Is there a trick to welding exhaust headers?

[Admin]

Q. Is there a trick to welding exhaust headers? I currently am using just a small portable 110V MIG welder with shielding gas that seems to work pretty well, but I keep having problems with the tubes cracking, not on the weld but right beside it especially the welds by the header flange. The tubes are made from 18 gauge mild steel. Do you have to anneal them after they are welded to prevent this from happening or is there a special process or filler wire that would help? (Submitted by: Shawn Cassidy from Burlington Ontario )

A. This is an ongoing problem with many race teams. Some will run a rose bud torch across the welded area after welding, but most are now welding these with TIG. They seem to last a lot longer. The TIG process is less "violent" of a weld transfer than MIG. Use an ER70S-6 filler with your MIG welder and a smaller wire size like .030 or .023.

[fun4now]

are you purging the inside of the pipe wile welding ??? if not you are asking for cracks. unprotected SS will christilize and generaly go to sh--. so wile you are looking at a nice MIG bead on the outside, on the inside its more like.
i suspect the TIGers are purging the inside of the tube. most TIGers are used to haveing to purge and will automaticly do so on thin SS i would suspect, i would. only took me one time trying SS without a back gas and thats not going to happen again you can bet on it.

[monte55]

He's using mild steel........not SS

[PUMPKINHEAD]

you fellas realize this is a 5 year ago Q&A thread. right?

[KB Fabrications]

you fellas realize this is a 5 year ago Q&A thread. right?

Yeah but at least the resident welding "expert" has set him straight, huh.

[Portable Welder]

Even though its an old thread I have talked to some guys that do custom headers and they are real stiklers about welding on the cold side to pevent as much blead through as possible on the inside that causes unwant turbulance.

I also believe that they use OXY/ACy. to widen the heat affected zone, I believe we have also had the same talk about air plane tube welding using the same process.

[trstek]

Where does the crack come from?

Differences in thermal expansion / contraction in the metal over time, vibration fatigue, the metal in the heat zone changing properties during welding to be more brittle, all the above or something else?

How is the rest of the exhaust supported? Is the mass of the system putting forces on the tubes at the weld? Does it allow for thermal expansion.

I used to deal with undercoatings for industrial ceramics. What we found is solutions are sometimes complex, but always determined by application and life cycles.

When we coated pistons and headers with ceramic, it was important to start with the right materials.

[fun4now]

Ii says that a exhaust header with segmented bends will out flow one made from mandrel continuous bends. They had flow charts as proof. They suggested the segments created small amounts of turbulence that kept the flow laminar and thus carried more fluid out in the same time frame.


so rather than bend it you should cut it at an angle and weld it back togather to make a turn ??

[Aerometalworker]

Ii says that a exhaust header with segmented bends will out flow one made from mandrel continuous bends. They had flow charts as proof. They suggested the segments created small amounts of turbulence that kept the flow laminar and thus carried more fluid out in the same time frame.


so rather than bend it you should cut it at an angle and weld it back togather to make a turn ??

Actually I believe he is talking about having weld beads on the inside of the tubing vs. a smooth wall. Unfortunately its impossible to say that something like a weld bead would help, without looking at what all is going on in the system. For one, your exhaust system will never see a "laminar" flow, the velocites produced push the flow regeim into fully turbulent. Also the basic fact that the flow is dynamic and not steady state tends to rule out some things.

-Aaron

[fun4now]

bummer, i had to cut and reweld to get bends in my truck exhoste i was hoping i got lucky and did a good thing.
oh well, i kinda figured it was not the best idea, but ya cant blame a guy for trying.

[SundownIII]

James,

Have you ever welded an exhaust header?

[fun4now]

no, but i just did the rest of my truck.
sorry about the SS part i though ti read he was using SS, must have caried that over from another post.

[Doug Doty]

I would love to see the mentioned charts, I would have never even considered the possibility of increased flow through the use of sectioning around a curve Infact I have always tried to fasion this type of curve in a header to replicate a mandrel as best as possible thinking it to be the optimal arrangment.

If I were trying to justify this sectioning as having some advantage I think it is sound to argue that there in more volume of area in the bend on a sectioned curve than a mandrel. If you calculate the area of any circle and then diagionally cut through a section of the same dia. tube the area at the plane of the cut increases so it might seem that there is pockets of larger volume alternating with pockets of volume equal to that of a mandrel as you sweep around a sectioned curve. Might be something here.

Mandrel bending was a great replacment for non-mandrel bends in terms of flow and also as compared to sectioned curves from a cost stand point but I can't say I have ever seen any evidence that it necessarily flowed better than sectioning. I will consider the jury might be out on this.

Hey, Fun, any way to scan and post those charts or e-mail them to me.

I see the japs in import racing using sectioned tubing on the compressed air side of turbo cars instead of mandrel like I pay through the nose for on my mustangs. I had assumed it was a budget issue but in reality they more normally seem to spend more $$$ than we do anyway. Maybe they think it is supurior also......

[ridesideways]

when i've TIG welded exhaust headers, i just used backing tape to protect the inside from icicles and contamination.