Difficulty with Mcquaid Ehn Carburization to Obtain PAGS

[kingsley47]

My company has begun attempting to perform prior austenitic grain size evaluations. I am attempting to use the Mcquaid-Ehn process but am having difficulty getting the PAGs to show upon etching. I'm typically working with very low P+S 4140 or 4340 that has been quench and tempered to produce a martensitic structure. Nearly every sample I run ends up looking like I took a cross section of the carburized layer. I am mounting two cuts on each mount, one with a longitudinal cut and with a transverse cut and both are giving me the same results. I have checked to see how much material my grinding & polishing process removes and it is about .025" when all is said and done. From my understanding, the case depth should be about .050". However, if I have been grinding through the carburized zone, and I am effectively looking at a cross section of the carburized layer at the edges of my samples, then my case depth is extremely thin because that layer is measuring at roughly .0012" from any edge. Not sure what would be causing this but I'm hoping someone can point out the error in my processing.


The process we use is as-follows
- take two cuts, longitudinal and transverse, on an abrasive saw
- pack samples in carbon granules
- heat treatment: no atmosphere but samples are packed in a container with carbon granules, 1700F for 10 hrs, furnace cool (decrease 50F every 30 minutes) until reaching 1100F. Remove and allow to air cool.
- Mount and grind / polish
- etch in nital

I would really appreciate some direction on what I should be changing in the process to get the PAGs to etch. Thanks in advance for anyone's help.

 

[EdStainless]

My first question is why are you doing this?
If the alloy contains Al/Cb/V it will be fine grained, you can't get otherwise.

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

 

[kingsley47]

EdStainless,

It comes from requirements that are written into a new industry specification that we are manufacturing to. The API specification requires the product manufacturer (us) to perform a grain size analysis on every heat treatment lot of bolting we manufacture. So it's really just a spec compliance issue and we don't have much choice in the matter.

 

[EdStainless]

Funny, SAE is removing it form all AMS (aerospace) specs if the chemistry is met.
There are some very good (old) guides to doing this test.

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

 

[Maui]

You didn't mention how thick your samples are, how you are measuring the case depth, or what your etching procedure consists of. Please fill in the blanks.

Maui

 

[redpicker]

Take one sample per heat as-quenched and etch with picral. Even with low P & S, the GB will show up well. It takes a bit of fussing with the picral (try adding a drop of dishwashing liquid) and don't throw it out when it turns black.

I was getting the steel suppliers to certify the heat as being fine-grained, which satisfied most auditors. But, API really should get with the program and accept that modern Low-Alloy steels are going to be fine-grained by default; you cannot buy commercially available 41XX or 43XX steels that are not fine-grained.

 

[kingsley47]

Maui,

The samples are roughly a 1/2" thick.

I was measuring what looked like case depth by using the linear measurement tool on our metallograph software. Basically I can see some austenite looking grains along the entire edge of all of my samples. But it's literally only one or two grains deep, which is corresponding to a layer of about .0012".

The etching procedure consists of using a 3% nital (ethyl) solution and immersing the samples in the etchant for about 10 seconds or until they cloud over.

Redpicker,

I'll give the picral etching a shot. I was hesitant to go this route because of exactly what you alluded to, the process is finicky. From what I understand the Mcquaid Ehn process should be pretty consistent. So I'm assuming I'm doing something incorrectly. I'd love to at least figure out where I'm going wrong, if only to satisfy my own curiosity.

 

[redpicker]

Redpicker,

I'll give the picral etching a shot. I was hesitant to go this route because of exactly what you alluded to, the process is finicky. From what I understand the Mcquaid Ehn process should be pretty consistent. So I'm assuming I'm doing something incorrectly. I'd love to at least figure out where I'm going wrong, if only to satisfy my own curiosity.

Well, picral isn't all that finicky as it is just a bit tricky to get the hang of it. With low P steels, you need to be using an as-quenched sample. A drop of dishwashing liquid helps, along with some water; maybe even a drop or two of HCl. Submerge the sample for a while; I'd start with 10 minutes. You are etching for grain boundries, not structure, so don't worry too much about overetching. Wipe off the smut with a cotton ball. You might have to lightly polish afterwards to set-off the grain boundries.

You might have to etch your first sample two or three times to get it. Once you get it, though, it is easy. Do not throw out etchant because it turns black. In fact, you aren't going to get a good etch until it turns black. We used to run a lot of Sour Service product which, because of the way the spec was written, we had to run a grain size on each heat. I tried a lot of ways and found the easiest way was just to ask production for one as-quenched piece per heat. The spec was 6 or finer; our process produced 10-12 and smaller. We still had to run one every time.

 

[EdStainless]

There is technique involved.
As for the etch my practice was to mix half old with half new when I needed more.
It has been 25 years since I worked for a company that would let me use Picric acid.
Whatever you do by prepared etch and not straight picric, I have seen and explosion.

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