Thread Rating:
  • 2 Vote(s) - 2.5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Testing for Overheated Edges
#21
Great post.  Thanks for the effort and sharing EOU.  Your micro images seem to be improving a bit.
 
These results don’t surprise me as I’ve never noticed much of any heating of the blade when I sharpen.  That said, I do have a very light touch when sharpening.  I'm open to the idea that heating occurs, but the evidence now seems to indicate that it probably is not a problem.
 
In eager anticipation of this test I posted in the Knife Making and Bladesmithing forum questions about the time and temperature required for tempering.  I was curious because I thought it possible that the edge could possibly warm into tempering range.  However, I reasoned that when judiciously sharpening, even if the blade did manage to reach tempering temperature, it does so for a very ephemeral period of time and would not hold temperature long enough to change the temper of the edge.  Regardless, so far, the Tempilaq tests apparently indicate that temps are not reaching even the lower end of tempering range, and I postulate that even if it did reach tempering temperature, the short duration of those temps would render the effect on the blade insignificant. 
 
Obviously enough pressure was applied between blade and belt to create a burr so we know that a blade can be sharpened on the Kally without significant heating.
 
I did some ferreting around and found some non-scientific observations stating that when applied to thin surfaces, specifically brass cartridge casings, that Tempilaq may be a little slow to trigger resulting in the material the Tempilaq is applied to may be at a slightly higher temperature than the Tempilaq indicates.  Care must be taken in understanding what is being said as the Tempilaq was applied to the inside surface and heated from the outside. Part of what is being discussed is thermal migration through the brass and not specifically how fast the Tempilaq reacts once the surface that Tempilaq ia applied to reaches trigger temperature:
 
http://www.65guys.com/cartridge-case-ann...equipment/
 
Here is some additional information regarding the properties of cartridge brass. This reference shows the annealing temperature to be between  800-1400 degrees F. If that’s the case, why is 750 degree Tempilaq recommended? We spoke with one of the annealing machine manufacturers and they explained that the reaction time of a temperature measurement system will be slower than actual temperatures achieved for something as thin as a case neck. As a result, the indicated temperature will be exceeded.
 
http://forum.snipershide.com/threads/6mm...ble.61253/
 
“Now, brass isn't a perfect conductor, it takes time for it to heat up and for the grain structure to change once it does, but this still probably has a lot less of a lag to it than the time it takes for the heat to transfer to the Tempilaq and melt, which is much less thermally conductive than the brass. So, if you're using 750 F Tempilaq inside of the neck, the brass has probably exceeded 800 F when its only started to melt and has shot quite a ways past it when its fully melted. I saw this. If I put 750 F Tempilaq and 850 F Tempilaq in the same neck, it was not possible to get the 750 F Tempilaq to fully melt or even partially melt without the 850 F Tempilaq starting to melt. 800 F Tempilaq surely would have melted to some point in between. Parts of the brass itself were likely over 900 F. I could totally melt the 850 F Tempilaq without fully annealing the brass. I have to go a bit longer to succesfully do it.”
 
Even taking the above into consideration, I would still postulate that if parts of the blade were indeed slightly hotter than 325° it would not be excessively so, and the period of heating so short as to be of little significance.  Furthermore, this experiment clearly indicates that the test blade did not get anywhere even close to annealing temperatures let alone tempering temperatures.
Reply
#22
Mike, thank you so much.
We know that the edge properties do change during grinding, we know that the mechanical component is always contributing, and your tests are to show whether the annealing ("detempering") component can be nulled by careful grinding.

Would you be able to repeat the same test on a finer grit belt or a buffing wheel?

It's a common knowledge that coarse grits grind cool.
For this reason, for example, grinding paper wheel is not made finer than #180, because with finer than 180 grit not enough grease and air get between the abrasive grains to cool the edge you sharpen, and it burns.

The majority of sharpeners deburr using near 5 micron compounds, i.e. JIS #3000 or ANSI #1500.
For completeness, would be great to also test JIS#1000 or ANSI 600-800, as this grit is common in fine grinding.
If you see no heat produced by that fine grit abrasives in your testing, the results will become conclusive for the whole of the practical range of grits.

For protocol, what is the belt speed in your experiment?
I assume you don't observe sparking while grinding? Or do?
http://knifeGrinders.com.au
Reply
#23
Can’t answer for EOU Mr. KG, but I have used extremely fine (1000 – 5000 grit) belts with the Kally and don’t notice much heating at all. As I have said, I use very light pressure when grinding. That’s not to say heat can’t be generated with the Kally. I can get stuff really hot with the Kally by applying a lot of pressure. Grinding off the head of a bolt against the platen works well for generating heat even with 60 grit Zirc belts. Way too hot to touch. But then I can tap an egg with a hammer and not crack the shell, or obliterate it into a disgusting mess with more pressure. Just depends on how the tool is used.

The thing is that when sharpening with light pressure the blade is in contact with the belt in any single spot for a very short time and with little pressure. Any one spot for what… maybe a second? Not very long at any rate, and then the blade cools in open air. At least I think that is what is going on.

