Heat treating of AISI 1070 versus AISI 1074? 5

[coreman73]

Hi all,

I currently have two ratchet cable cutter blades. The first was formed by hot forging of AISI 1070. The second was formed directly from AISI 1074 plate by wire EDM. During service, the forged parts are routinely cracking while the EDM parts are not.

I analyzed the bulk chemical composition of each sample and both matched to their respective material grades. However, the AISI 1070 blade additionally contained a Cr content of 0.134 wt% along with Cu of 0.167 wt%, which are not part of the 1070 specification.

My question is if the AISI 1070 sample was heat treated in standard fashion for this type of steel, would the Cr and Cu contents prevent proper heat treating? In other words, would the heat treat schedule have to be adjusted for the inclusion of these two alloying elements?

I would appreciate any help greatly.

Thanks!

 

[coreman73]

Ok, I have one more question regarding the material and then I'll drop it! haha Could there be a logical reason to explain why the 1074 blades (the EDM sample as well as some the vendor made by CNC) consistently outperform the 1070 forged blades? I've already established that heat treating is the same (or allegedly so according to the vendor).

Regarding the decarb layer on the EDM sample, it is only present on the top of the blade with continuation to the cutting edge tip. The entire bottom surface of the blade including at the cutting edge tip shows no such feature. Also, the decarb layer within the induction heat treated zone at the cutting edge is approxiately 10 HRC lower than the core area (~48 HRC versus 60 HRC). Within the regular heat treated portion away from the cutting edge, the surface hardness is approximately 97 HRB. The structure of the HRB layer appears as pure ferrite while a tempered marteniste structure is found in the decarb layer near the cutting edge.

 

[coreman73]

I have attached a photo illustrating the full length of the crack. After etching, you can clearly see that the crack extends only through the induction hardened zone and dead-ends at the transtion area. There is no sign of local damage or deformation at the entrance to this crack. There is also no sign of internal oxide inside the crack. The crack is very straight with no branching.

Based on this photo and the other information, is the concensus still that this is a quench crack?

The vendor confirmed that they only saw this crack AFTER the blade was used.

 

[dgallup]

I think the biggest problem is your heat treatment spec. 61 HRC is too high for this material. It is going to be very brittle at that hardness. You either need to go to a tougher grade of tool steel or reduce the hardness. 58 HRC would be more appropriate. As a general rule, a forged cutter should out perform a machined or EDM'd cutter if it is properly processed & heat treated.

 

[metengr]

Given the quench crack would have formed after induction heating and during ploymer quenching would not result in a noticable oxide build-up along the crack surface. Second, you should have broke open the crack surface prior to mounting for metallographic examination using liquid nitrogen to view the fracture surface versus mounting it whole.

Yes, it looks like a quench crack given the appearance and orientation. Also, just because the crack was not observed visually does not imply there was no crack. Surface NDT would need to be performed before use to confirm a crack along the edge.

 

[coreman73]

Thanks dgallup and metengr.

So, if it is a quench crack then the reason that it wasn't visible until after the part was used was because it simply propagated and became more prominent? Shouldn't there be some sign of surface damage/deformation at the crack site though or not necessarily?

metengr,
Where could I learn more about the liquid nitrogen technique? I have not heard of this before. The lab I work in doesn't have an SEM anyway so I would only be able to perform a macroscopic review of the fracture surface.

 

[metengr]

coreman73;
What this method does is allow for visual/SEM examination of an embedded crack. The crack containing region is extracted, and cooled in liquid nitrogen to induce brittle fracture, similar to a notched impact specimen. What you have is an exposed fracture surface for examination with little to no collateral damage. Obviously, this method works only for ferritic materials that exhibit a DBTT.

 

[redpicker]

It may be difficult to definatively determine whether these cracks are quench cracks. Quench cracks from in the brittle martensite as a result of the stresses resulting from the decomposition of austenite during quenching. The stresses are formed during quenching, not the cracks. The cracks form after quenching, which can be seconds, hours, days, or even months if stresses from the transformation are not relieved and martensite remains brittle. Tempering will both lower the hardness (and brittleness) of the martensite as well as relieve the transformaiton stresses.

[Disclamer] It is difficult to evaluate photomicorgraphs over the internet. Artifacts from sample preparation and the inability to examine the sample at higher as well as lower magnifications can cause such evaluations to be highly unreliable [/Disclaimer]

The photomicrograph you posted seems to indicate a wide crack existing sub-surface while near the surface, the crack appears to be very tight. This suggests that there is a variation in residual stresses within the hardened layer. Whether or not these residual stresses caused the crack to form, or merely assisted srevice stresses that lead to crack formaiton will be difficult to determine. The tight crack at the surface does suggest these may be difficult to see visually on new parts, however.

