Oxygen-Hydrogen Gas Welding

[Bill Tromblay]

Hi,

What grade is the parent metal? This is very important to developing a good weld. With new 1100 or 3003, wire brushed and cleaned with alcohol, no flux is needed on the base material, only the filler rod. With 5052, 6061, old material or mystery aluminum, flux is needed on the material and filler rod, wire brushed and whipped down with alcohol.

Hydrogen doesn't get you out of clean up work. If you can't brake the surface tension of the material or appears scabbed over like an old green bean casserole, more than likely you are oxidizing the surface with an oxidizing flame instead of a neutral flame. With a dark back ground, you should barely see a dull blue cone. If the cone is defined in shape or a brighter blue, it is oxidizing.

A side note. I'm trying to decide if it is worth dragging my oxy/hydrogen set up to Jim Hurys for the metal meet. Cliff, if you are not going to be there, I will leave it home.

B

[cliffrod]

Hi,

What grade is the parent metal? This is very important to developing a good weld. With new 1100 or 3003, wire brushed and cleaned with alcohol, no flux is needed on the base material, only the filler rod. With 5052, 6061, old material or mystery aluminum, flux is needed on the material and filler rod, wire brushed and whipped down with alcohol.

Hydrogen doesn't get you out of clean up work. If you can't brake the surface tension of the material or appears scabbed over like an old green bean casserole, more than likely you are oxidizing the surface with an oxidizing flame instead of a neutral flame. With a dark back ground, you should barely see a dull blue cone. If the cone is defined in shape or a brighter blue, it is oxidizing.

A side note. I'm trying to decide if it is worth dragging my oxy/hydrogen set up to Jim Hurys for the metal meet. Cliff, if you are not going to be there, I will leave it home.

B

Keep the welding gear at home this time, Bill. I haven't called Jim yet but I doubt I'm going this time. Eldercare obligations are the new reality here, so everything else is basically arranged around that.

As far as equipment and material, this sheet is supposed to be 3003 but may be 5000 series. It came from the shop where Will got his waterjet work done. It was the end of the previous order and they weren't certain which it was. I have been scrubbing it clean and was getting better results with a full flux of both base metal pieces and rod, so it's probably 5000. I have been mixing flux with alcohol (per one of Kent's suggestions in the past) but not wiping down material.

The whole process/problems seem very much related to flux/dirty issues, but I know this flux is barely worth using. I need to be more diligent, but finding the time to get into the shop around my mother's needs is a challenge right now. Thanks for the advice. I'll keep improving.

[Bill Tromblay]

Hi,

Contrary to Kent, I don't mix alcohol in my flux... Ever....

My reasoning... When heated, you exceed the flash point of the alcohol and it begins to burn. While burning it is releasing hydro carbons and other chemicals right into your weld pool... Use water, as recommended by the chemist at the flux manufacture.

Because 5000 series aluminum is corrosion resistant, flux is needed on the base material to remove the oxide layer that forms during the weld process. Even with new flux, this grade material is challenging.

Old flux, in 1100 and 3003 will work ok. A lot of old timers, including Kent, aren't quick to changing out old expensive flux for new, even more expensive flux.

Thanks for the note on Jim's event. I understand family commitments and the challenges you face. Wishing you well.

B

[cliffrod]

as always, you're way ahead of me Bill. Thank you. And thank you for the family considerations. This isn't an accidental situation but does take priority over most things.

it isn't just being cheap. This tub of flux is really old- i.e has self hydrated while screwed tightly shut over the past many years. No need for a spray bottle. or bottled water. or even much alcohol. Normal hygroscopic action has done it all for me, with plenty of standing liquid on top to spare. I've got a new sheet of 3003 that's now available locally but already have shaped parts from the 5000 sheet to weld, so need to figure it out. So your specific information is much appreciated.

[Bill Tromblay]

Hi,

The 5000 series will have a structurally fine weld, but will be unsightly. Roger on the flux..

B

[joeswamp]

Thanks to everyone who's contributed to this thread, especially Cliff who is documenting stuff that I've never seen written down before. Two questions that maybe you guys can answer:

1) How exactly do you set the flame? In the excellent redbackaviation link (which is the only place I've ever seen this discussed), they say that:

• "you’re looking for a continuous, uniform flame with a controlled circle as it strikes the firebrick."
• "too much hydrogen, and the flame will be washed out."
• "Too much oxygen, and the flame will be pinpointed"

Is this as hard as it sounds? It seems like there's not enough info to judge a neutral flame, i.e. how do you know when it's a "controlled circle"? And how small of a circle is a "pinpoint"? Just seems a lot more difficult than in oxyacetylene, where you have a solid criteria: when the two inner cones line up, the flame is neutral.

