Gas Carburizing Uneven Case Hardness 1

[metalman8357]

Hello all,

I am case hardenening a large part ( 12"D x 3'L cylinder, HSLA steel) and I am seeing strange results on the test bars that are treated along with the part. Just to back up a little the furnace is new, and I have been playing around with carbon potential and process time in order to dial in the correct settings to meet my surface hardness and case depth requirements. I have been using 1" x 4" rectangular bars of the same material in order to dial in the 'recipe'. I have since moved to treating a full-sized part and I have included a test bar in the furnace in order to check the case properties before sending the part to Q & T. I have checked micro hardness traverse on the test bar and It is showing very strange results. The sides of the test bar come in with different case depths of 0.045, 0.038, 0.058, and 0.066" when they should all be showing 0.053". I have run this same recipe before and got great agreement between the four sides within 0.001" and great agreement between samples of 0.002".

I'm wondering if now having the large part in the furance is messing up my recipe somehow? The furance is all under computer control with sensors to control gas flow to maintain the correct carbon potential which stays very constant through the process. It also has a very large circulation fan which does a great job of keeping the gas moving. My thinking is that the furance should treat any surface inside regardless of size.

Another option is that I never check the prior process history of the test bar and it's possible it could have been carburized before. The only strange thing with this is that I am getting only 0.038" on one side which is much lower than the expected 0.053" depth. The bars are media blasted on all surfaces prior to carburizing, and are hung in the furnace by an inconel wire.

Any ideas gents?

 

[TVP]

The first thing I would check is the rectangular test bar. Did it have decarb prior to the heat treating cycle? Alloy segregation? Media blasting should have removed most of any surface contamination, but you should probably dip the sample in an organic solvent and blow dry before placing it in the furnace.

 

[dbooker630]

If you have not already done so, can you confirm that the circulating fan is operating at the correct RPM, let alone whether it is still in place. Also you may need to confirm the CO2 level in the furnace. The carbon potential may be stable, but does the system have to inject CH4 to counter any CO2 entry. These are worst-case scenarios but anything can happen.

 

[metalman8357]

Is there any reason to believe that the test bar would carburize differently given that such a large part was in the furnace at the same time? It usually takes about 6 hours to get the right case depth and theoretically this time should be the same regardless of the part size in the furnace?

 

[Maui]

How often you turn over your atmosphere per hour, your furnace temperature uniformity, carbon potential settings, method of carburizing (boost and diffuse versus constant carbon potential), etc. can all impact the resulting diffusion profile.

I would suggest etching the test sample in an appropriate acid solution and viewing it under a microscope to look at the case carburized region. Does it look relative uniform in thickness, or is there a large amount of variation that can be seen?

Maui