Distortion During Induction Hardening

[induction31]

We are attempting to induction harden shafts in a vertical scanning operation, and we are getting so much distortion that the shafts cannot be straightened, (without breaking) after hardening. The shafts are 1.875” diameter, 1547 continues cast material, (Mac Steel, formerly North Star, Monroe) 2.5 DI, 43” long. The spec is Rc 40 at .400”. We are heating with a 1khz, 250 kw power supply, and quenching with 12% Ucon Quenchant A at 95 degrees Fahrenheit. We are running the case depth at the minimum. Does anyone have any suggestions how to reduce the distortion and still meet the case depth requirements.

 

[bklauba]

Are you sure that you have to harden these to 0.4" depth, on a shaft 1.875" OD??

 

[Carburize]

What is the condition of the in-coming material?
Is it quenched and tempered? If so, what process is used for the heat treatment.
Assuming taht the IH process is being done uniformly it would appear that there is a lot of residual stress in the parts.

 

[induction31]

Bklauba
Unfortunately, the spec is Rc40 @ .400.
Carburize
The material is hot rolled material with no pretreatment. I agree that there seems to be a lot of residual stress in the parts. I suspect that that it has something to do with the way the material is cast, in that it has a bend as they cast the billet, 6”X6” I think. I would think that these stresses would be relieved when the material is reheated for rolling, so I guess that I am lost at this point. We are going to normalize a test batch of the material to see if that helps, but it will be a few days before we know.

 

[swall]

What is the core hardness in the incoming h.r. bar, and how much does it vary? How are your quench rings arrayed and what quenchant are you using? Also,when we first started using Northstar Monroe for 1050 special quality bar, we found a "picture frame" segregation near the bar center.When Northstar turned off one of the two magnetic stirers in the caster, the problem went away.Such segregation could give rise to hardness variations in the core material, which could be a factor in your distortion given the case depths you are running.So check your incoming bar micro for macro segregation.

 

[metengr]

induction31;
According to "Induction Heat Treatment of Steel", by Semiatin and Stutz, ASM publication, part distortion during induction heat treatment can be caused by too high of an austenitizing temperature, improper induction scan rates, uneven heating, nonuniformity of quenching, and relief of residual stresses.

The book mentioned possible rotation of the part to enhance uniform heating. Regarding quenching, sizing of the quench ring holes (orifices) and angle of spray were listed as important variables. A 30 degree angle between the axis of the spray holes and the axis of the part was suggested for single-shot spray quench devices used for induction scanning setups.

This book contains very practical information. You might want to consider purchasing it for future reference.

 

[norrick]

As metengr stated, this book seems like a good idea.
from my own experience in the field of hardening, quenching is the proces which can cause the distortion.
As a tool and die maker, i dont surface-harden my parts, so i'm not really familliar whith that proces, but for any hardening proces a couple of things are to be kept in mind:
Heating uniformly, and not to fast, giving the material time to form austenite

Quenching should be done in such a way that the workpiece also cools uniformly.

we quench our parts ( 1.2379 / D2 ) in a salt bath at 230 degrees celsius /446 Fahrenheit after heating to 1030 degrees celcius / 1886 Fahrenheit for hardening
the parts are put in quickly, and for instance long and thin parts ( like shafts ) are put in vertically to minimize distortion.

 

[induction31]

To answer a few of the questions, the material has a core hardness of 18-20 Rc. It seems to be fairly consistant. We to have seen the picture frame in their material but it is not visible in this material. We are using 12% Ucon Quenchant A for this part. We have tried different quench rings, high pressure, low volume and high volume low pressure. I will keep everyone posted if any progress.

 

[swall]

Induction 31--I am assuming that you are rotating the shafts during scanning--is this correct?

 

[induction31]

Swall
We are rotating the shaft during hardening. We have tried different rotation speeds, the slower speeds distort less, but the shafts still have excessive distortion.

 

[IRstuff]

I didn't see where you described what the distortion looked like? Is the shaft saggy or out of round?

And what the physical configuration of the heating looks like?

Does the shaft go immediately into a max heating region? Is there or should there be a transition?

TTFN

 

[metengr]

Induction31;
How did you optimize your induction scan rate? In the ASM book that I mentioned above, they do list recommended induction scan rates based on material and diameter.

 

[metman]

What is the straigtness tolerance?

Have you considered hot straightening just below the tempereing temperature immediately after quench?

Jesus is THE life,
Leonard

 

[ibf]

Induction 31 - I guess you have already received a fair amount of excellent suggestions and guidelines; however, here is another one I think you should at least take a look:
HEAT TREATING PROGRESS - MARCH 2003 - AN ASM PUBLICATION.

The article was written by Mr. Fred Specht and refers to distortion control by using a two-turn inductor and retracting idler rolls to minimize distortion (horizontal scanner though).

 

[Carburize]

Have you done the test with the normalized material yet?

 

[EdDanzer]

How are the ends supported? With this length to diameter ration and case depth the expansion during heating could force the bar out of straight if the ends are held rigid.

 

[induction31]

I would like to thank everyone for their contribution to this problem and answer the questions. Irstuff, after hardening, the shaft has an excessive bow in it, app .500” run out, the shafts are straightened to .090” after hardening. We are heating with induction, we start with a low heat level, then apply more heat as we start to scan. Metengr, we tend to run at the lowest frequency possible that will produce the effective case depth without heating too deep. Metman, we are straightening the shafts immediately after hardening, if the shafts have excessive run out, we temper them at 350 F, then straighten hot out of the temper. Ibf, we are using a single turn coil on this shaft, the pattern run out at the end of the shaft does not allow us to use a two turn on this part. Carburize, we have not tried the normalized shafts yet, keep reading. EdDanzer, we are supporting the shaft between centers, and yes the shafts grow app. .100” in length during hardening. We use a very weak spring loaded top center to allow for the expansion. Late yesterday, we started running the same shaft, but from a different heat, the shafts have very little distortion. The problem appears to be related to this one heat of material. The two chemistries are almost identical. We have two more heats of this material to harden, if they harden without the distortion, we will contact the mill. I will post the results on the shafts that were normalized. Thanks again for the help.

 

[TVP]

We have experienced similar issues of excessive distortion after induction hardening that can be turned on and off by changing the raw material. The material is from the same mill, with very little difference in the composition from heat to heat. Obviously the mill processes can have a considerable effect: re-heat temperature, rolling temperatures, strain distribution during rolling, cooling practice, etc. You should definitely discuss this in more detail with the Macsteel.

 

[induction31]

I just wanted to let everyone following this thread know. The shafts that we normalized did not make any difference. They distorted the same as the ones that were not normalized. The problem was isolated to the one heat of material.

 

[Carburize]

Interesting! If you ever get a chance to section a piece of the problem heat check if there is any gross segregation.