Belt Sander VS Edge Temper

I just finished a knife from a file a few weeks ago. 1/8" thick and unsoftened, except for the handle to drill holes. The teeth were removed w/ a bench grinder, but the bevels were ground w/ 40 grit ZrO belts in a 1x30" Harbor Freight sander. I wont be doing that again. I also made a knife from O1, 1" wide 1/8" thick, full flat grinds, but the blade was only 4" long. That still took a couple of hours. Hopefully I'll be trying some more tomorrow, but we'll see.

On the topic of the post, I noticed a good jump in edge holding on my Kershaw Vapor after sharpening on the HF 1x30. I have also burned/ruined about 6 knives on the sander, so some practice is required. I also find the belt sander w/ narrow belts to be the easiest way to sharpen recurve edges. I may have to get another BM 710 now that I can sharpen the blamed thing. I had previously sworn off recurve edges. For touch ups on the Vapor, I use the sharpmaker stones freehand. My angle control w/ the sander is a little off, and using the 20 degree slots on the Sharpmaker make the bevels look a little funny.
 
Jerry Hossom said:
Any reputable peer-reviewed publication would require you to completely describe the exact method used to determine that your estimate of 250 gram of pressure on the knife edge on the stone is accurate.
Click to expand...

I have published in many such journals, this is complete nonsense. You would be required to describe the method in detail if you have origionated it. This I have done in the several ways I have developed to measure sharpness and other people are using said methods and have independently reproduced the same results I have done, and expanded the work. Such as :

initial_sharpness_M2_34deg.gif


This is from Steve Elliott who used the push cutting sharpness test I developed to show the effect of higher polishes and different abrasives in a quantitative manner and as well to measure the extent of edge blunting. He expanded this with photographs of the edge as it wore. Very interesting work on plane blades, well documented.

http://cablespeed.com/~sgelliott/blade_testing/index.html

There are a few guys doing very similar work. By using these methods of measuring sharpness and cutting ability I have personally replicated curves and shown behavior which agree with independent sources such as CATRA data and I did it before the CATRA data was published. I have also used independent data to show that the models I have developed are true not just for the data I collected but are generalities. Such as :

http://forums.swordforum.com/showthread.php?s=&threadid=71299

Note this is a precise comparison of a flat ground vs convex blade where the flat ground blade offered significantly better cutting edge liftime as well as high initial cutting ability. The effect was so dramatic that a 420HC blade outperformed both ATS-34 and BG-42. All were Bos heat treated. The reasons for this I have described previously in this thread. It is a matter of cross section and was discussed on rec.knives almost 10 years ago. This was well before you were even using convex edges.

Dog of War said:
I don't understand, Jerry. You're asking this of others, but when I asked you about the testing that you were involved in with convexed wood chisels, you didn't have anything specific you could provide to explain or back up the claims. Instead, you suggested I was calling you a liar.
Click to expand...

This is standard form, use data to support a viewpoint which is extremely vague to the point of being undefined and have completely undefined references like "other makers" "some guys", but yet place extreme demands on everything and anything which is held as a point of contention. The bias is clear and transparent. I am all for being demanding and critical of data - ALL DATA, not just that which is in opposition to your point.

The basic question raised in this thread is can you use a belt sander to sharpen an edge and not induce temper draw. The answer is yes. The misinformation about other issues such as the behavior of convex vs flat comes from makers like Hossom never defining his edges, which is ironic considering his last post. As I noted recently to Will York, you need to constrain the geometries you can say for example "A circle will have a greater area than a square if both have the same perimeter".

The problem is that the people who are promoting general curvatures never define them. The easiest way to understand why this is necessary is to consider that any convex edge can be considered to be a modification of a flat ground edge which is more obtuse (thinning the shoulder) or of a flat ground edge which is more acute (increase the apex angle). The convex edge will have the exact opposite relationship to those flat ground edges. It will in general (very general) cut better but be less durable then one but cut worse and be more durable than the other. Thus comparing convex vs flat is undefined you have to constrain the geometries.

-Cliff
 
Getting a blade too hot (changing color) does not necessarily mean the knife is ruined. I've recovered a few by stripping the knife to the blade and re heat treating. I don't currently have a heat treat oven so I use either the molten lead or "sand and 30 weight motor oil" techniques for final temper. Result is a really nice spring steel condition that will both take an edge and resist shock. I'm a gunsmith and often use either method when I fabricate flat springs and some other parts for guns when there's no other way. I have some old six shooter hammer springs I've made out there that have lasted for years of hard use by the action shooter types. I usually temper those in molten lead and use the sand and oil for firing pins and such. Point is, also works for knives and other edged tools. Best regards to all.
 
Cliff Stamp said:
Originally Posted by Dog of War
I don't understand, Jerry. You're asking this of others, but when I asked you about the testing that you were involved in with convexed wood chisels, you didn't have anything specific you could provide to explain or back up the claims. Instead, you suggested I was calling you a liar.
Click to expand...
This is standard form, use data to support a viewpoint which is extremely vague to the point of being undefined and have completely undefined references like "other makers" "some guys", but yet place extreme demands on everything and anything which is held as a point of contention. The bias is clear and transparent. I am all for being demanding and critical of data - ALL DATA, not just that which is in opposition to your point.

The basic question raised in this thread is can you use a belt sander to sharpen an edge and not induce temper draw. The answer is yes. The misinformation about other issues such as the behavior of convex vs flat comes from makers like Hossom never defining his edges, which is ironic considering his last post. As I noted recently to Will York, you need to constrain the geometries you can say for example "A circle will have a greater area than a square if both have the same perimeter".

