17-4 wide hardness spread after aging 3

[Lyrl]

My company has aged at 900° for 1 hour some small pieces of 17-4 material (120 pieces of 1.9" long bolts that weigh 0.022 lb each). Our customer says these parts were machined from the same bar as the two previous orders we processed, where the hardness results were as expected.

On the current order, the hardness came out at 36-37 HRC (converted from HR15N), below the expected hardness of 40-47 HRC for the H900 condition. We re-aged for a second hour and then had hardness of 38-49 HRC. We tried a third hour at 900°, with no change in the hardness.

Any idea what could cause this wide hardness spread with both soft and hard parts?

 

[israelkk]

This is a known fact that for same heat treatment even if the tensile and yield are within spec the hardness can vary in a range of 10RC. If you look at MIL-HDBK-5J you will see that the spec only ask for minimum yield and tensile properties there is no requirement for minimum hardness.

 

[metengr]

On top of what was mentioned, converted hardness values adds another level of variability versus direct hardness measurement using HRC scale.

 

[EdStainless]

The only way that re-aging could raise the properties is a) if your original age was not hot enough (below 900F) or b)if your original age was too short of a time.
There are only a couple of sources of scatter, either the two things above or a variation in the annealing temperature.
If you want to check material re-anneal a couple of them and then age them. I would do this to a couple of the softer ones and a couple of the harder ones. 1900F x 30min, cool to room temp, 900F x 45min


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[tbuelna]

There was no mention of the heat treat specification used (if any). I checked AMS 2759/3 and the hardness range for 17-4PH H900 is Rc 40-47. The H900 aging set temp is 890-910 degF and the aging time is 60-75 minutes. Additional aging is permitted only if the part hardness exceeds the required maximum. Regarding metengr's comment about variation in conversion of hardness values, AMS 2759/3 table 7 lists a minimum tensile strength requirement of 190ksi for 17-4PH condition H900.

 

[Lyrl]

My previous experience has been consistent with EdStainless's statement: that with our hardness results the original age must have been too cold or too short. But in this case we run monthly TUS (±10°F) and weekly SAT on the aging furnace and are very confident the logged temperature of 900°F for 1 hour was what the parts actually saw.

We re-solutioned the order, re-aged it (in the same aging furnace) and now hardness (testing the same 5 pieces that previously were 38-49 HRC) is 45-46 HRC equivalent on all 5 tested pieces.

It is very strange.

 

[metengr]

Based on what you presented there is nothing strange only a case for re-evaluation of your in-house process controls.

 

[MagBen]

It was clear to me the solution heat treatment issue: the bar was not completely transformed to martensite due to either too low annealing temperature, or in-sufficient oooling. Re-aging won't solve your problem.
Further, it was not due to too high solution temperature (ferrite formed), in which case even re-solution will unlikely solve your problem.