Testing for Overheated Edges

[EOU]

Very useful and not confusing Geno so thank you. As we detailed in our description of the Tempilaq test, the procedure leaned toward giving belt sharpening every opportunity not to  overheat the edge. Just like all of you, we've seen metal turn cherry red during a grinding operation so we know that, given enough force and time while grinding, metal can get very hot. We did everything we reasonably could in the early phases of the test (outside of water cooling) to avoid overheating including beginning with a knife that was already sharp. We never really "leaned in" with much force nor did we maintain contact with the belt for long periods of time. In the later phases we did begin working, back and forth, over a relatively small area and that might tend to create some additional heat.  Generally, we would describe the final stages of the test as "normal grinding methodology" as if we were simply, "sharpening a knife". 

Hardening on the surface and softening beneath? Sounds like the fodder that many Knife Forum discussions are made up of and, possibly, could be quite accurate but, of course, very difficult to test for and validate.  This applies to all discussion of HRC levels of edge apexes because, once again, we can't actually conduct an HRC test at or near the edge apex. If someone has actually figured out how to accomplish this then we will stand corrected.

There is at least one variable in our test that might prove significant; our test knife was a cheap stainless steel model purchased at the discount store some time ago. We'd guess HRC55 or so. We'd could do better than a guess at hardness because we do own a Rockwell bench tester but Mark Reich won't fly down here and show us how to use it. Harder knives are harder to grind. A different steel and a higher HRC number could add up to a significantly higher grind temperature. Just how significant we don't know. Also, higher HRC knives are tempered, generally, at lower temperatures making the potential for altering hardness, it might seem, more likely. This would assume that tempering temperatures are among those that we are concerned with but the answer to that question is not yet clear to us.

Here's an unexpected benefit of our test and one that gives us additional confidence in our results; as you already know, the business end of a knife is the edge apex. This is the area of our concern and is only a very few microns in height and  much less in thickness. To observe melting Tempilaq in the immediate area surrounding a region so small is a questionable task even with the aid of a digital microscope. We were successful, during the grinding process, of raising a contiguous burr that was several thousandths of an inch in height (25.4 microns = .001"). The thickness, or thinness as you prefer, of the burr metal should approximate the thickness of the edge apex (therefore approximating the thermo mass characteristics of the apex ) but over a much larger and more easily observable area. The Tempilaq that was adhered to the backside of the burr steel remained opaque.  

Note on the above: We would have liked to have described the Tempilaq on the backside of the burr as being "clearly opaque" but that just doesn't sound right does it?

[Bud]

So your're back at it again EOU. I'd think you'd be tired of people spray painting hate messages on your garage door by now. I've never been worried about temperature while grinding on the Kally before. I take care not to overheat but maybe my edges get hot and maybe they don't. Not to mention that any knife I buy has already been power ground at the factory. I don't think that any big knife manufacture sits around with 500 people sharpening on stones in the back of their warehouse. I do know that my edges seem to last a reasonable amount of time and they are sharp enough for me. It also doesn't take much time to sharpen them either. Maybe ignorance is bliss.

This temperature liquid thing seems like a very reasonable experiment to me. I'm surprised that the stone and the belt grinder guys haven't had more to say one way or the other. Its sure been speculated about enough here and on several different occasions. If its a good experiment then maybe more of it should be done.

[KnifeGrinders]

Many findings on this forum draw knife people from the bliss of ignorance to the discomfort of voluntary ignorance.
Not everyone is willing to change his sharpening methods in view of new facts, let alone the equipment.
Even the BESS sharpness tester - I've watched it too often in the last Knife Show how knife sellers' attitude shifts from aspiration to hatred as soon as they realize how dull the knives they sell are.
I had accuracy of the sharpness tester questioned by end users when they struggle to improve their sharpening skills.

On the other hand, thanks to all the advance we've made here, I am delighted to get messages coming from all the corners how our followers who struggled to get their knives to 150 BESS start scoring 50 BESS, those whose knives dulled on dressing one deer now dress 10, or the knives sharpened in or workshop that could fail through one beef carcass now last through 5 etc

[Bud]

I'm a little different Mr. KnifeGrinders. I knew I had problems and a PT50 only confirmed my suspicions. I'm still not cranking out any 50's but I'm a solid 150 grams better than I used to be and taking half the time to do it. You've just got to figure out what works for you and what doesn't and when you find yourself headed down the right trail then stick with it until you're happy with the results. What told me I was on the right trail? A PT50. I'll tell the average sharpener this much, everything you need to know can be found right here from the good people on this forum and that's a fact.

[grepper]

I'm with you there Mr. Bud. Without a sharpness tester how do you really know? It's just a guess. It shaves well, or cuts paper or it is "scary sharp" or "wicked sharp" or whatever.

