aluminium manifolds - heat sink plate material...

[FASTdan]

Just after a bit of advice on heat sink materials.

I manufacture manifolds similar to the ones shown in this thread:

http://www.mig-welding.co.uk/forum/showthread.php?t=23592

However mine are aluminium. The baseplates are 10mm thick, the tube 3mm thick either 45mm dia or 51mm dia.

I run a 200A AC inverter machine. To achieve the fillet to the plate I use a 3.2 tungsten (was using white tip, now trying grey).

To minimise distortion I fasten the assembly down to an aluminium plate using the actual manifold fastener holes. I find this works well and means the workpiece is very easy to manipulate without lots of large clamps etc. So I have a couple of different plates (15mm and 25mm thick - just what I had lying around) with lots of holes drilled for the variations in manifolds.

I do however find that I am almost continually running at 200A (so the limit of my machine) to do the fillets. Does 200A seem about right for this type of work? I wondered if I would be better or worse off changing the material of the heat sink plates? Would I get heat into the job quicker if I used steel plates and thus achieve a more consistent weld? - the initial welds are hard work (slow moving, poorer appearance) until the workpiece has heated. I am not preheating purely due to lack of facility.

I guess the obvious answer is to address the pre-heat issue.....

My products are (I would like to think) of reasonable quality and appearance, but I am always looking to improve. Any advice would be appreciated

 

[Paul.]

I'm not surprised your having problems, clamping the job to a 15 or 25mm ally plate makes a hell of a lump of metal to heat up and keep hot, mild steel would be better, and even better still stainless steel, you could drop down in thickness to probably 10mm or so, or even use 6mm and reinforce the backside with ribs to prevent distortion, this will allow you to drop your amps and use a 2.4 tungsten,
I used to make aluminum heat exchangers which involved welding tubes into header plates, these headers were 600mm Plus long and could easily end up with a 15mm bow in the middle if not handled properly, the assembly fixture was a lump of 100 x 100 x 6 stainless box section drilled and tapped to match holes in the header just as you are doing now,

 

[FASTdan]

hmmm yes just as I thought lol. Thanks for the advice.

its one of those things, I started doing the odd manifold - and my set up sufficed, but then I've just persevered as my volumes increase without giving it much thought (partly laziness!). I should have addressed this about a year ago!

best get myself some Mild steel bar then!

 

[PhillipM]

Clamp it down on a couple of old gaskets - will slow the heat transfer to the jig plate massively.