Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Chill block, warpage... TIG welding thin aluminum questions

126 views
Skip to first unread message

Stuart Wheaton

unread,
Jan 9, 2010, 9:36:10 AM1/9/10
to

Monday I will be starting a project which involves tigging very long
outside corner seams in 1/16th aluminum sheet. The seam is set up with
just the inner edges touching. Using 1/16" lanth. tungsten with 1/16"
4043 wire filler. I have a gas lens. I plan to tack every 4-6 inches,
then I will stitch every 3rd or fourth segment to try to keep the
piece from getting too hot. I need to keep warpage to a minimum. My 2'
test sample got rather warm, but I was using too big a tungsten, (1/16"
will be delivered Monday) and the amount of metal (width of sheets) was
less than the actual plates, but some of these seams may be 8-10 feet in
total length.

If anybody has words of wisdom, I'm eager to learn from your mistakes so
I don't need to learn from mine! I think my employer prefers that way too!

I have seen mentions of using a copper chill block to help avoid blow
through, how big a block of copper is needed to work effectively?

Thanks,

Stuart

dca...@krl.org

unread,
Jan 10, 2010, 10:00:10 PM1/10/10
to
On Jan 9, 2:36 pm, Stuart Wheaton <sdwhea...@fuse.net> wrote:


>
> I have seen mentions of using a copper chill block to help avoid blow
> through, how big a block of copper is needed to work effectively?
>
> Thanks,
>
> Stuart

The copper block does not have to be very big, but it adsorbs heat.
So the bigger it is, the longer before you have to let it cool down.
I would use something at least an eighth inch thick, but would prefer
something 1/2 inch thick. For long seams, you might consider kludging
up something with circulating water.

Dan

Martin H. Eastburn

unread,
Jan 11, 2010, 9:51:36 PM1/11/10
to
I would not re-design something unless it is third-world design.

Aluminum is just fine if greased correctly and the Al is larger
by a percentage than the copper.

The metal tab and mounting header is a copper heat spreader
made for impulse heating. If you spread the heat across a surface
then it remains active.

Simple air between - a.k.a. mechanical attachment without grease
is not a good idea at all. If not greased, buy some at Radio Shack
or DigiKey. Use it between the SCR/Triac and the insulator? and there
to the large heat sink.

Martin

Stuart Wheaton

unread,
Jan 11, 2010, 10:30:35 PM1/11/10
to
Martin H. Eastburn wrote:
> I would not re-design something unless it is third-world design.
>

Um... no, it was designed right here in Cincinnati...Pretty simple
though, a few pieces of curved aluminum plate, makes a 3D shape when
welded together...I guess you could design it in a third world
country... Besides, I'm paid to weld, and the design I got is the
design they want. Do you routinely redesign the things you weld?


> Aluminum is just fine if greased correctly and the Al is larger
> by a percentage than the copper.

You routinely weld greased aluminum? How does that work out for you?
Do you use a silcone type grease to avoid trapping hydrogen in the
molten aluminum?

dan

unread,
Jan 12, 2010, 6:07:05 PM1/12/10
to
What's that Lassie? You say that Martin H. Eastburn fell down the old
sci.engr.joining.welding mine and will die if we don't mount a rescue
by Mon, 11 Jan 2010 20:51:36 -0600:

>
>I would not re-design something unless it is third-world design.
>
>Aluminum is just fine if greased correctly and the Al is larger
>by a percentage than the copper.
>
>The metal tab and mounting header is a copper heat spreader
>made for impulse heating. If you spread the heat across a surface
>then it remains active.
>
>Simple air between - a.k.a. mechanical attachment without grease
>is not a good idea at all. If not greased, buy some at Radio Shack
>or DigiKey. Use it between the SCR/Triac and the insulator? and there
>to the large heat sink.
>
>Martin

I think you got the wrong post. This thread is about welding aluminum
sheet, and controlling the distortion with chill blocks,
not how to properly attach semiconductors to heat sinks.
--

Dan H.
northshore MA.

Martin H. Eastburn

unread,
Jan 12, 2010, 9:06:23 PM1/12/10
to
NO NO.

Si Grease is used to pass the heat from a semiconductor to a heat sink.
Martin

Stuart Wheaton

unread,
Jan 12, 2010, 10:28:25 PM1/12/10
to
Martin H. Eastburn wrote:
> NO NO.
>
> Si Grease is used to pass the heat from a semiconductor to a heat sink.
> Martin
>

Aluminum is not a semi-conductor...

