Failure analysis of cracked pliers 3

[coreman73]

About a week ago, I got word that some of our plier products were showing up cracked after heat treating. At that time, I was told there was reason to believe the cracking might have been related to a raw material defect or possibly introduced during forging. I was sent some raw material samples for evaluation. I found multiple sites around the perimeter of the 1070 bar stock showing rolled in scale defects. These were as deep as 0.017" and were fully decarburized. If needed, I can provide photos.

Rather than assume these defects were the root cause, I had some actual cracked pliers sent to me in heat treated condition. The attached file "cracked pliers" illustrates the results.

Cracking is only associated with parting lines. For some reason, these samples were treated by an acid bath after heat treating in order to reveal the cracks. I have no idea why this was done. I've never heard of this. Unfortunately as a result, as shown by some of the photos in the second attachment, there is both general surface pitting as well as pitting along the crack paths.

Hopefully the attachment provides enough information. For me, it seems that the cracks originated during forging, since they're found exclusively along the parting lines and the cracks show extensive scale. I don't see how the raw material defects could be responsible since cracking is not seen anywhere else except for these parting lines. What's a little confusing is that there are some deep internal voids adjacent to the crack paths that are filled with scale. If it's raw material related then I didn't find any such defects on the raw material I originally examined.

Any opinions on the root cause of cracking would be appreciated.

 

[swall]

Quench cracks along parting lines of forgings are fairly common. The corrective action is the same as for any quench cracking problem--reduce the severity of the quench and make sure there is no delay between quenching and tempering.

 

[coreman73]

swall,

By "quench cracks" are you referring to the quench after hot forging or the quench after austenitization related to finishing heat treatment?

If the latter then how could there be so much scale along the crack path and so deep?

 

[mrfailure]

I'd look closely at heat treatment. Core and surface hardness are much softer than nominal 58 HRC expected. You also expect no decarb or corresponding lower hardness at surface.

Parting line location of cracks may be purely the result of geometry - the geomoetric discontinuity of the parting line may make this the most likely first cracking location.

Aaron Tanzer

http://www.lehightesting.com

 

[swall]

I meant cracks formed during the quench phase of your hardening cycle. The reason quench cracks like to form in parting planes is that this is where all of the inclusions and other junk get squeezed to during forging. Along with the short transverse grain flow created in the thickness direction of the forging (i.e. perpendicular to the parting plane), you are setting up a metallurgical notch and this will be the preferred site for cracking, absent any macroscopic notches. The 850F tempering temperature is a little low to account for the scale observed in the crack, however. It may be that the 2+ hour tempering cycle plus residual chemical from the washing could account for this. I'm going to re-read your report and see if there is something I missed.

 

[swall]

One more comment--your observed hardnesses are consistent with a higher tempering temperature than 850F--perhaps as high as 1000F, which could account for the scale in the crack.

 

[coreman73]

Thank you for the responses swall and mrfailure. It seems that you guys both concur the issue lies with heat treatment of these parts.

I have another question. The cracks appear very wide/separated. Wouldn't this normally indicate the crack has been present longer than normal and maybe went through more process related stresses? Or maybe the acid bath opened the cracks more than usual?

Swall, that is an excellent comment about my observed hardness readings being consistent with a higher tempering temperature. I definitely hadn't considered that along with it accounting for more scale in the crack.

 

[redpicker]

Just for an alternate view, I'd look at the annealing treatment.

You say the process flow is:
(I assume Cold)-Shear 1070 blanks
Hot Forging
Annealing
Cold Trimming
Abrasive Cleaning
Edge Coining
Flat Coining
Machining
Heat Treat-Controlled Atmosphere, Oil Quench, Hot wash, 850F temper.
Expected Hardness 58 HRC (?)

Cracks can occur during the shearing operation if the mateial is not adequately annealed prior to shearing. If this was the case, I'd expect heavy decarb and oxides along the sides of the cracks resulting from the heating to forging temeratures. You don't have any photomicrographs in the etched condition, but I assume if there was heavy decarb along the crack, you would have shown it.