There are many processing variables that could lead to this. For example, if the forged parts were processed in the winter so they were very cold when the induction hardening took place but the EDM parts were much warmer when they were hardened. Or, maybe it is only the sharper cutting edge of the EDM blades that causes the difference. I still am confused why the EDM blade has decarb while the forged one does not.

Regardless, a tempering cycle of 1 to 2 hours at 350F to 400F might be all that is needed if they are, in fact, quench cracks.

rp

 

[TVP]

I vote quench crack too, for the reasons metengr elucidated.

 

[coreman73]

I appreciate all the help guys. Thanks metengr for the information on the liquid nitrogen method. I will have to find some way for someone to possibly demonstrate this to me.

dgallup or others,
I would agree that a tool steel should be a better option for achieving such a high hardness/toughness since that's what they're primarily used for. Is there any literature or graphs that would better prove that the 1070 steels aren't the best choice for manufacturing these cutting blades? I would like to present at least something to the vendor as another way to correct their issue.

redpicker,
Is it not simply possible that the EDM process actually created the decarb layer in the first place and that subsequent heat treating simply was not able to get rid of it? As for the tempering recommendation, these parts have already been tempered. Are you suggesting a double temper?

 

[dgallup]

Something like S7 will be much tougher. You can get some data sheets here:

http://www.simplytoolsteel.com/data-sheets.html

I would be tempted to try your forged 1070 at a more realistic hardness first. 57 to 58 HRc.

 

[redpicker]

I guess I am just having a hard time understanding how the EDM process would result in 0.010" of decarb. That seems excessive to have resulted from EDM processing.

If you have some unused parts from each process, why not get some NDT performed on them to see if any have pre-existing cracks? It may answer some questions.

Without knowing more about the application, it would be difficult to know what materials will out-preform the existing material. To make an informed decision, the cause of the cracking will need to be better understood.

rp

 

[swall]

I'm glad you brought that up, redpicker. I can't see how EDM would cause .010" either. But, I didn't want to burden Coreman73 with another issue to sort out, so I never mentioned it.

 

[coreman73]

Thanks dgallup. I will recommend them at least experiment with a lower minimum required level of hardness.

redpicker and swall,
So what would be the more expected depth of decarb from the EDM process? Also, since we're talking a bit more about it, wouldn't heat treating normally remove most of this layer anyway? I'm now beginning to wonder why it's even still there if the EDM was done prior to heat treating, which it was.

Do you guys have any explanation how this could be possible? I went ahead and attached a few photomicrographs illustrating this layer. You'll see that the depth is approximately 0.006-0.007" deep. Microhardness data reflected this depth as well but appeared to be closer to 0.010". This layer is nearly pure ferrite.

Etching did not reveal such layer along the induction hardened portion of the blade but microhardness did show it was still there (with higher hardness of course). In contrast, this layer appeared as tempered martensite.

I'm about to reach frustration level on this job! I may simply send my findings and be done with it as I hate to postpone sending out the reoprt much longer. hah

Unfortunately, I do not have more of these parts to send out for NDT. I will suggest this to the vendor though as another option.

 

[swall]

The decarb shown in the photomicrograph is consistent with what would be expected on a piece of hot rolled 1074 plate that had been subsequently austenized, quenched and tempered. Not sure where this leaves us.

 

[coreman73]

Thanks swall. Does this simpy leave us at the conclusion that this layer was a result of heat treating and NOT the EDM like I had originally thought?

If so then wouldn't this also suggest that the vendor was not correct and that heat treating was really not the same for both forged and EDM blades?

Any ideas/speculatoin are certainly welcome.

 

[swall]

I was suggesting that the decarb was from the original plate. Can't comment about the propensity of the heat treat process to cause decarb, as you haven't described the process.

 

[coreman73]

swall,

Unfortunately, I don't have details of heat treating but it was clearly stated by the vendor that the process was exactly the same for both forged and EDM parts. Keeping that constant variable in mind, it should be safe to conclude that heat treating couldn't have caused it.

I will simply have to suggest to the vendor then that this decarb likely came from the original plate, which does make sense.

Ok final question, and I will wrap this job up. As a theoretical question, is there a way that a pure ferrite decarb layer originating from wrought hot rolled 1074 plate could still be present even after heat treating? If so how? It seems like if austenitizing was done correctly this would be impossible.

 

[swall]

In order for decarb to go away, you have to add carbon via an enriched furnace atmosphere. This is known as "carbon restoration". Your layer of partial decarb is acting like a cladding of 1020 steel over 1074. If you re-austenitize and quench, you end up with the 1074 and 1020 that you started with, unless you add carbon and apply the appropriate amount of time for restoration. Carbon restoration can be kind of tricky, BTW.

 

[coreman73]

Thank you swall and to everyone else for helping me work through this job. I greatly appreciate it and have learned lots along the way.