Did you actually read the thread?

BobH

unread,
Jan 13, 2010, 12:21:59 PM1/13/10
to
Stuart Wheaton wrote:
> Martin H. Eastburn wrote:
>> NO NO.
>>
>> Si Grease is used to pass the heat from a semiconductor to a heat sink.
>> Martin
>>
>
> Aluminum is not a semi-conductor...
>

What I think Martin was referring to is using the silicone heat sink
grease AWAY from the weld area to make the chill block couple to the
aluminum being welded better. He was probably referring to mounting
semiconductors because that is the use that most people are familiar with.

I would be nervous about doing this for three reasons:
First, it would be a requirement to keep it out of the weld area.
Second, it's a major PITA to clean up.
Third, its fairly expensive stuff.

I think that Martin's idea would be worth experimenting with, because a
chill block is not going to carry off much heat if it is not in solid
contact with the aluminum. Clamping chill blocks to large pieces is a
problem.

Regards,
BobH

waynesbane

unread,
Jan 13, 2010, 7:55:21 PM1/13/10
to
On Jan 14, 4:21 am, BobH <Wanderingmetalhead.NOSPAMPLE...@yahoo.com>
wrote:

Given that the OP is worried about blowing through, one could be
forgiven for thinking that the OP does not have the requisite skills
for the job.
16th inch should not pose a problem.

Stuart Wheaton

unread,
Jan 13, 2010, 8:20:39 PM1/13/10
to

As the OP, I'd say that if I felt fully confident in my abilities, I
wouldn't have asked. If you think the only people who should post are
people who don't have questions, it is going to get very dull around here.

My usual stuff is frames MIG welded from 1/8" Al or 16 ga steel and
aluminum 1x box tube. I have occasional call for TIG, but I have not
tried to weld 10 foot seams in 1/16" sheet before. For me, trying to
limit overpenetration in an outside corner is a challenge.

The parts were delayed, I started the job today. So far, not too bad,
Some sections look downright pretty, most are acceptable, a few will
have to get ground down and done again. By stitching, I have been able
to keep the heat down and getting a little oil-canning on the top
surface, but not seeing warpage.

Since you claim to be an expert at 1/16" Al sheet, what are your
suggestions? That is why I asked in the first place.

Martin H. Eastburn

unread,
Jan 13, 2010, 9:43:32 PM1/13/10
to
The SCR /Triac is semiconductor that needs a heat sink.
It was wired to the pcb and mounted on Al to keep it cool.

Martin

BobH

unread,
Jan 13, 2010, 3:01:11 PM1/13/10
to
Martin,
I think you are confusing this thread with the one that Dan posted about
servicing the electronics on a welder. This one is about TIG welding
long outside corners of 0.060 aluminum.

Warming up for a taste of crow,
BobH

BobH

waynesbane

unread,
Jan 13, 2010, 10:39:33 PM1/13/10
to

Skill is a matter of self-development. It cannot be gained by a text
message.


>
> My usual stuff is frames MIG welded from 1/8" Al or 16 ga steel and
> aluminum 1x box tube.  I have occasional call for TIG, but I have not
> tried to weld 10 foot seams in 1/16" sheet before.  For me, trying to
> limit overpenetration in an outside corner is a challenge.

Ok if you want to compare experiences - Ticket for Unlimited Thickness
Structural Steel, 30 years experience with Mig, Tig, OA, Submerged
Arc, Steel, SS. Al, Pressure Vessels, Road making machinery, Yacht
building...


>
> The parts were delayed, I started the job today.  So far, not too bad,
> Some sections look downright pretty, most are acceptable, a few will
> have to get ground down and done again.  By stitching, I have been able
> to keep the heat down and getting a little oil-canning on the top
> surface, but not seeing warpage.
>
> Since you claim to be an expert at 1/16" Al sheet, what are your
> suggestions?  That is why I asked in the first place.

From what you have said it seems that you are not doing to bad. The
fact that you can weld it at all means you have reasonable skill to
start with so I will give you the same advice that my welding
instructor gave me 30+ years ago - PRACTICE. One of the tasks he set
for us was to tig coke cans together bottom to bottom. Yes it is
possible, mind you it is best if they are empty first.

Take your time, relax and keep your eye on the puddle. Don't be afraid
to play with your settings.

