Shaft reclaimation challenge - solidification cracking? (Rush job)

[WeldQA]

Question:
How much weld metal (0.9 SOLID ES6W503AH using mixed gas 75% Argon 25% C02 ) can be deposited in a build up situation until other alloying elements are required?
Concerned of solidification cracking occurance with too much weld metal and not enough parent metal dilution.

Background:
230mm diameter shaft has stripped the keyway and made a mess full circumference. Shaft was tested and found to be too soft.
Urgent repair required to return to service until new shaft is manufactured; approx 3 weeks.

Parent Material as per analysis report:
"Metallography was conducted on a sample piece of the shaft. The material has an iron base with a carbon content of 0.20% and a manganese content of 1.13%. As stated in the mentioned report the material conforms to AISI/SAE grade 1518 or 1522. With a carbon equivalent of 0.47, its composition is indicative of a carbon manganese steel of the type AS3678 G350/G400/G450. Thus, it is estimated that the yield strength of the parent material is in the range of 350MPa to 500MPa with the most likely range to be 350-400MPa due to the size of the member. It must be noted though that the material tested does not comply with the Steel Alloy 4340 to AS1444 as per drawing."

Restrictions:
We have no metal spray or sub arc option. Semi automatic process only owing to time factor allowed for the repair. In stock is the afore mentioned solid wire in 0.9 or 1.2mm, or an E71 T1 (Lincoln or ESAB Flux Cored wire in 1.2mm).

My preliminary recommendation:
Machine damaged area down to 150mm diameter tapering ends of prep
Crack test
Pre heat to 170ºC
Use spiral welding technique as opposed to longitudinal stringers
Post heat to 350ºC hold for 1 hour per 25mm diameter. Cool at 55ºC/hour.
Crack test
Machine diameter and keyway
Final Crack test

Consequences of failure for this repair are not critical & no person can get injured if it doesn't work, or fails in service.
Boss may make it hard for me to sit down for a while though.

 

[metengr]

First - there is no issue with solidification cracking using the stated base material and filler metals. Second, what the heck is this filler metal??? I would use a filler metal equivalent to carbon steel and under one of the SFA welding filler metal specifications for either SAW, GMAW or FCAW. Your concern should be to minimize distortion. Watch the start and stops.

Your rather brief write-up is acceptable. Get rid of the short cut jargon - Crack test means surface NDT using wet fluorescent MT

 

[WeldQA]

Thank you metengr for your reply.
I appologise for my jargon, and Australian classification of the filler. It is:

Material type: ER70S-6
AWS Classification: A5.18
ASME Specification: SFA-5.18
F Number: F6
A Number: A1

 

[Tmoose]

Ts the damage so extensive the entire diameter needs to be reduce 80 mm ( over 3 inches )?

Or, are you creating a new uniform thickness welded shell in which to generate the keyway?
I'd be inclined to leave as much original material as possible, and excavate locally for the keyseat repair, fill the key to the repair diameter, and then start welding circumferentially.

I'm concerned with why the keyed component loosened and caused damage.
I don't think it was just because the shaft is a little soft or under alloyed.
Is the component large, or what drives the shaft?
Are setscrews involved? Perhaps with a history of loosening before, despite wishful anti-loosening measures like Loctite, double setscrews,or "killing" (welding) ?
if so The new shaft may be doomed to a similar fate.
API and others insist on a few thou' interference fits for keyed couplings etc starting with shaft diameters smaller than yours.

 

[Duwe6]

ER70S-6 or E71T-1 will both be 'soft', with a UTS of about 80-85 Ksi, and a brinnel hardness of about 180BHN.

 

[WeldQA]

Tmoose; cheers for your reply, the answer to both your first questions is yes.
Missing parent material starts at full depth of the key and in direction of rotation tapers out roughly at between 180 and 250 degrees. Though there is surface damage to the full circumference. And as Metengr noted the control of distortion is a factor.
I am not aware of the fault origin as yet.

Duwe6; I received the hardness report this morning. Using a D type indentor the hardness of the shaft in the keyway was found to be 150 - 160 BHN.

 

[WeldQA]

Tmoose, you are correct. Investigation report has identified the poor fit of the key as the instigator of shaft damage.

To any others interested:
The weld procedure and welder qualifications with the ER70S-6 A 5.18 wire complied.
Shaft build up was done with balanced longitudinal stringers, not spiral technique.
All NDT with wet flourescent MT complied. No rework or repairs.
Shaft is ready for installation.

Thanks to Metengr, Tmoose and Duwe6 for your input.

 

[kingnero]

Why was longitudinal welding executed, instead of circumferential as prescribed?

 

[Tmoose]

What was the diametral fit of the keyed component?
In my opinion Relying on a key to handle alternating loads is looking for trouble. The micromotions will reletlessly erode the key, keyseat, shaft and hub. The hub needs to be locked to the shaft either with diametral interference (straight bore or taper) or with heavy axial clamping against a stout integral shoulder.

If your machine's hub is a slip fit and uses setscrews and a key I'd be checking the setscrews for tightness frequently.
Adding Loctite (a fine product), double setscrews, or welding the setscrews will not keep the hub from loosening and destroying the shaft again (or the new shaft).
Initial set screw torque should be right up to HoloKrome or similar specs maximum.
Setscrews should be positioned to drive the shaft hard against the hub bore, so should all be in the same 90 degree quadrant.
(I think Some pillow block bearing were, and are made with 120 degree setscrews locking the inner race to the shaft, claiming this is "optimum." I remain sceptical, as I am of ALL setscrew mounted bearings. Eccentric locks are especially weak. All are compromised significantly by industrial shafting typical undersizes)
Setscrews for hub retention should never be at 180 to each other.

Is the keyed hub part of a belt drive? They are REAL good at loosening up cylindrical fits.
Is the key hub subjected to oscillating loads, or the whole thing driven by a reciprocating engine? They too are REAL good at loosening up cylindrical fits.

 

[madmantrapper]

I find these worth their weight in gold. Pulled me out of many a problems on power presses.

http://www.ringfeder.com/en/international/products/ringfeder/?g=382

 

[Duwe6]

That hardness probably indicates undiluted weldmetal. Nice and low for cracking resistance, kinda low for a shaft with keyway [you can't have both - if the shaft was mine, I would be happy with your results].