The Kally runs at 1725 RPM. The 1/3 HP motivates the belt to 548.64 m/sec (1800 ft/sec).
Reply
#24
Forgot to mention:  I NEVER see sparks when sharpening.  Grinding the head off a bolt - lots of sparks, but never when sharpening.  Two very different operations.
Reply
#25
Thank you Grepper, this all is meant between us, but is worth mentioning aloud that we are talking about spark-free judicious sharpening with no or very little warming of the blade felt by fingers.
I am excited in anticipation of Mike testing the finer grits, as the negative result is as important as positive.
http://knifeGrinders.com.au
Reply
#26
Totally understand. Fingers are one thing, Tempilaq far more accurate. I have no doubt that 325° Tempilaq can be triggered by applying more pressure/time.

I too am sleepless, waiting with baited breath for further Tempilaq testing, and have also wondered about finer grits. I also believe that Mr. Mike is considering testing the fences with lower triggering temperature Tempilaq.
Reply
#27
Mike, in the coming experiments would you be able to apply a thinner layer of the thermal lacquer, as its response must be mass-dependent.
http://knifeGrinders.com.au
Reply
#28
For those wishing to do additional testing themselves with Tempilaq it’s not hard to find:

McMaster-Carr:
https://www.mcmaster.com/temperature-indicating-liquid

I don’t know anything about the best places to get stuff in Australia, but for what it’s worth I found several places, among them:

http://www.temperatureshop.com.au/tempil...ting-paint
Reply
#29
Thanks KG and Grepper. Yes we think that reaction time could be a factor of some significance. We did test that variable at some level by placing a pyrex glass dish in the oven at 350°F for about twenty minutes, removed it from the oven and painted on a streak of the 325° Tempilaq. The Tempilaq triggered about as fast as we could paint. So given enough thermo mass and thermal conductivity the stuff goes off pretty quickly. 

We didn't thin the Tempilaq in our experiment (thereby supposedly yielding a faster reaction) at all and something that could be tested in the future. We think that this is an experiment for the masses. It's cheap, it's fast and it's easy. KG is right. Finding the fail point is important as well. Around here we call this "locating the fences". This then would let individual sharpeners know just how much margin they actually have. I think that if we were going to conduct another experiment we'd get a bottle of 250°F and see what happens under the same circumstances. Kwakster will be using some even lower temp stuff so it will be interesting to see what he finds.

In the arena of having fun and satisfying curiosity we made a few passes on the edge with a Scotch Brite belt and tested the edge in the area where we had conducted the tests. BESS 135. From the micro pics we took we think it could have stood a little more deburring but not too bad. In our experience chisel grinds test sharp. Sharper than double bevels quite often. We weren't quite as successful at converting the test knife into a genuine chisel grind as thought. After we cleaned it up we could see that some bevel still remained on the back side. Here's a couple of pictures, grind and backside, of the cleaned up test area. The lopsided width of the bevels are pretty obvious.

                                                                   

Grepper raises some good questions concerning the relationship of temperature and time and then the speed with which cooling occurs. If the steel nevers get to a temperature where any loss of temper effects can come into play these questions may be moot from a practical standpoint but if it does prove necessary to hold steel at a certain high temperature for a significant amount of time before deleterious effects occur, it could provide additional peace of mind for sharpeners.
Reply
#30
So first of all sorry for the delayed answer time did not allow to write something.

No I am not a member over there but an avid reader for quite some time.

As for the Tempilaq experiments I remeber reading about some simmilar tests or reports thereof but cant find it at the moment (Will search for it).
Here is what I remeber on top of my head (may be completely off): The results where not pleasing to them as the material was ground away and the adjecent material was too far off to get a reading of the temperature on the grinding surface. This observation was then taken in relation to older research which I already linked where on page 19(17PDF) is a graph of a fast running disc (320FEPA @ 19m/s) with watercooling that shos peacs of temperature of up to 1200°C up to a depth of 59µm and that after ~1sec that temperature is already fallen as low as 450°C. Now this is an exptreme excample but shows how high and fast temperatures can become. The research also describes how the generated temperature heavily depends on how much material is removed (preassuer applied) in a given time Page 29(27PDF). Here different media wer used with diferent binders and grits: 320grit sic with ceramic binder, 50/320grit sic with magnesia binder and aio 280/320grit with magensia binder. The graph shows that the aio ground the coolest with temperature as low as ~450°C while taking away 20µm @ 19m/s uncooled grinding.
A Page further in the same was done with coolant (water) which shows a given the right media even fast grinding can be done below 200°C.

On the "detempering" it was argued that the most surface layer gets so hot that there is a rehardening effect going on which makes it harder but the layer directly below is softened up so that the hardened layer has not enough support to hold up and just dulls/crumples away quicker.

General consensus on the german board is that it is better to watercool while belt sharpeing but it is also acceptable to grind uncooled when the belt is run slow and only light passes get taken with cooling the knive off in between.

I hope this does not spread more confusion then it aleviates ...
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)