Connecting (soldering?) aluminium sheet and pipe 1

[drodrig]

Hi there,

We are designing a cooling shell (ethanol, at -25ºC / -13ºF). It consists of a pipe 10/8mm and an aluminium sheet (as thin as possible: 1 or 2 mm)

Here some pictures of the current design:

The sheet is ca. 800mm / 30inches (maximum length)

The pipe will (probably) be AW 6082, CNC bent, the sheet we can decide.

How can we connect them to have a good thermal transfer?

The problem with soldering is that the sheet will deform because of the temperature

Glue or thermal paste don't have good thermal conductivity

Maybe some clamps.

My favourite option would be soldering, adding material for a good contact. Is there any option here?

thanks

 

[EdStainless]

There are glues that are loaded for improved heat transfer, that is as good as you will get. Keep the bond lines as thin as possible and go with it.

https://www.masterbond.com/properties/thermally-conductive-epoxy-adhesives

https://www.henkel-adhesives.com/us...-material/thermally-conductive-adhesives.html

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Compositepro]

This is often done using furnace brazing. A brazing paste is applied to fixtured parts and the whole assembly is evenly heated in a furnace. An inert atmosphere will keep the parts clean.

 

[EdStainless]

The catch here is that Al is difficult to braze other than in vacuum.
With Cu this would be a piece of cake.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[drodrig]

EdStainless,
These glues could be an option, but they have poor thermal conductivity.

Compositepro,
The furnace brazing seems promising, I'll dig into that

Unfortunately, we need to use aluminium for this project

thanks!

 

[EdStainless]

Those glues will be a 1000x better than a mechanical fastening system, and if you are careful and keep the fit tight so that the bond line is only a few thousandths of an inch they will add an insignificant amount of resistance to heat transfer.
It would help greatly if you could slightly flatten the tube so that it had more contact area against the plate. We used to use square tube for projects like this.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[mfgenggear]

@drodig
Was your question solved?

 

[drawoh]

drodrig,

Adhesive has poor thermal conductivity specifications. If the adhesive has a large cross-sectional area and it is thin, it may still be very conductive.

Is there any way to use square tubing?

--
JHG

 

[drodrig]

Hi there,

I'm talking to the company to see the possibilities for square section tubes. They want a lot of money for the bending tools.

cheers,

 

[EdStainless]

Use round tube and pull it between a pair of rolls to flatten it some (the shape is called 'double D'). In Cu square and rectangular tube is fairly common.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[gtaw]

Dip brazing is your best chance of success if both the sheet and pipe are aluminum alloys. The aluminum braze filler metal melts at about 50 degrees F lower than the melting point of many aluminum alloys. With such a small temperature window, it is difficult to precisely control the temperature using methods other than dip brazing.

Best regards - Al

 

[drodrig]

I found a company (in France - I'm based in Germany) which can do it (after some negative answers)

They can do it with vacuum. It's now know how much the plate will deform, so they'll run a first test.

EdStainless,
I'll ask them about the double D option. It sounds good.

gtaw,
If this vacuum doesn't work I'll check the dip brazing

thanks