How to weld onto a turbo exhaust housing?

[tig_21]

cast iron is heavily used on turbo exhaust housings and manifolds....i don't know why you would think they can't take the heating and cooling down....?

[jakeru]

Take a magnet to your turbine housing. Does it stick as strongly as you expect it would stick to steel?

If it doesn't stick very well, it's likely an austenitic cast stainless alloy, such as NiCr-202. The better turibne housings are cast stainless alloys such as this, and not "plain cast iron" alloys.

I would weld to a nonmagnetic turbine housing using 309 stainless steel filler rod (whether you are welding stainless or mild steel to it.) 309 has plenty of Nickel and Chromium, and a thermal expansion rate about halfway between austentic stainless steel and mild steel, so it's a very versatile choice regardless of the alloy.

If your turbine housing is a not an austentic alloy (e.g., if a magnet sticks to it very strongly), then I think I would rethinking doing any sort of welding to it at all. Also, I would be suspicious about it holding up very long.

[MikeGyver]

It's kinda difficult... Here's one I did.
I preheated the housing to maybe 500 degrees with a torch, (you could use an oven too), and just used 308L filler. The most important part is getting it to cool down slowly, I mean like hours. This is when it will crack. You can put it in kiln dried sand (not sand from the sandbox, that's way too moist), or put it in a firepit/bbq charcoals, or even the oven and turn it down 100 degrees every half hour or something. Also lightly peen the weld with a hammer immediately after each short weld to help relieve stress... and use plenty of filler, don't do an autogeneous weld.
Ideally I think you would use Ni-55 or Ni-99 filler, it's expensive but you can sometimes find small quantities on Ebay.

Originally Posted by astaindsoul5446

It's not iron, they are steel! I've ported about 20 turbos, all. Cast steel. Cast iron wouldn't take the heating and cooling cycles

You mean like cast iron stock exhaust manifolds that only last oh, quarter century or longer?
I did a spark test on the housing pictured above, it has standard cast iron sparks; few, orange, and feathery. I've handled many cast iron things and as far as I can tell it's the exact same stuff. I have also seen cast (stainless?) steel parts like the wastegate housings on turbosmart gates, those are something entirely different but still cast.
Turbo housings could be grey cast iron, nodular iron, ductile iron, or malleable iron for all you know, any of these would probably hold up. There are all different kind of cast irons, with very different properties. Turbo housings are absolutely not cast steel in the sense that it's anything like mild steel, if it is 'steel' it's something very, very different (cast iron like).

[chimmike]

Thanks, that's what I was thinking. I was under the impression most turbine housings are iron, only a few are steel unless specific manufactureres prefer. I've had most of my experience with Garrett turbos and they've all been iron from what I could tell.

I know a few people "braze" internal wastegate flappers shut on these housings to go external gate. Seems sketchy to me. Not sure what filler they've used.

[con_fuse9]

Originally Posted by astaindsoul5446

309 is for carbon to stainless. Your dealing with cast steel. Build or buy the proper adapter for the turbo housing bolt pattern. It's not difficult, you can buy the v band piece off eBay or where ever then take a piece of stainless plate and cut it to the exhaust scroll flange and wel the vbband flange to the plate. It is a lot better way to accomplish what your trying to do. What turbo are you doing this with? What is it on? Post a pic of the flange and bolt pattern. They are pretty easy to come by...

309 is a commonly available high nickle stainless rod. The high nickle allows for some ductility when it comes to joining two pieces that expand differently under different temperatures. That beings said, I don't think it will work on an turbo exhaust housing were temperature variations can be very high on top of different expansion rates.

There are higher nickle rods that are probably a better choice. I have done some non-critical welds on cast using 309. 309 (and 308) weld nicely and normally blend in nicely with the parent metal. I had a crack on an old tool. Decided to try and fix it. Ground out the crack. I had to be carefull not to go beyone 100% penetration because the back side was machined flat. I found it difficult getting the puddle right. In the end the weld held (tool is used at room temp only) but what surprised me was the shrink back when it all cooled. Ended it up having to re-machine it flat anyway. Weld held (again, room temp only).

[astaindsoul5446]

I apologize, I had it backwards, meant cast iron. It cast steel.

[jakeru]

I just checked the magnetism of four turbine housings I have around. Two were quality OEM units (Garrett and KKK), and the other two were generic import knockoffs.

Both the OEM Garrett and the KKK were nonmagnetic. The KKK turbine housing had raised cast letters in it "Ni Cr 202", indicating it is indeed 202 stainless. I just looked up the alloy for 202 stainless, and it has 18% Chrome, 5% Nickel, 8.5% Manganese, and 1.5% Carbon (I was surprised it had that much carbon... it's a lot, but equivalent to the carbon rating in 303 stainless as well. Apparently austentic alloys can hold a good bit of carbon without necessarily being brittle.)

I'm not sure what alloy the Garrett was, but it also was nonmagnetic (so likely some cast austenitic stainless steel, having good amount of nickel in it and some chromium.)

Both knockoff brand turbine housings were indeed magnetic... so I would guess gray cast iron.

[gottarollwithit]

Wow, this thread started off sketchy, but really resulted in sparking some great thoughts and ideas.

The way it stands now is that i have a spare V band and a spare exhaust housing. I really don't stand to lose much by trying to weld it up. If either break or get destroyed, i can just do this job "properly" and buy an exhaust housing that already has a larger exit hole.

It seems that 309 rod and a preheat is kinda the consensus here.

So, i guess the plan is.... Grind/clean all of the surfaces, warm up the exhaust housing with a MAPP torch, plop the V band on top and work my way around the circle by welding a bit, then switching to the opposite side of the circle. Probably start with the outside of the V band, then do the inside diameter of the V band afterward?

