[jkhackney]

Hi,
this is my first post here. I've used MIG on steel for about 9 years and I also solder aluminum with oxyacetylene and now I'd like to TIG thin aluminum.

I bought a used ESAB DTB250 square wave (380V), rented a bottle of argon, and worked through some online tutorials and TIG drills. I have no pedal for it, but I think that I'm pretty confident in practice tacking (with and without filler), butt, lap, and corner joints on the 1mm-1.5mm metal, that I can tackle most of the spots I have to repair.

However I want to be able to spot weld with it: 1/16th" thick floor support channels to a 3/32" floor. I want to use the TIG to melt through the thin metal to afix it to the thicker floor.

I tried many techniques on scrap but almost always just melt the thin metal away, with a small pool forming in the thick metal: with and without first drilling a hole; with and without a heat sink; with ramping current and with immediate full current; different cup sizes and tungsten sizes. I once even melted the pink cup and broke it. Rarely will the utterly destroyed top piece actually stick to the bottom piece.

I've tried with a heat sink (steel) on the top piece around the cup, mimicking the Lincoln spot welding kit, and this succeeded in making a passable spot weld only once, but also melted to my cup once!

Is this an impossible thing to do? The youtube videos and advertisements only ever show someone doing it on steel ... Can anyone offer a magic tip?

Thanks!
Jeremy

[Olivero]

If I was you I would drill a 1/4" or 1/2" hole in your top piece and just weld it to the lower piece that way, kind of run in circles with the hole until its filled.

Repeat that procedure in multiple areas and you should have yourself some decent "spot welds"

[AndersK]

I've experimented a bit with this too and only success was clamping really hard, otherwise air between oxidised sheets oxidized bottom sheet and made fusion impossible. You fight different thickness but a good heatsink on top might help as you describe.

I was never satisfied with strength of the ones I made so have ended with drilling and plug welding so far.

[jkhackney]

Thanks,

I'd be happy to achieve a plug weld, like I do with the MIG on steel, 6-7mm diameter or so.

I have yet to develop that skill with a TIG on aluminum! The hole gets bigger and bigger until it hits the edge of the thin top piece, without melting the bottom piece, then when it hits the edge I have a chance to make a tack. It sticks but looks really bad.

Is the secret a lot of pressure the heat sink? What amperage settings, relative to a "normal" weld?

I can't clamp the work since it's an entire floor, about 3x4'. To get pressure on my practice pieces, I've been standing on my heat sink (it's a big open-end wrench). But it mostly just blocks my view. I also make sure there's NO heat sink on the bottom piece, so it gets as warm as possible.

I'll search the forum but if anyone has tips for plug welds I'd be glad to have them here, too!

Jeremy

[AndersK]

The main problem with a small hole plug weld is to get the arc at the bottom piece, it likes better to jump to the edge of the drilled hole. Try a small blunt tungsten and set AC frequency to max, helps focusing the arc. Also make sure you have good grounding of the bottom piece.

Some might say high frequency pulse might help too but I haven't tried that.

Welcome to the forum, by the way

[cj737]

Actually Jody did a video on this and used DC to spot tack aluminum. You can get penetration without the heat that melts and distorts thin aluminum. Search his website or YouTube videos for the tutorial.

[Olivero]

I would use a sharp tungsten, it won't last but it will get you started on the right piece.

[jkhackney]

Thanks for the welcome!

And the tips!

I looked at the DC electrode positive videos, which seems to be useful for thin plates and worth a try.

DC negative seems to require helium and is for thicker pieces.

I can't adjust the AC frequency nor do I have a pulse or other special effect to plug into my machine. I can get an aftermarket pedal or make one, if necessary.

I'll report back after the weekend's experiments with the ground clamp on the bottom plate on AC, vs. DC negative!

Happy new year!
Jeremy
PS- it's a Land Rover floor whose stiffeners were originally spot welded on (in the 1950s!), which is why I wanted to mimic that appearance and strength. My alternatives are rivets or adhesive if I can't get the TIG welds right.

[Oscar]

How much amperage do you have at your disposal and how much of it are you using? Also, how long is it taking you to ramp up to your amperage? How long are you holding it there?

[Erich]

Drill a hole in the floor, clamp the parts together and fill the hole with weld. It is called a rosette weld and it will look similar to a spot weld and will be as strong or stronger. I used this technique while replacing rusted panels on a 69 Mustang.

[motox]

i have done this on aluminum 1/8 sheet to square tubing.
pre drill the holes in the sheet and then use a 82 degree
(or 60 or what ever you have) and bore a chamfer.
the thin edge from the chamfer lets you make a plug or rosette weld
without chasing the arc.
craig

[motox]

hope this helps
craig

[jkhackney]

Hi,
thanks again for the tips.

