Aluminum Radiator Design 1

[valmeidan]

I have some tubing I want to insert into some mechanical aluminum tubing,cap the ends and add barbs to create a radiator which will hold glycol/water mix for cooling. Here are a couple questions.

1. I am using 6061 Aluminum,is this the best aluminum grade to use in this case?

2. Is it difficult to punch the mechanical tubing to hold the tubing in place in the orientation you see? Normally you would see the tubing rotated 90 degrees. 1" outer diameter, with about 1.5-2mm wall thickness..

3. Would it be better to solder or braze or TIG, taking into consideration the design, durability, and clean joints will as little flux or other material that could contaminate the fluid or aid in corrosion etc?

 

[MikeHalloran]

1. Yes, it's what you have. If you are making trainloads of product, other choices might be better.
2. Yes. You may have to mill the slots; punching a tube with any precision (the kind required for brazing) is difficult to impossible. You also need an assembly fixture to hold all the parts in the right position during bonding, however that happens.
3. It would be better to have _far_ fewer joints. Look at transmission coolers and especially power steering coolers.



Mike Halloran
Pembroke Pines, FL, USA

 

[tbuelna]

Ideally you would flare the holes in the header tubes. This will provide a much stronger joint. If you are making more than a prototype quantity of cores, you can have the header tubes hydroformed to add the flares without too much cost. If you are only making a small number of cores, you can also make the header tubes with flared holes using a rapid prototype process like DLMS.

The best method for joining the tubes to the headers would probably be furnace brazing. The braze joint is conductive and the process does not create residual stress in the core like TIG welding would. You could also use an adhesive bond, and this is commonly used on automotive radiators, but it can be more difficult to get right than furnace brazing. 6061 aluminum will work for a water/glycol heat exchanger core, and 5052 aluminum is also a good choice.

 

[MikeHalloran]

Speaking of flared joints, there are tools sold into the refrigeration industry that can pierce and pull a tee branch from something like your header tube, in copper. The branch is round, so you would have to flatten the cross tubes but leave the ends round. I'd say the tool has a fair chance of working in 5052-O or 5052-H2. No way would 6061 tolerate the necessary elongation.

Note also that if you are relying on airflow normal to the major plane of your radiator, the cross tubes are facing the wrong way and there are too few of them.





Mike Halloran
Pembroke Pines, FL, USA

 

[ornerynorsk]

Piercing tube as shown in your illustration can be done very precisely and economically.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.