Similarly, cracks can form during cold triming of the flash, particularly if the parts have not been adequately annealed after forging. I would expect these cracks to form exactly where you are seeing them. They would be relitavely shallow, would not exhibit decarb (becasue of the protective atmosphere HT), but would show temper scale in the cracks from the tempering operation. In short, they'd look like what you have. What is the typical harndess after the annealing cycle? Can you investigate the particular annealing cycle these parts were given?

I would be surprised to see quench cracks to be this shallow. Generally, conditions that produce quench cracks in long, slender parts like this create mulitple pieces, or at least, run to the center. None of the parts seem to have pieces breaking off, or even coming close.

I question your 58 HRC hardness since I would expect 1070 material with a 850F temper to be in the low to mid 40's, exactly what you are seeing. Unless this is a modified chemistry, I don't understand this. I also wouldn't expect plier handles to be 58 HRC; I'd think that would be too brittle. The jaws induction hardened to 58 HRC, OK, but not the handles.

I think it would be difficult to quench crack 1070 in oil, unless they were left out is sub-zero temperatures in the as-quenched condition for an extended period of time. Assuming they went in the temper within an hour or so from getting out of the washer, I doubt they are quench cracks.

rp

 

[swall]

The big problem I have with a cold shearing crack or cold flash trimming cracking scenario is that the cracking appears to be, in part,intergranular, which would be inconsistent with cold cracking.

 

[coreman73]

redpicker,
Thank you for the post. It definitely opens up some more possibilities. I asked the engineer that sent me the samples to confirm the hardness. It looks like he gave me the tempered hardness instead of the final hardness of 38-44 HRC, which is what I got. I will ask for more information on the anneal but I don't believe the records of this particular anneal will be documented.

That's correct. I didn't see any sign of decarb along any of the crack paths. So it looks like that would rule out cracks picked up from cold shearing.

swall,
I will give etching the grain boundaries another shot to see what fracture mode occurred.

 

[redpicker]

Cracking from cold trimming can certainly be intergranular, depending on the microstructure at the time. Look at the crack tip in Fig 3 d). How do you get a quench crack to produce mulitple crack fronts like that? I wouldn't rule out cracking during the coining operations, either, but I think the cold trim is the most likely candidate, particularly since the cracks appear where the stresses would be the highest during the cold-shear operation.

Are all of the cracks on the same side of the parts? In the samples you have shown, they are all on the "top". Would this side go "down" in the cold shear? That is, would the shear blade come from the flat side of the plier to the side with the joint recess?

I'd expect to see quench cracks at areas of cross-sectional change or emanating from the holes. Some of these cracks run past areas of cross-sectional change, but do not seem to be influenced by them. In fact, the crack shown in Fig 2 b) seems to stop right before it gets to the hole.

coreman73, you seem to have a very interesting job. At least, the parts of it you post to this board. I'm jealous.

rp

 

[swall]

Redpicker--in lieu of large changes in cross section, quench cracks can occur at a metallurgical notch, such as the flash line, because as I previously noted, this is where all of the inclusions and other material inhomogenities show up. I frequently see quench cracks in hot coiled springs, made from large diameter bars, that start from seams. No cross sectional change in a round bar, just the metallurgical notch of the seam. These types of quench cracks tend to stop a short distance from the seam bottom rather than head to the center of mass. They do, however, tend to meander somewhat from the plane of the seam. I can't agree with you about i.g. cracking in a cold sheared edge. I see tensile cracks all the time in high strength spring wire (high Rc 40/low Rc 50 range)with a martensitic structure and they are quasi cleavage with a mix of dimple rupture. I agree that the drilled hole represents an area of interest--not sure yet where to go with that line of thought. I would also suggest that trimming cracks could have been present and that they did not manifest themselves until they cracked open further during quenching. Had they not opened up,they would not present any issues in the final product. But once they open up during quenching, you have a visual defect that can't be ignored.

