding

c-using 2 x 2 angle with ground bevels on the edges, weld along the sides of the angle--only to the radii on the 2 x 2 tube. The corners of a tube are much more resistant to weld pulling than the flat wall.

d-using heavier wall tube, as other's have mentioned, helps limit the distortion. 1/8 wall moves all over the place--as you now know.

[Craig in Denver]

"b--Heating the topside to shrink it, is the 'other way'."

Dave:
Isn't the topside already too short?

Craig (I hate questioning the wizards, see my sig)

[dave powelson]

Originally Posted by Craig in Denver

"b--Heating the topside to shrink it, is the 'other way'."

Dave:
Isn't the topside already too short?

Craig (I hate questioning the wizards, see my sig)

Heat with or without wet cooling will pull the bowed-Down top back up--to some degree. I don't mess around with heat shrinking. I'll defer this heating biz--to someone who's into it.

I plan my work on paper, etc.--and consider distortion, what may happen before the fact, how it will be controlled, what's acceptable, etc. then check my fitup as I go......I worry like an OLD WOMAN.

I suppose I should post, some real world examples 'your's truly' has, of cambering beams and successfully pre-stressing beams/assemblies on the first try.

To repeat--once again:

That silly little sketch and 'how-to', and that elusive, nebulous, conundrum of 'how much?', in the link below

http://weldingweb.com/showthread.php...445#post190445

is 3,945% more than any 'expert' can provide you with on pre-stressing for light or intermittent welds....and references to the uniform load deflection of a beam, for the technically minded.

[I suspect, there's others in this forum that may have more real world data--if they will share it. Sure, this is proprietary to Blackbird Co.--I'm the owner and disclosing this, does not 'hurt' me. The name of the game, is to help each other do a better job. In 20 years in my shop, I've helped and been significantly helped by my 'competition'.]

[e57]

Gonna give this 'stitching' a shot - let you know....

FYI it was only 3/16" but toally blows what I was going to do with the rest of this.

Oh and the drawings are by "sketchup" the free version - really easy once you get the hang of it.

[dave powelson]

Originally Posted by e57

Gonna give this 'stitching' a shot - let you know....

FYI it was only 3/16" but toally blows what I was going to do with the rest of this.

Oh and the drawings are by "sketchup" the free version - really easy once you get the hang of it.

Suggestion--If you can spring these beams uP some, before the stitching--this will help. Like with temporary tube, steel, wood blocking. Don't spring more than the flat for starters.
Be patient. If it needs another pass, so be it.

(ideally without springing the bottom beams Down)

[Craig in Denver]

Dave:
Thanks for the link. I had seen that thread, but before your post; which contained the answer.

and that elusive, nebulous, conundrum of 'how much?'
Amen.

[Broccoli1]

If you don't have a torch, weld beads on the top side where the brackets are and then grind the beads off.

[Oldtimer]

Heating it red hot on the opposite side of the tab will pull it back. It is a guess how much heat to use when you do this, too little heat and you have to do it again, too much and you have a problem the other way. Let it cool slow or quench it, either way will work. I would use a hydraulic jack in this instance. Block up under the bottom rail so you won't bend it, put a piece of 1/4" p;ate slightly wider than the tube under the top rail and push the top rail slightly past straight, let the pressure off and check it. Two jacks would be even better. With a little patience you can get it perfectly straight.

[e57]

Originally Posted by dave powelson

So anyway, the main questions are: 'How do I correct it?' (Most important at the moment) and 'How do I avoid it?'

2-Correction? You'll need the straight edge to gauge what's happening. You can either camber by welding (easier) or by heating. This cambering will not provide a dead flat surface--due to the welded angles, but it can raise the basic contour uP. At this point, you also may have some twist in the tube tops-relative to each other. This cannot be readily corrected.

a-Quick and dirty is to camber the beams, by stitch welding
-mark off 4" segments on tube bottoms the entire beam length. You will not be stitching over the angles, just up to them.
-This stitch welding is in the center of the beam-mark a centerline
-Set the welder for a small, not hot, not super penetration, high,ropy kind of bead--
one that's just wetting into the edges. (you can always do more passes--but, once the bead is down--that's it-as far as the amount of camber it's going to pull.)

-Beginning at one end--'Backstep'-that is, make direction of this weld--towards that closest end. Repeat on the other end, starting away from the close end, moving towards it. Back and forth from one end to the other, stitch, stitch, stitch........until you finish at the center of the beam.

-During this stitching, pay attention to what's happening with your straightedge. Finish this first ropy pass, even if it doesn't appear to be correcting the unknown amount desired.....then let it air cool and do again, if you think you need to.

Tried the torch, jacks, just short of starting over - but I have to say this was the best new thing I have learned in some time - thought it was possible but did not think it would have such facinating effects. The item in red.... missed that wrapped the other way - horizontaly... Corrected it the same way. Also noted that if you over correct you can grind tham off and lessen the effect of each. Anyway - its coming out pretty well now - nearly dead flat - and about to start the wood portion soon.

[dave powelson]

Originally Posted by e57

Tried the torch, jacks, just short of starting over - but I have to say this was the best new thing I have learned in some time - thought it was possible but did not think it would have such facinating effects. The item in red.... missed that wrapped the other way - horizontaly... Corrected it the same way. Also noted that if you over correct you can grind tham off and lessen the effect of each. Anyway - its coming out pretty well now - nearly dead flat - and about to start the wood portion soon.

Good to hear some feedback!

On other fab. items, one can camber beams, prior to fitup, to counteract deflection-IOW, pre-stressing the beam. Running a sample beam first, logging the weld parameters, helps.