Board logo

Grrrr, exhaust manifold cracking!!!
NS Dev - 9/8/07 at 07:14 AM

My own work, so can't complain, but my exhaust manifold keeps cracking!!!!

One runner broke off some time ago at a butt weld which lacked a bit of (tig) penetration, I can understand that one.

But yesterday I spotted a crack at 90 degrees ACROSS the 2" to 2.5" collector, running at 90 deg through the weld...........ehhhh??????

Took it off to weld that one up and found cracks in two more runners.

Its made of 1.6mm 304 stainless, mandrel bent and tig welded, and I have no idea why its cracking so much!!!

(covered 600 miles)


NS Dev - 9/8/07 at 07:17 AM

PS its not s complete hash job:


exhaust 4
exhaust 4


here's the collector at the time of making, the crack goes top to bottom in the pic, from 10mm above the centre of the weld seam to 10mm below it!!!????


[img][/img]


designer - 9/8/07 at 07:23 AM

The length of the primary tubes is applying stress to the welds.

It would be best if the primary turned down 90deg and then another 90deg lower down to exit the engine bay.


BenB - 9/8/07 at 08:23 AM

?? some bracing pieces needed between the various collectors to spread the load.... Unusual, though pretty much the same thing has happened on my nearside collectors.... I'm letting it blow at the moment.....


907 - 9/8/07 at 09:22 AM

It's that cheap old welder you bought Nat,

either that or "operator error"


Sorry, serious mode now.
How solidly mounted is the can? It looks from the pic that it's fixed it 2 places, and they look like metal.

Systems should be allowed to flex, or any engine movement or vibration causes stress cracks at the weakest point.

Atb
Paul G


jimgiblett - 9/8/07 at 10:34 AM

I had my old manifold crack at a weld. The firm who made it suggested it might not have been heat relieved. After rewelding they used an oxy torch to heat up the weld and surounding area to near melting point and then gradually turn down the torch allowing the metal to cool more slowly.

I am no metallurgist but it makes sense that tig or mig heats and then cools rapidly and like quenching makes the metal brittle.

- Jim


smart51 - 9/8/07 at 10:40 AM

quote:
Originally posted by jimgiblett
I had my old manifold crack at a weld. The firm who made it suggested it might not have been heat relieved. After rewelding they used an oxy torch to heat up the weld and surounding area to near melting point and then gradually turn down the torch allowing the metal to cool more slowly.

I am no metallurgist but it makes sense that tig or mig heats and then cools rapidly and like quenching makes the metal brittle.

- Jim


How hot does it get in service? I'd imagine any heat treatment would have to be really hot to make a difference over in service temperatures.


jimgiblett - 9/8/07 at 10:45 AM

Depends where in the system it is. Yes in service the primaries can get very hot. I watched my primaries glow cherry red on the dyno but the collector didnt. And it was here that the crack appeared.

- Jim


Minicooper - 9/8/07 at 10:47 AM

Have the engine mounts gone soft? does the head have a mount like it would be in the original car, minimising the engine movement? Is the exhaust rubber or solid mounted? Sounds like the engine is moving

Cheers
David


907 - 9/8/07 at 11:29 AM

Hi.

anorak on. (sorry)

18/8 (304) stainless is austenitic at all temperatures,
so the chemical structure remains the same regardless of temperature or cooling rate.

There can be a problem with the carbon being absorbed by chromium when the metal
is held at high temperatures for long periods (600-850 deg C) but the structure can be
restored by heating (1100C) and quenching in water.

This is counteracted by the addition of niobium and /or molybdenum and keeping the carbon content to a minimum.

Hence for headers 316L (20/12/3) is a better grade to use.


(info taken from AC Davies, p79)

anorak off.


Boring ol' fart arn't I

Paul G


tks - 9/8/07 at 11:56 AM

if the can is mounted on 2 points than the headers take all the flex when the engine wants to move.

if the can was only mounted on 1 point than it could turn/twist.

Tks


NS Dev - 9/8/07 at 12:26 PM

quote:
Originally posted by 907
It's that cheap old welder you bought Nat,

either that or "operator error"


Sorry, serious mode now.
How solidly mounted is the can? It looks from the pic that it's fixed it 2 places, and they look like metal.

Systems should be allowed to flex, or any engine movement or vibration causes stress cracks at the weakest point.

Atb
Paul G


the can is mounted by one rubber bush at the back end, no mount at the front. The rubber bush is fairly solid though, maybe too solid!

The first weld to break did have slightly poor penetration, but the others were 100%, and as for the collector, I have NEVER seen a crack propagate at 90 degrees to a weld, on both sides, into "virgin" steel!!??


Simon - 9/8/07 at 07:25 PM

If it's a stress crack caused by too rigid a mounting, the way I did mine may help.

I have a rubber bush bolted to the exhaust can, which is bolted through a piece of steel. This piece of steel then goes into the mounting bracket and can swivel therein allowing for for aft expansion.

My (new) stainless manifolds are made from 316 butt weld (about 5mm thick), though as yet they are untested.

ATB

Simon


MikeRJ - 10/8/07 at 09:17 AM

I reckon it's more likely stresses due to unequal expansion of the primaries and secondaries. This would explain the direction of the crack propagation.

Slip joints on the collectors would probably help a great deal rather than being solidly welded.


NS Dev - 10/8/07 at 03:08 PM

mmmmm!!

I was thinking that one Mike!!

Most of the decent manifolds use slip joints and indiviudual primaries.......................

I've never seen a broken one from.....arrgghh name fails me, that bloke that makes em up in yorkshire and only uses mild steel, and he uses slip joint on it all!


Mark. - 10/8/07 at 10:57 PM

Dave Swarbrick? Or does he only do stroker spannies....elsie with swarbricks eeee that takes yer back