 

[coreman73]

redpicker,

To answer your question about crack location, there are a couple of parts that are cracked on both sides along the parting line. From a second review of parts I received, cracks are definitely initiating from the parting lines. The drill holes don't have appear to have anything to do with them. The cracks actually terminate before reaching them.

I actually do have a very interesting job but am having to learn a lot on the job (as is clear from all my posts!). I was trained by a PhD metallurgist for a couple of years but for various reasons he was let go and I was put in charge of the lab. It's been rough. I do enjoy what I do though.

swall,

I will re-etch in a bit to maybe shed more light on crack mode.

redpicker/swall,

So basically at this point, the consensus could be the likely root cause of cracking is from cold trimming due to the localized stresses at parting lines? We could also say that there may have been inadequate annealing after forging that didn't sufficiently remove these stresses, which made this area more susceptible to cracking by the cold trim. Finally, quenching during heat treating further opened the cracks making them clearly visible defects.

As swall noted, we could account for the scale along crack paths coming from tempering at the higher temperature or even from the acid bath.

 

[swall]

We all agree that cracking started at the parting line.If cold shearing caused the cracks or initiated small cracks that opened up during heat treat, the problem could be either material hardness or tooling related. I would guess that hardness of not greater than Rc 25 would be desirable to prevent cracking. The hardness after annealing all depends on annealing temperature and cooling method. If the cracking is due to dull trim dies, improper die clearance or excessive mis-alignment of the top and bottom of the forging, these issues need to be examined. I think a metallurgist should start with the annealing process and resultant hardness, since that is easy for him to investigate. As to the flash trimming issues, I think I would query the forge shop about what trim die issues are important.

 

[coreman73]

swall,

I think we could rule out cold shearing causing the cracks since there is no sign at all of decarb along any of the cracks, which would have been introduced by the forging process. Wouldn't this be correct?

I have a separate question related to this. If decarb along a crack is introduced after forging, is it possible that carbon restoration during heat treatment (similar to how these parts were heat treated) could remove it? The engineer is really trying to blame the forging process for these cracks and it seems like I should be able to easily disprove this by telling him that it couldn't be possible since no decarb was present along crack paths.

 

[redpicker]

The primary purpose of the annealing operation is not to relieve stresses, but to soften the material. While relief of internal stresses will occur, the use in the process you have listed is to soften the material to allow the subsequent cold-work operations (trimming, coining, machining) easier to perform and reduce cracking tendencies.

One of the reasons annealing cycles can appear to be confusing is because the actual cycle used is very dependent on not only the alloy used, but also on the part size and the needs of the annealed product. Because it is an intermediate heat treatment, the annealing cycle is often not viewed as a critical step and often process controls are very lax.

I'd guess these parts received what is called a "process anneal" or "sub-critcal anneal", that is, ~1200F for an hour or so and then air cooled. If temperature uniformity was poor or some other condition arose which resulting is some of the parts not receiving as good of an anneal as others, you may see some developing cracks during subsequent processing.

I hope I have given you some ideas as to where to look. Good Luck.

rp

 

[redpicker]

Do you see any decarb on the ouside surfaces, away from the flash trim line? If so, then you aren't going to restore the carbon in the depths of a crack. I don't think you'd see much carbon resotration of a 0.70% C steel with a CP of 0.7 at 1550F in two hours.

It is possible these are forging cracks, but I doubt it. I'd expect forging laps at internal fillets or confined to the ends; finding them mid length, where the material is flowing away from, not where the material is flowing to, is what makes me suspicious.

rp

 

[coreman73]

redpicker,

Yes, I'm seeing roughly 0.007" decarb on the outside surfaces away from the flash trim line. Considering this along with still not seeing any decarb at all along crack paths pretty much rules out forging cracks entirely.

You have given me lots of valuable information to use as usual. I learn lots from you as well as swall and the other members that respond to my posts. I really do appreciate it.

 

[swall]

When I wrote "cold shearing", I was referring to trimming of the flash, not shearing of the bar prior to forging.Agree with redpicker on the unlikliness of carbon restoration of the faying surfaces of a crack.

 

[coreman73]

redpicker and swall,

Thank you both again for all the valuable information.