Brazing or welding thick copper?

Make that a 75 CuFt acetylene tank... Only reason I mention this is that I'm not sure I can use a #3 or higher tip with that small of a tank.

Interesting question. I will be following this one because it seems impossible. My vision shows forging end cap T's onto the copper.

It might be possible with the TIG if you can insulate and preheat enough but copper is such a good conductor of heat that it wouldn't be easy.

As for brazing it's definitely possible if you use the right filler rod. If this is really copper (and not brass or bronze) then I'd recommend using Sil-fos, Sil-flux or any of the other trade names for the 90-95% copper, 5-10%silver, and trace amounts of phosphorus for fluxing rods. Refrigeration contractors use this rod for joining copper tubing and it's great for that job. No flux is needed do to the phosphorus in the rod.

Wayne Cook Shamrock, TX

Hello Wayne,

Thanks for the suggestions. I really am using 100% copper (Alloy 110).

I already figured out the insulating bit when I made the hinges. When I was welding the hinges on the metal table, I couldn't get a puddle. When I put the hinge in a little cave of fire bricks, I was able to get a puddle.

I can try welding two 1' sections of the bar. However, considering that copper seems to just suck up the heat, I'm not sure it would be a good test considering that I'll actually be joining a 4' length to a 6' length.

I like the end cap idea, but not sure how I would fabricate those. I can always screw on gussetts to the bottom since there will be a copper panel on the lower portion of the gate that would hide the gussets. However, gussets on top would be ugly.

When/if I finish this, is this the kind of thing people would like to see in the drop box?

Copper melts at about 1984F, somewhat lower than steel. The problem with copper is it's high thermal conductivity, which results in the whole mass sucking heat away from the weld and conveying it by radiation and convection to the environment.

Perhaps if you wrapped the bar in ceramic blanket, just leaving a narrow region open for access to the actual weld site, you could reduce the heat loss enough to get a good puddle going. It might take a while for the TIG to heat up the thermal mass of the whole bar, but it'll eventually get there if the blanket reduces the heatloss rate to less than the heat output of the welder. The blanket would also shield you from a large bar glowing within. If you run into duty cycle limitation on the welder, you could preheat with O/A, with blanket in place to limit heatloss rate. Might take awhile, but a

75 cu ft acetylene tank has a fairly long duty cycle...

but I figure that 4400 watts (185 amps x 24 volts) will heat up a .5" x 1" x 10" copper bar to near melting in about 64 seconds if there is no heat loss. The blanket won't eliminate heat loss, but it may reduce it enough to make it almost like welding unblanketed steel. I used 10", figuring 5" of shielded region on either side of an exposed

1/2" to 1" wide weld zone about the same as 1" of bare steel since copper is 5X as conductive and assuming metal beyond that distance to not be much of a factor. . Here is a source for small qty of Durablanket S at reasonable cost:

I'm no expert, but there was once a time when I had to weld copper. It was 3/4" thick, 4" wide, and a little over 2 feet long. No way could I weld it with a 300 amp tig torch. I called the welder who usually did our fancy stuff and he said a good pre-heat would do it. So, using fire bricks I was able to fill in the 1/2" deep 1" wide goof in the part using copper wire as filler. It seems to me that after heating the bar I covered it with fire brick. I for rested my hand on it while welding. Even with the preheat I think it took at least 250 amps. And I may have been using helium. We used helium for some aluminum jobs and ran the torch DC. The metal had to be real clean but boy, could you really lay in the rod. Eric

Piece of cake if you braze it. Damned hard if you weld it, because of the conductivity.

Are you going to patinate the copper afterwards?

I'll second Wayne's recommendation of phos/copper brazing. It's relatively inexpensive and a pretty good color match for pure copper. I've brazed heavier (though not as long) sections of copper bus bar than what you're looking at using oxy-propane.

Ned Simmons

Hi Aaron,

Andy is right. Brazing (not brazed) is the best that you can use. I suggest you use AWS - Bag-1 Silver Brazing Alloy CD 45. it would cost $1 in USA.

For more info on How to do Heating and Brazing of Copper with CD 45

Austri Basinillo (Philippines)

Most of your comment I generally agree with. But, I think you will need to blanket much more of the part, out to several FEET on each side. The thermal conductivity of copper is amazing, and a heat sink several feet away can prevent you from heating a part to the temperature you need. I have some experience with soldering and brazing large copper parts, and it is always amazing how much heat comes out the far end of the part!