Am i missing anything? Have any other tips for actually doing the weld?

And.... Think i can get away with doing this weld on an 80A Harbor Freight TIG welder? I have access to a much bigger and better TIG, but is a lot of juice really needed here? If i crank up the heat super high, won't that just blow through my V band?

[gottarollwithit]

And... Do ya think 308L filler rod will work?

I just happen to have some in my garage...

[jakeru]

Yes, you are missing something - taking a magnet to it to find out what it is.

If it's austenitic stainless steel (nonmagnetic), then weld it like it is austenitic stainless steel. That means, no pre-heating is necessary (although you could if you wanted to weld it with less amps), and use a filler rod like 309 stainless.

It's it is magnetic, then it's probably grey cast iron, so then, weld it like its grey cast iron. That means probably using a filler rod and techniques designed for welding cast iron - best I know of for welding "cold" is nickel-55 filler rod. I would not recommend welding gray cast iron with 309 filler rod. I've tried it with awful results - hairline cracks appearing on one side of the weld bead. I got a much better result welding cast with a nickel-55 rod.

Different materials call for different techniques, so find out what material you have before you determine the technique.

[chimmike]

IMO if you preheat it, you should do so by placing the thing in an oven so it heats uniformly, too.

[gottarollwithit]

Well, the housing is definitely magnetic.

It seems that other guys that have welded this housing have used a 309L rod. I'm going to try to find out what the housing is actually made of.

[gottarollwithit]

Yep, one of the guys on the car forum claims that the housing is made of cast iron.

[T1488]

Guys is I ever ask a really dumb question or if I'm to lazy to look up my own stuff up on the web please BAN me.

[dave powelson]

Originally Posted by gottarollwithit

Well, the housing is definitely magnetic.

It seems that other guys that have welded this housing have used a 309L rod. I'm going to try to find out what the housing is actually made of.

http://weldingweb.com/showpost.php?p...1&postcount=25

2 days ago, I suggested in the above link-post #25 that you spark test it,
to determine whether it's cast steel or cast iron.

[astaindsoul5446]

Wanna take bets on how many bumps it takes to break it off with 309? Anybody in?

[gottarollwithit]

Originally Posted by astaindsoul5446

Wanna take bets on how many bumps it takes to break it off with 309? Anybody in?

Whoa... Don't join the group of d-bags in this thread!

You're aware that the only reason i'm looking into welding this V band on is b/c i already have all of the materials, right? By starting this thread, i'm simply investigating the feasibility of the project. I might not be a pro like you guys, but i at least have enough sense to seek appropriate information before making a decision that could waste a bunch of my time.

The "proper" way to address this issue is to simply buy an exhaust housing with a larger output diameter. There's more than one way to go about doing that as well, so this option is undeniably still on the table.

[SundownIII]

gotta,

The problem here is that you didn't seem to read ANY of the responses given.

If you planned to do it "your way" with "materials on hand", why did you even post in the first place.

[MikeGyver]

Originally Posted by dave powelson

http://weldingweb.com/showpost.php?p...1&postcount=25

2 days ago, I suggested in the above link-post #25 that you spark test it,
to determine whether it's cast steel or cast iron.

what is "cast steel"?

Everybody is throwing that term around like it's actual nomenclature. "Cast steel" doesn't mean anything. Be more specific to what the material is, not the process.
Is it mild steel that has been cast? Malleable cast steel like a crankshaft? ... cuz if you spark test that it looks like cast iron... Nobody is being very helpful, which is understandable if you don't know what you're taking about (which I'm kinda getting that vibe).

[sarg]

Since most of what my friend Andy has welded have been cast steel turbo manifolds from Borg Warner, I decided to give a cast iron manifold a try that I had laying around. The manifold is cast iron and I tried the magnet test I did not try the grinder test though. The pipe I welded to the manifold was stainless steel 304 pipe. I used a 309 rod at the suggestion of the local welding shop. Bear in mind I am still learning to TIG and this was not the easiest thing in the world, but it did stick. To attempt to release any tension I would tap the weld with a small hammer after welding an inch or so, let it cool, do the next weld for an inch and repeat. It was a pain for a welder of my meager skill. Getting the amps high enough to melt the thick cast and not blow a hole in the thin stainless was a challenge!

I then proceeded to let it all cool down, heated it back up a couple times to red hot with a torch and let it cool off (trying to simulate heat cycling from driving it). I beat the crap out of the tube with a 5 lb mallet and it bent the crap out of the tube with no hints that I could see of the stainless or the iron cracking.

I know this is probably not the most scientific method of testing, but it made me feel a little better about my upcoming project where I will more than likely do this with some other GM manifolds so that they have a v-band flange on them rather than the factory 3 bolt or doughnut flange.

[MikeGyver]

It seems like cast iron usually cracks right away if it's going to. You can usually tell pretty quickly if it's going to work or not.

[usmcruz]

309L on cast!:what: Ni Rod

Sent from my DROIDX using Tapatalk 2

[sarg]

Donno, maybe they didn't have that in stock, lol.

[usmcruz]

Originally Posted by sarg

Donno, maybe they didn't have that in stock, lol.

Lol

Sent from my DROIDX using Tapatalk 2

[dunemetal]

Originally Posted by MikeGyver

what is "cast steel"?

Everybody is throwing that term around like it's actual nomenclature. "Cast steel" doesn't mean anything. Be more specific to what the material is, not the process.
Is it mild steel that has been cast? Malleable cast steel like a crankshaft? ... cuz if you spark test that it looks like cast iron... Nobody is being very helpful, which is understandable if you don't know what you're taking about (which I'm kinda getting that vibe).

^^^ Agree, the actual alloy as well as the shape of the casting are some important criteria in how and what to weld with. 1020 vs 4140 vs 8620? Different animals.