Exhaust system: dual or single

[Homemade WRX]

6al-4v is Gr5...yup. What's really impressive about it are how minimal the differences are between forged and cast for component use.

Regardless, vibaration/resonance will be what causes the weld to fail, with my guess. Old Greg's idea for testing actually isn't bad. Now to keep in mind, the exhaust isn't (or shouldn't be) rigid mounted. So a flex section/bellows will allow it to wobble/swing and then is is still suspended by rubber or poly exhaust hangers. I wouldn't worry over cracks is overkill for anything behind the flew section. The exhaust manifold which is rigid mounted to the engine (a large vibrating unit) where the pipes are growing in multiple axis', that is where your concern for ti tubing weld failures should be IMO...see usage of slip joints and bellows.

I plan to play with some 2xxx series alum for the catback. Probably 2024. It costs was less than ti and does reasonably well with heat.

[old greg]

Originally Posted by Dimman

The weight used would be proportional to the intended weight of the exhaust?

Doesn't need to be. The idea is to get two or more data points so that you can interpolate, or just use them for direct comparisons of welding techniques.

Originally Posted by Dimman

The loading that it's subject to is basically itself moving around from inertial and vibration, right?

We're also hoping that slip-fitting it will reduce some of the expansion issues.

Yep, inertial and vibration as well as thermal expansion/contraction and engine movement. Slip fits are good for thermal expansion, but I'm not sure how well they do with bending.

[Dimman]

Originally Posted by old greg

Doesn't need to be. The idea is to get two or more data points so that you can interpolate, or just use them for direct comparisons of welding techniques.



Yep, inertial and vibration as well as thermal expansion/contraction and engine movement. Slip fits are good for thermal expansion, but I'm not sure how well they do with bending.

For the slip fits, the thing that the welder pointed out is that they would be lap joints rather than butt joints. Since we don't have a hydraulic swaging setup or end-finisher, we will be welding the joints. So there will be some overlap on both sides of the joint and any bending force should be a bit more distributed.

[serialk11r]

You showed me that link before

I still want to try aluminum out because at temperature they're all about the same strength, and 10000 psi yield is still just as good as say nylon. Perhaps not for the bends, but for straight sections, with slip fit...maybe some reinforcement on the bottom.

[Dimman]

Originally Posted by serialk11r

You showed me that link before

I still want to try aluminum out because at temperature they're all about the same strength, and 10000 psi yield is still just as good as say nylon. Perhaps not for the bends, but for straight sections, with slip fit...maybe some reinforcement on the bottom.

Sorry, memory fail number two.

Homemade brought up an interesting point of using 2024. Stronger at high temp than 6061 or 7075. Plus the fact that there are aluminum mufflers. Could probably do a reasonably reliable axle-back with it. Problem is all the pre-bent tubing (U and J bends, etc...) for fab is usually 6061 so you would need access to a bending setup (I will need this if I do a Ti system as well...).

[koyv90]

it comes down to which one increases hp the most/price!

[serialk11r]

2024 and all 7xxx are expensive, and 2024 is not very corrosion resistant due to copper content. However on Onlinemetals (among other places) you can get 6063-t52 round tube which is cheaper, weaker, but less brittle than 6061-T6. I suspect that at 500F it's probably around the same strength as 6061. Apparently 6063-t52 is easier to bend than 6061-t6, although since this is already pretty ghetto I might just reuse stock bends or get stainless steel ones.

My idea is that using a strong exhaust tube isn't the most mass-efficient way...the tube only needs to hold in the gas and sorta not deform. Using a larger number of aluminum fasteners that spread the load across a greater area of the tube should be able to cut cost and weight. Hopefully. Thin walled tube is pretty cheap so I'm definitely trying this someday, keyword being someday.

EDIT: sorry 6063 only comes in 1/8" wall on onlinemetals...I probably was looking at 6061. But still on the quest to find cheaper aluminum! 5xxx would be nice since it's particularly corrosion resistant.

[old greg]

There's no point in using anything hardened for exhaust tubing. Between the welding and the high service temperatures it's all going to end up annealed anyway, so just start with T0 and make life easier on yourself.

[Dimman]

Originally Posted by old greg

There's no point in using anything hardened for exhaust tubing. Between the welding and the high service temperatures it's all going to end up annealed anyway, so just start with T0 and make life easier on yourself.

Soft aluminum is horrendous to work with, though. Gummy, clogs tools, doesn't deburr nicely. Most mandrel bends are 6061-T6.

I don't know where the '6061-T6 has bad machinability' comes from. Unless there is a major difference between the official 'T6' and the industry-common 'T-651' treatments. AFAIK T-651 is just T-6 (precipitation hardened and artificially aged) that is pulled or something, some minor cold working.

I actually had a little argument with a project planner at work a few days ago about this. Worked with some 'mystery' aluminum and it was awful.

[old greg]

Originally Posted by Dimman

I don't know where the '6061-T6 has bad machinability' comes from.
...
I actually had a little argument with a project planner at work a few days ago about this. Worked with some 'mystery' aluminum and it was awful.

I was thinking in terms of bending tubes. Machining-wise I love 6061-T6, other than getting built-up edges on tooling I've always found it quick and easy to work with.

[Dimman]

Originally Posted by old greg

I was thinking in terms of bending tubes. Machining-wise I love 6061-T6, other than getting built-up edges on tooling I've always found it quick and easy to work with.

Fluids take care of the edge build up. (Haven't done any bending, myself.)

Problem I can foresee is where the steel joins to the aluminum. If you want maximum weight savings it will require more more aluminum, which means closer to the engine, and hotter. Then the expansion, corrosion and vibration/fatigue stuff becomes worse. Especially at the joint. Flanges will be no-good, slip-fits I can't see working due to different expansion rates, which leaves V-bands, maybe?

Properly welded 20gauge full cat-back titanium is the way to go, Serial! You get the weight benefits from less density (though more than aluminum), plus the reduced material amount from the much thinner walls.

[serialk11r]

Gahhh too much $$$ for my cheap ass :P
I just remembered, there is water in the exhaust, and the aluminum will rot away where it touches steel, if it does at all. I think silicone can withstand exhaust temperatures at the pipe (it's good up to 600F or something), so maybe some silicone between the pipes?

[old greg]

Originally Posted by serialk11r

I just remembered, there is water in the exhaust, and the aluminum will rot away where it touches steel, if it does at all.

Sacrificial Anode? What's the melting point of zinc?

[serialk11r]

High enough? lol
Silicone insulation should work I think...it loses strength at 300C (er that's like what, 572F?) but probably won't go up in flames until well above that (I hope) It's used in cookware, and doesn't explode into flames upon contact with an open flame so...