2) How do you know how far to hold the torch from the workpiece? In oxyacetylene, you have all this control because you can see the inner cone, and by holding the inner cone a fixed distance from the work you get very uniform results. Do you just try to hold the torch a fixed distance from the work? It seems this would be much harder, because you don't have a reference point that's close to the work. I'm admittedly a poor welder, but I watch that inner cone a lot.

Sorry if these are dumb questions, but I've always wanted to try oxyhydrogen and these are two things I've always wondered about.

[cliffrod]

From the novice perspective, I'm still learning about the proper flame adjustment. When I say the flame is hard to see, imagine looking at a shard of matte non-shiny glass or similar plastic. You can see it but it's transparent. the blue or orange color in the flame is very faint. If you adjust the torch, the orange and blue details change. I also read that adjusting the flame "a little hotter" produced a desired result with the same tip. Performance does seem to vary, but i probably lost benefit of a neutral flame. with better info from Bill, I'll adjust my flame differently.

Distance? with external lighting to illuminate the puddle area, you will see the same behavior as the aluminum heats that is visible with oxy/act. It just happens more slowly, so is easier to manage. The area will heat, typically with a fine border line around the HAZ that travels away from the specific puddle area. The aluminum surface then changes a little more, roughness appears, as the puddle begins to form. Then it starts to wiggle, then shimmers as it becomes a fluid puddle. Then it falls away. But it's all viewable because there's no visible flame to distract you and it all happens much slower than with oxy/act. All this time, the torch is fairly close- closer to 1" than 4". I don't think about it much because I hold it where it produces the results I want. The angle of the torch is more important- vertical for max heat fast to horizontal/parallel to seam for no heat. Because the flame is cooler, alignment of parts and equitable direction of the flame at both edges (not prioritizing one edge)is really important. The flame needs travel directly along the seam, not angled the the right or left of the line of the seam. If one edge gets the flame and the other doesn't, its a fast hole. With a cooler flame, there isn't a large enough HAZ to melt the distant edge and have the two separate puddles accidentally join like it can with oxy/act.

Like I said, I'm learning. NOT an expert. But I would highly recommend anyone thinking about trying or pursuing oxy/hyd to spend some time welding aluminum with oxy/act first. Even though I haven't done tons of aluminum with oxy/act, it was a big help moving towards oxy/hyd. Between the cooler flame, less obvious flame adjustments, more critical psi, fitment requirements, etc, oxy/hyd is like an exaggeration of all good & bad related to oxy/act. I wouldn't be able to notice many of the unique oxy/hyd details if they hadn't been part of the oxy/act learning curve. Before spending $$$ to get an oxy/hyd rig, get and use a plain cheap & much more versatile oxy/act rig. It's the foundation.


I said it many times before- Choose your Masters wisely. Finding info and experts that you can trust (and I consider Bill to be an expert to be trusted) is tough. One thing about Bill that matters to me is he's not trying to sell a proprietary product or video at every step along the way. The same with Peter. Even though he had great & proven products, Peter doesn't follow every recommendation with "by using this HandBuilt" hammer or whatever. I don't think I'm a moron and don't need to listen to an endless commercial from someone who apparently thinks I can't look & see what he has for sale. There is also a lot of contradictory info. IRRC, the RedAviation link recommends setting line pressures far higher than the pressures that produced acceptable results for Bill. even though there's some good info posted, I don't know (or necessarily trust) RedAviation. I do trust Bill. I hope posting all the ups & downs of what I'm learning about oxy/hyd might help others sort out information as good or bad for themselves.

[Bill Tromblay]

Hi,

Thank you Cliff for the kind words. I too, am always learning. I'm now getting consistent results and have high confidence in my welding.

I have tried the "fire brick trick" and I didn't find it conclusive. When you bring the flame to the brick, regardless of the fuel to Oxy mix ratio I saw the same orange "washed out" appearance. I tried it several times and never saw the same out come as redback aviation did. It would be great to visit with them in person to see their outcome.