Ernie Leimkuhler

unread,
Jan 14, 2010, 3:26:40 AM1/14/10
to
In article <ce01e$4b48945a$d06602ac$23...@FUSE.NET>, Stuart Wheaton
<sdwh...@fuse.net> wrote:

Be very careful in ever using copper chill blocks on aluminum sheet.
It is the same problem with TIG welding copper on an aluminum welding
table.
The copper and aluminum can fuse very easily creating a spot of
aluminum bronze on both parts.

You will want to use a steel backing block with a very small hammer to
tap the edges together after tacking.

You will want tack welds every 2 inches before welding.
It's a shame you couldn't have a flanged edge on both pieces.
Makes it so much easier.

You will find a pulser makes this process so much easier.

Tungsten size doesn't really matter that much.
Your amperage will be very low.
1/16" = 0.0625" = 63 amps, outside fillet weld means -1/3 = 42 amps.

Stuart Wheaton

unread,
Jan 15, 2010, 7:22:18 PM1/15/10
to
Ernie Leimkuhler wrote:
> In article <ce01e$4b48945a$d06602ac$23...@FUSE.NET>, Stuart Wheaton
> <sdwh...@fuse.net> wrote:
>
>> Monday I will be starting a project which involves tigging very long
>> outside corner seams in 1/16th aluminum sheet. The seam is set up with
>> just the inner edges touching. Using 1/16" lanth. tungsten with 1/16"
>> 4043 wire filler. I have a gas lens. I plan to tack every 4-6 inches,
>> then I will stitch every 3rd or fourth segment to try to keep the
>> piece from getting too hot. I need to keep warpage to a minimum. My 2'
>> test sample got rather warm, but I was using too big a tungsten, (1/16"
>> will be delivered Monday) and the amount of metal (width of sheets) was
>> less than the actual plates, but some of these seams may be 8-10 feet in
>> total length.
>>
>> If anybody has words of wisdom, I'm eager to learn from your mistakes so
>> I don't need to learn from mine! I think my employer prefers that way too!
>>
>> I have seen mentions of using a copper chill block to help avoid blow
>> through, how big a block of copper is needed to work effectively?
>>
>> Thanks,
>>
>> Stuart
>
> Be very careful in ever using copper chill blocks on aluminum sheet.
> It is the same problem with TIG welding copper on an aluminum welding
> table.
> The copper and aluminum can fuse very easily creating a spot of
> aluminum bronze on both parts.

Oooh! Thank you, I'm glad I didn't try to build a chiller Actually,
the aluminum is cooling faster than I expected. This is just so
different from the stuff I've been doing for the last 20 years. The
chance to do this came up, and we were slow, so I'm glad I bought the
Syncrowave and taught myself TIG!

>
> You will want to use a steel backing block with a very small hammer to
> tap the edges together after tacking.

Yep, and I finally designed and fabricated a decent clamp that actually
holds the parts in the proper place for tacking, and accommodates all
the differences between the sections.

>
> You will want tack welds every 2 inches before welding.

I've been doing about 4" spacing, I'll try tightening it up. I need to
work on making the tacks disappear under the welds.

> It's a shame you couldn't have a flanged edge on both pieces.
> Makes it so much easier.

Yeah, remember when you used to work for other people? Asking the guy
who will actually weld it about his opinion on the design might cause
trouble, better to get it cut quickly, before anybody thinks too hard!
:-)

I have suggested that if we ever do this again, we should ask for flanges.

>
> You will find a pulser makes this process so much easier.

Got a spare PC 300 under your desk? Send paypal invoice!

>
> Tungsten size doesn't really matter that much.
> Your amperage will be very low.
> 1/16" = 0.0625" = 63 amps, outside fillet weld means -1/3 = 42 amps.

Yep, I was using the Ernie Rule. I found that shading the arc balance
toward penetration made a positive difference. The later ones are
looking very good. I expect to have a lot of re-work on the first one
once it gets cleaned up, but the others are pleasing me. Maybe the
first one is only surface ugly.

I'm getting real good at picking up my pedal between my feet as I scoot
my chair between welds.

Thank you for your help. Ernie posts are always worth reading! Did you
ever get that text done?

Stuart

Ernie Leimkuhler

unread,
Jan 23, 2010, 4:51:39 AM1/23/10
to
In article <8d270$4b5106ba$d06602ac$32...@FUSE.NET>, Stuart Wheaton
<sdwh...@fuse.net> wrote:

You mean the long rumored, yet never completed, TIG welding bible?

Not with my current work schedule.

0 new messages