Sample preparation guidelines for Vickers hardness testing of weldments

[DefenderJ]

In the testing of weld hardness for welding procedure qualification I find that I am concerned over the preparation route which may be affecting the values I am getting.
As the hardness limit is 250HV10 (sour service applications) and materials are strong we are often working close to the limit in the heat affected zone.
Recently I had one sample prepared and the results looked high by 20-30 HV10, two further ones were identical to each other and confirmed the incorrect readings of the first.

I would like to define a preparation route for hardness testing of weldments to eliminate potential variability. The perceived variation would be from work hardening during cutting and milling or from overheating if coolant not used. Should I be concerned over pressure applied during polishing stages?
Materials involved are relatively strong steels 60 - 85ksi yield strength and corrosion resistant alloys such as Inconel 625. Both butt welds and weld overlay

The typical route that is used at the moment appears to be:
Band saw cut with or without coolant
Mill parallel with or without coolant
Possibly surface grind with coolant (I hope!)
Hand polish progressively with grit papers moving on to the next when the scratches from the previous stage are removed
Finishing 1200 grit or 1 micron.

I am thinking of introducing a set amount of material removal by initial grinding on a coarse qrit paper to remove say 8 thou / 0.2mm to remove some of the effects of the initial preparation.

Does anyone have any advice or know of a good resource for this question?
Many thanks

 

[metengr]

I would say your current proposed sample preparation is more than acceptable for hardness testing. Frankly, looking at the effects of local work hardening from sample polishing is not going to produce improved consistency or more favorable results. As you are aware, hardness testing will show variability and, as such, this is what I believe lead to your initial results.

 

[EdStainless]

The greatest risk lie in the initial steps, cutting and milling.
The potential damage done in a finishing operation in proportional to the material removed.
Each step is designed to remove both the roughness from the previous step and the damage.
Making sure that you machine cleanly is critical.
But even given good sample preparation hardness testing is only testing a very small spot.
If you hit an inclusion you will get a large value.

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

 

[DefenderJ]

Thank you both for your input, it is very helpful to get some opinion as I have not been able to find much literature on this.
I will develop procedures for the test houses we use for our work - concentrating on the initial "heavy" preparation.
This way we should be able to get consistent preparation and hopefully representative hardness measurement values.

 

[berkshire]

The steps involving polishing are more in line with preparing for microscopic examination than for an indenter.
You do not want to be surface grinding even with coolant. The grinding wheel will surface harden the material so that it will appear to have been heat treated. Hand sanding through the grades will give you more consistent results.
You should be able to use an indenter on a milled surface with consistent results.
B.E.

You are judged not by what you know, but by what you can do.

 

[EdStainless]

I have to disagree Berkshire, good grinding practices create far less structural damage than milling ever will.
The real problem is that people have detailed procedures for the hardness measurements, but no standard practice for sample preparation.
We took a hear treated bar that we knew to be withing a couple points HRC of the limit and cut it up. We made each operator prep two pieces and then took hardness measurements. Sure enough two of the guys had the same (lower) value and one was a few points higher. We wrote a procedure and then trained to it. And we had a way to show that it worked (the procedure and the training).

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

 

[DefenderJ]

Thanks for the further comments.
Lack of procedure for preparation is an obvious area for inconsistency.
We have decided to eliminate differences in machining practices by doing an initial hand grinding on a 60 grit Zirmet grinding paper (with water) to remove approx 0.2mm of material. Zirmet has a longer cutting life than standard SiC papers and allows this amount of removal using the same disc.

We did 2 samples from the same weld and got consistent results and lower than a previous sample that had minimal grinding after milling.

We plan to adopt this additional material removal in the grinding process as standard practice for our weld testing.

Regarding level of surface preparation - The same sample is used for the weld macro - therefore a high level of preparation is used. However, I would think that Vickers testing would need more careful preparation than a milled surface.