My experiments on AC have been with 150-180A and 50% balance, obviously I cooked the 1/16th tungsten and narrow cup with the higher amperage I tried the rosette/plug welds with about 100A. This metal welds nicely with 60-70 amps depending on whether there's a heat sink behind it. The welder goes to 250A, not sure my 380V/16A socket delivers that much current though!

I was going to weld the supports (1/16") to the floor (ca. 3/32"), rather than the other way.

Are rosette/plug welds easier from the thicker to the thinner metal, or vice-versa? My problem is the top layer melting away, so a beveled hole may help. I did read that a less-pointed electrode will concentrate the arc in the center better. I haven't tried that, yet.

The hole for these rosette (plug) welds: what diameter would you recommend for welding a 1/16th thick piece to a 3/32 and vice-versa? Also what cup, electrode diameter and amperage?

Meanwhile I made progress with DC + electrode today until I ran out of argon. 150A, 1/8" tungsten, size 8 cup (so it won't melt with the hot tungsten). I made a heat sink/pressure foot from scrap steel. These are the original 60 year old supports I have to replace on the underside of the floor; the holes in them are where I drilled out the original spot welds.





The welds with the arrows would be acceptable on the car, though a smaller diameter would be better. All these welds left a penetration mark on the opposite side. I haven't tested the weld strength, yet ("nugget" test?). These pieces are trapped in place by crosspieces that are rivetted over them. The welds predominantly have to withstand shear as the floor bends.

Jeremy

[cj737]

I would say that you are using way too much heat. 150a for 1/16" and 3/32" is overboard. When you use a 1/8" tungsten, you are depriving the weld from some amount of heat as the electrode is consuming current.

I'd be using a #5/#6 cup, 3/32" electrode, and probably a lot closer to 100amps. I'd rather loiter on the 3/32" alloy waiting for it to heat than blow it away with too much amperage. by the time it heats up, the 1/16" support is already hot so fusion is simple.

Just my worthless opinion.

[Oscar]

Here is the recipe:

1 part: 5/32" hole in 1/16" thick aluminum
1 part: 3/32" 2%La tungsten sharpened to a 90-130° included angle.
2 parts: 0.3-0.5s spot timer
1 part: 75% DCEN balance
1 part: 100-150Hz AC frequency
2 parts: 0.040"-0.050" arc length pointed into 3/16" material, centered
2 parts: 200-225A primary welding current

That's what I would use. Serves 4. Adjust ingredients to your liking

[Keith_J]

16 amps @ 380 volts? Yes, more than enough for 250 amps GTAW from an inverter machine.

[jkhackney]

... yeah I bid on this welder in an online auction without knowing how really BIG it is!

Last summer I was required by our electric utility to install a "smart meter" in my garage and while they were at it, I asked them to install a 380V 3-phase plug, just in case I might want some big toys !

Back to welding the car, I don't have all the settings that Oscar listed, just a square wave of fixed frequency and the ability to linearly ramp the application of full current. I can alter AC from 40-60% EP and I can use DC EP.

I got some spot welds on 1/16" aluminum that I was happy with using a #4 cup, 3/32" pure tungsten, 160 A, DC EP, no hole and no filler rod. Whenever I drilled a hole, it just melted away from the tungsten and made a mess.

When the welds break they leave a good "nugget" of metal behind, but they're not super strong, so I planned on welds every 1.5" or so, which was the interval on the original spot welds.

Unfortunately, the welds warped the pieces just enough that I couldn't continue pressing them together close enough for further welds to "take". There was always a tiny sliver of air in the joint. If I had a really big clamp, it would have worked. But with just my body weight pressing down on the car's floor, it wasn't enough.

I had to use rivets to finish the job. But about 1/6th of the job is welded, some of it pretty nicely. In this photo, the horizontal support at the bottom is welded to the floor sheet (with some rivets to strengthen some of them) and the other supports are rivetted:





Thanks for your help! I'll stick around the forum for sure, as I learn to TIG. There are still a lot of cracks to repair on the aluminum body of this car.



-Jeremy

[noddybrian]

Well it worked out better than I expected - 5251 of 1950ish vintage is next to impossible to clean & weld as it's more corrosion than metal normally so congrats ! most people use rivets for structural & fiber fill for anything else

[AndersK]

Indeed good work

If you skip welding and use rivet it you could also add some body glue to add strength.

I've used Bond & Seal from Wurth to glue aluminum panels.
This inner structure is only glued to the panel.

IMG_20170114_51247.jpg (30.04 KiB) Viewed 2221 times