Garrett turbine weld

[3sgteluva]

Hi All,
as I have to weld a flange on a cast steel Garrett turbine housing, which are the better options in terms of flange material and filler rod?
Thanks

 

[500e]

What does it say about repairs in the Garret Maintenance manual.
If allowed it will give you a procedure, including materials to use if not you are letting yourself in for a lot of hurt if it goes pear shaped

 

[3sgteluva]

What does it say about repairs in the Garret Maintenance manual.
If allowed it will give you a procedure, including materials to use if not you are letting yourself in for a lot of hurt if it goes pear shaped

Garrett says nothing, but I know for sure it has been made multiple time with success.
Just trying to get the better options

 

[frank horton]

depends on what u wanna do....add or modify a flange....??
then I'd use braze if poss....less heat to distort stuff....
could be wrong tho...

 

[KemppiFrog]

I have had some success with 'cast' rods with a DC stick welder.
One assumes this a turbo on a road vehical, not a Garret gas turbine!

 

[3sgteluva]

Yes, I'd need to weld a flange on a cars' Garrett turbine housing, exhaust side.
Distortion won't be an issue, as the flange is thicker than necessary to have room for a skim.
Would a 304SS flange be ok?

 

[Jelly_Sheffield]

Yes, I'd need to weld a flange on a cars' Garrett turbine housing, exhaust side.
Distortion won't be an issue, as the flange is thicker than necessary to have room for a skim.
Would a 304SS flange be ok?

Based on the max housing temperature given by Garret for their G-series turbos, and the service conditions expected, a high temperature corrosion resistant steel would be typical.

Sources online vary, in suggesting either 321 Grade Austensic, or a Duplex steel with a composition similar to X2CrNiMoCuWN25-7-4 (S32760) is used.

Given how hot turbos can get (especially on the exhaust side) and the extensive exposure to vibration, I personally wouldn't want to weld Austensic to Duplex in this application, as welds of that type are characterised by embrittlement of the HAZ unless welded with a procedure designed to ensure the correct α/γ ratio, excessive austenite formation in the HAZ will give rise to a vastly increased risk of fatigue related failure.



Given that using a 321 flange on a duplex casting, or a S32760 flange on a 321 casting would both give rise to that failure mode...

The ideal approach would be:

- to find the grade of steel for the flange which most closely matches the original casting,

- pick a filler which will minimise the dilution of the parent metal, (for instance MasterWeld 2594 Super Duplex, if you were using S32760 steel)

- design the joint to minimise the amount of weld metal required.



If you have a large Scrap Merchant near you, they may be willing (for a fee, or perhaps a round of butties) to ID the steel composition in the turbo housing with an XRF gun, so you can look up the right materials.

If that's not an option, then whilst there's very little info on spark testing of duplex stainless, you could work with the fact that Austensic grades tend to produce relatively few reddish sparks with minimal forking, whilst Feritic grades produce more and whiter sparks (again with minimal forking).

So you might be able to tell by comparison to a known piece of 304/316... If you're seeing more, and whiter sparks when grinding then there's probably some ferrite present, indicating it's the duplex.



Obviously if you think it's made from 321 grade, using a 321 flange and filler rod would be best.

If it is made from a duplex steel and you are still going to use Austensic Stainless for the flange then:

- 309 (or indeed 316 or 321) would be better suited than 304 (due to the higher operating temperature, which is close to the Garret Spec), and

- it's critical that you use an appropriate joint design (full penetration only, no fillets) and appropriate duplex or 309Mo filler rod... IMOA has a useful guide for this.

If you're not going to be pushing the turbo to its limits, and don't need it to hold up to daily service over an extended period of time, that's probably an entirely acceptable trade off.

 

[3sgteluva]

Based on the max housing temperature given by Garret for their G-series turbos, and the service conditions expected, a high temperature corrosion resistant steel would be typical.

Sources online vary, in suggesting either 321 Grade Austensic, or a Duplex steel with a composition similar to X2CrNiMoCuWN25-7-4 (S32760) is used.

Given how hot turbos can get (especially on the exhaust side) and the extensive exposure to vibration, I personally wouldn't want to weld Austensic to Duplex in this application, as welds of that type are characterised by embrittlement of the HAZ unless welded with a procedure designed to ensure the correct α/γ ratio, excessive austenite formation in the HAZ will give rise to a vastly increased risk of fatigue related failure.



Given that using a 321 flange on a duplex casting, or a S32760 flange on a 321 casting would both give rise to that failure mode...

The ideal approach would be:

- to find the grade of steel for the flange which most closely matches the original casting,

- pick a filler which will minimise the dilution of the parent metal, (for instance MasterWeld 2594 Super Duplex, if you were using S32760 steel)

- design the joint to minimise the amount of weld metal required.



If you have a large Scrap Merchant near you, they may be willing (for a fee, or perhaps a round of butties) to ID the steel composition in the turbo housing with an XRF gun, so you can look up the right materials.

If that's not an option, then whilst there's very little info on spark testing of duplex stainless, you could work with the fact that Austensic grades tend to produce relatively few reddish sparks with minimal forking, whilst Feritic grades produce more and whiter sparks (again with minimal forking).

So you might be able to tell by comparison to a known piece of 304/316... If you're seeing more, and whiter sparks when grinding then there's probably some ferrite present, indicating it's the duplex.



Obviously if you think it's made from 321 grade, using a 321 flange and filler rod would be best.

If it is made from a duplex steel and you are still going to use Austensic Stainless for the flange then:

- 309 (or indeed 316 or 321) would be better suited than 304 (due to the higher operating temperature, which is close to the Garret Spec), and

- it's critical that you use an appropriate joint design (full penetration only, no fillets) and appropriate duplex or 309Mo filler rod... IMOA has a useful guide for this.

If you're not going to be pushing the turbo to its limits, and don't need it to hold up to daily service over an extended period of time, that's probably an entirely acceptable trade off.

Thanks Jelly, very useful info