welding 5120 H , and PWHT

[hansonbrest]

I have been repairing blocks of 5120H in an extremely high fatigue application, (they hold a 120 kg steel component which oscilates at around 300 cycles per minute),weight is around 150-200 kg per block.

They are originally toughened and hardened (to approximately 63 rockwell) , and i need to anneal them before welding.

I preheat to 400 Celcius + and weld with a fluxcored consumable to E110T5-K4 (Cigweld Tensi-cor 110 TXP in Australia)

Its composition is ;
Carbon 0.08 %
Si 0.4
Mo 0.4
Mn 1.5
Ni 1.9
Cr 0.3
and tensile class 800 Mpa / yield 720 Mpa /impact 47J at - 50 C.

I have left the PWHT process up the the Heat treaters i use, but they are "reluctant to divulge it" , and i want to know precisely what the most appropriate PWHT is to obtain the best toughness ,and resistance to subsequent failure, and if there are any recommendations for an improved consumable over what i am currently using.
I have about a 96 % success rate with this process , even after the items have been in service for 5-6 years after repair , however , the failures i get generally occur within a couple of months of commissioning , and i suspect the cause may be variations in the quality , or method of PWHT.

any ideas ?
i realise that i should start testing the failed items for hardness and consistancy to get a clue as to the reasons for failure,but i generally have a fast turnaround requirement to get them back in service .
The failures are clean cracking ,and not necessarily in the repair zone.

Thankyou.
Mark

 

[Metalguy]

5120H won't go to Rc63, so if that hardness is correct the blocks must be case-hardened. Where are the blocks cracking? A good photo would help, with a sketch of how the load is applied.

 

[hansonbrest]

The application of this item is as a toolholder in hydraulic Rock hammers

yes , there is a case hardness , but i am not sure what process is used, and the manufacturer never tells you what their processes are , the only way i could identify the steel was to do a spectroscopic analysis , and even then , the 5120 H match was the "closest match" ,but not spot on.

I should point out , that i grind away about 1.0-1.5 mm of the case in the weld zone before welding , and blend/polish the finished weld against the parent metal to remove notching.

I had assumed the original case was a secondary effect after a core toughening PWHT ????

 

[Metalguy]

No. The case hardening probably involves heating the blocks while surrounded by carbon-containing material, or by a C containing gas. The C level is raised from the ~.20% of the base metal to much higher levels-4 or 5 times as much, maybe more. Then the block is tempered at a fairly low temp. ~350 Deg F.