Over the last 3 years, I have purchased over 15 books on aluminum gas welding and have meet with 6 of the 11 known oxy/hydrogen gas welders in the United States. (Note: More than likely, there are more than 11 oxy/hydrogen gas welders in the U.S., but these 11 welder only know of us eleven.) Cliff is correct and I have stole his phrase many times, "Choose your Masters wisely!" Because so few people do it, it tends to be, not an exact science. With so few people doing it, skill levels very, quality very and out come depends on the care and attention to detail of the operator. (Note: there is some bad advice out there)

How I set the flame...

My last book purchase was from Kaiser Aluminum publish date 1967. In the Oxy/Hydrogen section, I found this statement which I have never seen before. "In larger factory environments, central located bottles with regulators are installed and individual flow meters are used at each workstation" also " 3 to 5 parts Hydrogen to 1 part oxygen to obtain a neutral flame".

This got me thinking and with a simple Amazon purchase I was up and running with a consistent result. There are 4 settings that we can do with any Oxy/Fuel torch system.
1, Pressure in Psi set at the regulator. I use 1psi oxy, 1psi hydrogen as outlined by all of the manuals I have.
2, Tip Size. I use the tip sizes as publish in all of the manuals I have.
3, Volume. How much you open the torch valves.
4, Mix ratio. How much you open each valve to set the fuel to oxygen ratio.

On my gas saver stand, I mounted two "Oxygen Approved" flow meters that I use to set the Volume and mix ratio.



I set the mix ratio by opening and closing the valves of the torch to maintain the ratio. I continue to open both valves until the volume is enough to melt the aluminum in 4-5 seconds. With the flow meters, I tend to not look at the flame anymore, I set it and forget it.

When I shared my idea with some of the other welders I got mixed opinions. Most set it by eye and sound and go. But most don't care what the appearance is, because they grind and hammer the weld so it is never seen. For me, in aviation, often the weld is left and not metal finished so appearance is important.

Distance,

If you get too close, A black circle will appear on the surface. If you move the torch away, the circle disappears. The rest is as described by Cliff.

Hope it helps,

B

[cliffrod]

Great stuff, Bill. This is why I say the things I say. Thank you.

[joeswamp]

Thanks for the help guys. I guess I am too reliant on looking at that inner cone, perhaps I am using it as a crutch.

The flowmeters solution for setting the flame seems like the way to go here. I can imagine getting good with the firebrick method, but only after you have a true reference like the flowmeters to understand what's actually going on. I remember discussing this with Kent White several years ago, he told me that when oxyhydrogen welding was ramped up in factories during WWII, a supervisor periodically set the flame for the weldors. They probably used a setup like this.

One piece of info that's not obvious: the numbers and units on those flowmeters (technically "rotameters") are only correct if the flow is exhausting direct to atmosphere (or to some other pressure they were calibrated at). This shouldn't matter to you, because all you care about is that the ratio between the two is correct -- you don't really care what the actual flow is. Also your pressures are so low that the numbers should be very close.

I believe you will get more accurate results at the meters if both oxygen and hydrogen have the same restriction downstream, i.e. if you set the torch knobs to the same number of turns, and you adjust the flow at the regulator rather than at the torch. But this is a detail. Thanks again for sharing your setup, solves a mystery for me!

[Bill Tromblay]

Hi,

So I understand correctly. We went from being accurate by looking at the flame, to not being accurate enough with the use of flow meters? Why I hate forums...

[joeswamp]

So sorry about introducing confusion. Your setup is great, it's almost certainly the best way to establish the "neutralness" of the flame.

I just mentioned that detail about rotameters because I got burnt by them in the past. At work I once inherited a contraption made by someone else, and he had incorporated rotameters in multiple places on his device. In the middle, he was measuring flow at high pressures (like 50-100psi), and I spent a couple of days troubleshooting this setup because the sum of the flows at the end were wildly different from the sum of the flows in the middle of the machine. It turns out that these things read differently depending on the pressure they exhaust to. You can in fact buy them in different calibrations for different exhaust pressures.

In your implementation, this issue does not matter because:

1) You are running at very low pressures, so the error introduced is tiny (probably on the order of a couple percent).
2) It doesn't matter what the numbers say, you only care about the ratio between the two flowmeters.

If you want to be super-duper double extra accurate, you should set the knobs on the torch the same, and adjust the flame from the regulators. Then your exhaust pressure will be close to the same on both flowmeters. This level of accuracy probably doesn't matter, but it might be worth trying to see if you notice a difference in welding.

[cliffrod]

I've had time to do some practicing.