Jon

Do you have a rosebud tip for your oxyacetylene outfit? It generally works much better that a regular welding tip for preheating. Even if you gas braze, you could probably preheat with the rosebud, quickly swap tips, and fire up the actual torch tip. The rosebud that I am familiar with is an entire section like a cutting torch attachment and not just a tip, although they may make a tip attachment also.

Seems a little odd to me that nobody has asked why you are doing this. I'm not a history buff, but I'm sure the bronze age was a long time ago and the copper age was earlier.

Today copper is a lot less available than other metals like steel. Why not ask how to coat some steel item with a thick coat of copper? Maybe the stupid waste of materials and disregard for strength is the point of your work.

Sorry. Just my opinion after reading what you are up to vs what I would be inclined to do.

Other than that -- traditionally copper gets soldered. I guess the whole point here is excess, though. I don't really mean that in a bad way, just a comment on what you are up to.

Xray, you don't have anything constructive to say, so why bother posting? Why is it material I use to make a gate? The customer likes they way aged copper looks. And if you bothered to google for copper gates, you would find that there are plenty of gates made out of copper.

By the way, the total cost of copper was $400. I just finished a wooden gate that cost the same amount in materials. Do you go and complain about people making furniture out of walnut? Or do you believe everything should be made out of MDF? And as to coating steel with copper, do you have any great ideas on how to do that in a small shop?

Well, it's usenet, just killfile the people who aren't worth reading. There'll always be more, but no reason to not keep the signal:noise ratio as good as you can.

Your 185 amp TIG might be a bit light for the task at hand. But don't get in a hurry with preheat. Even insulated, it'll "suck up the heat" until you have all of the copper within some distance hot -- which will be never if it isn't insulated.

Bulk thermal conductivity of copper is 5X that of steel. Resistance is resistivity * length / area so a bar of given length and cross section will have 1/25th the thermal resistance of a same-size bar of steel. (I missed that first go, Jon). That suggests (as Jon did) that you'll need to insulate for some length, perhaps the whole bar. If sufficiently insulated, then the conductivity doesn't matter. With heat input, the bar will continue to rise in temperature until heat loss rate = heat input rate. If you insulate for enough length, then *once preheated* it shouldn't be appreciably different than welding a same-size bar of steel -- perhaps easier, given the lower melting temp. Once adjacent copper is hot, its ability to "suck heat" is limited. However, it may take a while to pour enough heat in to get the part of the bar inside the blanket hot enough to proceed with welding.

The foregoing assumes that the insulation contains heat considerably better (25 times better) than a bar exposed to free air. I don't know if that's a reasonable assumption for an inch or two of ceramic blanket, but it might be worth a try. As said before, I'd preheat with O/A or oxypropane (lower temp but more heat) so as not to run into the dutycycle limit of your welder.

By all means, please do!

True enough. My bad.

I agree also. Brazing is the way to go.

B.H.

I brazed some 0.5 inch round rod, used my oxyacetylene torch. since I did not have a regular rosebud, I used the cutting tip. It worked. Even a number 4 welding tip was too small. Copper just sucks the heat away.

By the way, I often TIG weld 0.040 inch copper sheet, usually set the TIG welder at 90 amps, so it would take lots of current, and probably preheat, to weld 0.5 inch thick copper.

I like the suggestions for preheat and insulate, but I have no experience with that approach.

richard

woodworker88 wrote:

This has been bothering me since we started this discussion. I have this image in my head of Aaron K. getting this thing half built, and it is going slower and slower as the pieces are assembled. Finally he gets to a point where he's going to need every Oxy/Acetylene torch in the entire county all at once to do the preheating! The only other alternative is to do the entire job while sitting inside a furnace! That's what furnace brazing is for.

Maybe I'm overstating the heat loss problems here, I don't have a clear idea how many cross members there are. If this thing is a square frame with vertical bars, and no other members, maybe it can be built, but it may not support its own weight. If this is for a person to pass through, that may not be much of a concern. If it is to span a driveway, I really have doubts. Either way, if it has a few more horizontal bars of the .5 x 1" size across the middle of the gate, it is going to get messy as the joints are brazed and the number of heat paths increase.

Jon