IMHO, Mr. Mike changed the landscape with his sharpness testing instruments. Now we can all be on the same page. Not only that, but we can communicate exact sharpness readings and everybody world-wide understands what it means. I doubt that I will ever in my lifetime see another actual standard being created, but that seems to be what has occurred. BESS is now a world-wide understood standard. Thanks Mr. Mike.

[Mike Brubacher]

Thank you for the encouraging words Grepper, Bud and KG but the credit for bringing this system to the cutting edge industry should go to you, all of our customers and the volunteer board members at BESSU. Our BESS Partners have contributed tremendously by printing BESS scores on their sharpened products in both the consumer and industrial market places. Without  this proactive support quantitative edge testing would have been no more than a momentary blip on the radar screen.  The BESS is now an accepted and universal gauge of edge sharpness by many and in use on every continent (yes that includes Antarctica). It does seem like a good idea to quantify edge sharpness levels doesn't it? That's what allows us to speak the same language and share edge sharpness information, without ambiguity, worldwide. 

Please remember this; you don't have to own an edge sharpness tester in order to understand and benefit from the BESS. All you need know is that a DE razor blade is 50 and that a 500 edge should be sharpened. With those two reference points in mind, be you shaving your legs, carving up a rib roast or shearing yarn in a textile factory, anyone can make a quick and accurate judgement regarding the suitability of any edge that displays a BESS score on it.  A very large thank you to all!

[Tman]

I had a good chuckle over this one from Bud "So your're back at it again EOU. I'd think you'd be tired of people spray painting hate messages on your garage door by now."  I have come to the conclusion that all edges roll as a result of your edge rolling tests (SET I guess) as a result of not only your tests but also from what I've witnessed here in the plant. I have to admit though that I have problems with this grind temperature test. We take such care here not to overheat during the grinding process. This applies not only to the edges we use in the production process but also the edges we ship to customers that are part of the products we manufacture and market. I'm willing to admit that there is grinding and then there is grinding but everything we do is multiple light passes and through most stages, water cooling. This applies to both our hard grinding wheels and our belt grinders. I'm just having problems believing that what we have been doing for years is just wasted motion.

I really wonder two things, (1) I'm really thinking that the temperature sensing product that was used in your test was just too slow to react. I see your reasoning behind your conclusion that even the sensor paste attached to the massive burr didn't react but it looks to me like there is much more sensor paste than burr so therefore more thermo mass. (2) I'm certainly not opposed to conducting this test ourselves. We don't temper to high Rockwell values any longer. The only thing it seemed to buy us was longer grinding times and chipped edges. Therefore, I would think that our edges might be able to handle even more heat without altering their hardness. Don't hold me to that theory though because it hasn't been researched enough.  Water cooling is a mess and we have to purchase special belts for the belt grinders in order to accommodate it. Maybe we have been wasting our time and money. I guess that you would have to describe me as open minded yet skeptical. 

In any case, thanks for putting in the work on this. I like the fact that you are not invested one way or the other in these results and that adds to the credibility of your findings. Your instruments have increased the quality of our products and decreased our costs so I'm more than willing to save/make some more money with you. I'll drop you a email soon and get your thoughts on test products and methods.

[EOU]

Well we got your email and then phone call TMan before we could respond to your post. We agreed, the temperature response time of Tempilaq while grinding edges is an unknown factor. We also agreed that the response time for Tempilaq could be described, at worst, as a very few seconds which led us to Geno's reference to Holloman-Jaffe and work done by the Germans regarding time, temperature and loss of hardness. All of this then leaves us still with two questions. (1) Is the grinding temperature much higher, albeit very briefly, than our Tempilaq test indicates?; and (2) If the temperature increase is greater than indicated by the Tempilaq is it sustained for a period of time long enough to soften the steel significantly?

We look forward to hearing from you once you have completed your experiments and thank you for your post.

[grepper]

(2) If the temperature increase is greater than indicated by the Tempilaq is it sustained for a period of time long enough to soften the steel significantly?

I too have that question.  If we were to assume there is heating, I would think that it would occur at the very thin apex of the blade.  This super thin metal would heat and cool extremely rapidly.  I think the whole process would occur in only a second or two.

I think that for all intents and practical purposes it is irrelevant for general sharpening, but sort of interesting nonetheless.

[Jan]

Critical parameters of heat-treating process are time, temperature and cooling rate.

Some phase transformations of the steel (e.g. tempering) are controlled by diffusion processes, which are strongly dependent on time and are relatively slow.
 
On the other hand, Martensite formation (by quenching austenite to room temperature), depends only on temperature and not on time. Fraction of a second is sufficient for this type of transformation.
 
I expect similar processes also during grinding. In this case we expose the edge to short heat pulses, which are probably not long enough to enable diffusion processes, but may locally rise the temperature sufficiently to initiate some shear-type  transformation of Martensite, which solely depends on temperature.
 
Jan