Inconsistant dimensional growth after case-hardening 8620 1

[vcz]

Hello,
I am having trouble with a heavy clutch plate with internal
spline teeth. The part is 16"OD 5"ID 1.1"thick 8620 steel.
After case-hardening 54-58Rc .040"total deep, the pitch dia.
of the spline grows .020"-.040" on different parts in the same heat. Is this to be expected?

 

[EdStainless]

I'll assume htat you have checked the case on soem large and small parts and found that the depth and hardnesses are the same. If not, do it.
Have you taken a batch of parts and thermal cycled them with no hardening? After the same time/temp but without the case you may find a lot of veriation due to how the parts were originally manufactured.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[israelkk]

I suspect that you mean 0.0020" - 0.0040".

Are you sure it is just dimensional growth or the part distorts a little bit loosing its roundness causing the changes in pitch diamter? I am not sure how much but heat treated low alloy steels tends to grow too.

In the ASM vol 2 (8th edition) on pages 41,42 there is a disscussion and results of case hardening of 8620, 8625 carburizes gears which agrees with the range of 0.0020" - 0.0040". It also show the differences between various carburizing processes and following tempering or austempering, martempering etc. and how they affect the range of distorsion and dimensional changes.

 

[vcz]

Hello,
The between pin measurement pre-heat-treat was 4.955".
After heat-treat low part was 4.973",high part was 4.995"
The order was 33 parts. The large diminsion change is what
I am questioning. Thanks.

 

[Carburize]

If the parts were all from the same heat the first thing I would suspect would be that the pre-treatment may have varied. The second possibility would be variations in the carburizing furnace either parts were racked differently or there are major temperature variations in the furnace. A third but less likely possibility is that different amounts of surface residual stress were introduced into the parts during tooth cutting - if the hob or shaper was getting close to needing a re-sharpen then parts cut at the end of teh batch could have significantly higher surface residual stresses than earlier ones.

 

[BillPSU]

The amount of size change is not consistent with heat treatment of this size of part. What condition are the blanks in when you machine the parts, normalized, as cast, or as forged?

I think you have some other problem with the processing of these gears. How are they clamped on the hob. If they are clamped on the ends of the gear faces were the faces parallel and flat? You may have distorted the gear when you clamped the blank, cut the teeth and measured the size in the hob and then released the part adding distortion to the gear. Check you hobbing fixture it may be inaccurate also. If you have a lead checker you may see the lead is inaccurate also telling me there is distortion somewhere.

There is also the final choice operator error. Wrong pin, inaccurate measurement, or just hiding a screw-up.

 

[diamondjim]

Billpsu,
It is an internal spline and
cannot be hobbed. I kind of
agree that there may be some
in house problems other than
heat treating. Wrong gage
or something before and then
after?

 

[vcz]

Hello,
The pitch diameter measurements are correct. The material consisted of sawed blanks 1.25" long from a forged and rough turned bar 16.6" dia. I milled slots to various depths and checked hardness with the following results: .000"-55Rc, .040"-47Rc, .070"-42Rc, .100"-40Rc, .140"-38Rc.

 

[Carburize]

How were they carburized?
What kind of furnace?
How were they racked?

 

[stephenh15]

As a gagemaker, I would expect to see it shrink approximately .025-.030. This may seem a rather inane suggestion, but I have seen it happen many time (including doing it myself). If your gage has an electronic indicator, make sure that it is setup for the correct polarity. On better indicators there are usually a +/- button to change the polarity.

 

[mewhg]

Get the heat treater to exactly tell you how the parts were carburized. You may be able to find an anomaly in processing.

What furnace (for consistancy sake)
carbon potential
furnace temp
quenchant, quench temp, temp of part before quench
temper furnace, temp, time

In my experience most heat treaters do not control the process very well and this can lead to inconsistencies.

 

[gearHTr]

are you heat treating the blanks prior to machining? A normalize, quench, and temper to 28-34 HRC will take out a lot of your movement at carburize, as well as improve machinablity. I agree with mewhg that the heat treater may not care about distortion - see if he can control pre-heats and cooling rates at carburize.

 

[Steve898]

This is a common problem on short run parts with internal splines. There is simply no way to tell what a given lot of material will do when heat treated, there are just too many variables that will murphy you.

If the spline does not have to be 54-58 HRC, leave a little stock, then mask them off, carburize, harden, and temper, then finish them later. The splines should come out about 30-40 HRC.

Another suggestion for these parts: Freeze -120F for 2-4 hours, retemper. Try one see if it's worth it. This has worked for me, but not every time.