-No additional flowmeters yet.
-Welding at 1psi (set with the regulators and confirmed with the pressure gauge) I've tried 2 psi Oxygen and 1 psi Hydrogen as mentioned in another thread and 1.5 psi each as my own experiment. results were consistent, regardless of pressure.
-flame adjusted no obvious specific blue cone or excess of Oxygen
-supposedly 5000 series .063 aluminum. prepped to excellent fit-up. then degreased with lacquer thinner. then scrubbed clean with ss brush. then washed with isopropyl alcohol, then wiped clean with new dry disposable paper towel, then immediately fluxed immediately .
-brand new TM flux, mixed with tap water that's stood open for several days to outgas all chlorine. my own built brush, with natural hair and ss construction. base metal and rod are fully fluxed.
-Meco Midget, N3 tip
-mostly coupons measuring at least 1' x 1' (up to 1 1/2') before being split into two pieces to be welded back together. Each coupon is cut, welded, then cut again parallel to weld so I can weld one coupon numerous times.

I done at least 40 ft of welding. I'm having difficulty getting consistent full penetration. The backside isn't properly fused. Seems like I almost have to touch the tip to the puddle to make that happen or produce a very wide puddle on top to begin to get full penetration on backside. I don't have an N4 tip but understand this N3 should be adequate for this application. The bead looks like a typical cold bead if I add filler (new rod or offcut of metal being welded). There's no rushing the process. Unless you're on the outside corners at the edge of the pieces. It takes a long time to warm the work before you produce the puddle. It's so slow it's more likely to cause my discomfort from cramps or have the metal cool prematurely from my dripping sweat.

When I wash/scrub it, it's clean but appears grainy. There are no obvious sugar or inclusions that don't look like aluminum, but the shape/contour is there. if they're includions, they're fully encapsulated. It's not a smooth, shiny bead. Not sure if this is the "not pretty" bead Bill is talking about in reference to 5000 series. My oxy/act welds aren't movie-star pretty, but they didn't & don't have this grainy characteristic. Not really getting a scab like Bill mentions, but the puddle is typically not an open, shiny puddle of molten aluminum. There's only a small area that's shiny.

Whenever I stress test the beads that do have full penetration, these grainy-surface welds prove stronger than the base metal.

Right now my picture posting ability is limited to taking pics with my digital 35mm camera and getting them through computer to put them on here. the old ipad tablet I use will no longer log into the forum. I wanted to get some pics of good beads as well as bad ones, but that's still just a goal...

i haven't tried oxy/hyd welding the 3003 aluminum I have yet to see if results are different.

I expect the problem has as much to do with me as anything about the process details. I'll get some pics. Any ideas or advice?

[cliffrod]

My very smart wife may have solved my pic dilemma.

these images show front and back of the same two coupons. The first two are from an early coupon. The last two are some of the better welds, in terms of how easy they were to produce. But they still aren't going to win any contests...

[Chazza]

... There's no rushing the process. Unless you're on the outside corners at the edge of the pieces. It takes a long time to warm the work before you produce the puddle. It's so slow it's more likely to cause my discomfort from cramps or have the metal cool prematurely from my dripping sweat...

🤣🤣🤣
Two thoughts; the bigger tip would speed things up but maybe too much.

What are you doing with the flame? Keeping it still-ish or weaving a bit? Weaving would create a wider puddle and therefore better penetration,

Cheers Charlie

[cliffrod]

🤣🤣🤣
Two thoughts; the bigger tip would speed things up but maybe too much.

What are you doing with the flame? Keeping it still-ish or weaving a bit? Weaving would create a wider puddle and therefore better penetration,

Cheers Charlie

I don't know, Charlie. supposedly, a #3 tip is adequate for .063 aluminum and others produce successful results with an #3 tip. That points to the technique and the operator... a larger tip may be what I need but I haven't ordered any yet. I've weaved, pushed straight, held tip vertical like I'm tacking, angled forward & backwards (specifically to sink more heat and widen the puddle). iirc Bill posted a pic of his oxy/hyd bead somewhere. It was nice and narrow, not 3/4"-1" wide on one side.

This process reminds me of when someone goes camping, stacks up a bunch of big thick sticks or actual split firewood with no tinder or kindling, then lights one match & throws it in thinking it will turn instantly burst into a roaring bonfire... There's only so many btu's available to make it happen. Got to do it exactly right or it's not going to happen.