The problem is that the people who are promoting general curvatures never define them. The easiest way to understand why this is necessary is to consider that any convex edge can be considered to be a modification of a flat ground edge which is more obtuse (thinning the shoulder) or of a flat ground edge which is more acute (increase the apex angle). The convex edge will have the exact opposite relationship to those flat ground edges. It will in general (very general) cut better but be less durable then one but cut worse and be more durable than the other. Thus comparing convex vs flat is undefined you have to constrain the geometries.
Click to expand...
Cliff - of course I didn't really put that question to Jerry expecting an answer. But you raise some interesting points .... in fact I'd like to thank you if I haven't before for all the great info you post - I've learned a lot.

Some time ago I took a few moments to think about and visualize convex vs. flat grinds, and it became obvious that if the final edge angles are the same (disregarding microbevel if any) a convex grind has to have greater relief than a flat grind. So right there's the performance difference, if any, that I think people observe. Or the biggest part of it IMO. I'm not sure there's much need to further define the exact geometries being compared, unless you wanted to compare blades with different final edge angles. (BTW I particularly cringe whenever I hear the claim that "convex edges(grinds) hold up better because there's more metal behind the edge to support it" when of course just the opposite is true, if the final edges are the same angle. Maybe people have a hard time seeing how both hollow and convex ground both have greater relief than flat ground?)

None of which is to dismiss convex grinds or edges .... extra relief IMO is generally a good thing, plus convex edges can be easier to sharpen and maintain with simple equipment. Besides at the relatively acute geometries I prefer, there's hardly any difference between an edge or grind deliberately convexed, and a flat grind/edge done offhand.

As for the original thread topic, I've seen a lot of expensive tools reground and rough sharpened on grinders, and for someone who has the patience and the skill, ruining the temper isn't a concern .... and these are guys who work with wood, knowing only enough about steel to do their jobs. Just because many people haven't developed the needed skills to reprofile or sharpen on a belt grinder - skills I personally don't have - doesn't mean anything. Few people can make use of the inherent accuracy of a custom benchrest rifle, it doesn't mean there's a problem with the rifle.

OT ..... is there any way when quoting on these threads to have previous quotes automatically "embed" in the new quote? (I have to enter the older quoted text cut&paste, then the html-ish flags to quote it.)
 
"

Dog of War
said..

because there's more metal behind the edge to support it" when of course just the opposite is true, if the final edges are the same angle. Maybe people
have a hard time seeing how both hollow and convex ground both have greater relief than flat ground?)

I think that is probably a good question. How do you determan what the angle of the convex edge is? Is it the angle of the knife that is held to put the convexing on? Or is it the angle measured at the very edge? and really isn't the angle at the very edge a 0 angle?
 
db said:
I think that is probably a good question. How do you determan what the angle of the convex edge is? Is it the angle of the knife that is held to put the convexing on? Or is it the angle measured at the very edge? and really isn't the angle at the very edge a 0 angle?
Click to expand...
If I needed to measure what I'm calling the "final" edge angle, which would be the angle of the tangent (I think) on the curve right at the very edge ..... I think I'd just send it to Cliff! ;) Seriously, I'm sure it can be measured, but probably takes equipment most of us don't have.

If you're sharpening the blade, then you can control the final edge angle just by not raising the blade above that angle against the stone, belt, or whatever (of course if you're using a slack belt, or paper on mousepad, that's not going to work too well.) What I've done quite a few times when thinning out a convexed blade is to reprofile it like a flat grind, and then convex behind the edge - and that really proves the point, because you're removing metal to make the edge (or entire grind, depending) convex.

Pretty much it's more a theory "thing" .... just visualizing the profiles of two blades, one convex and one flat, as if you were looking at them from the tip. If the angles at the very edge are the same, the convexed edge has to be thinner as you move up from the edge.

It took me a good while to see this, BTW .... for some reason a convexed edge or grind "looks" like it's got more metal behind the edge, when actually the opposite is true if the angle of the final edges are the same.
 
Just occurred to me, to measure the final edge on a convexed blade you could do something like the "magic marker trick" (I prefer machinist's blue) .... just mark up the edge, prop up a Sharpmaker or crock-sticks rig at a really acute angle, then start lightly stroking the blade against the rod on one side at gradually increasing angles. Checking with a microscope after each stroke, keep working until you see when the rod is removing marker or bluing right at the edge. Then measure the angle of the Sharpmaker rod. Repeat on other side.
 
I'm still not really sure you can compare a curve and a flat bevel as a same type of angle. wouldn't the flats of the rods give a false measurement to the curve?
 
db said:
I'm still not really sure you can compare a curve and a flat bevel as a same type of angle. wouldn't the flats of the rods give a false measurement to the curve?
Click to expand...
I'm sure the Sharpmaker 'method' would produce pretty crude results, but in theory it's sound, I think .... where the rod removes the marking is the tangent to the curve at that point. So you could map the curve of the blade or edge's convex profile. Difficult with home workshop equipment of course....

Just speculating a bit ..... where it might become really interesting doing these comparisons would be cutting performance on different materials that bind. I think you might find a few situations where a convex grind with a more obtuse final edge cuts through some binding materials as easily or better than a comparison flat grind, due to the convex's greater relief. However you're probably not going to see that on all materials, so there'd still be plenty of debate.

Very interesting to consider, db, thanks for the discussion. Sounds like it might be an overwhelming project to do testing like this, but if you're thinking about it I'm really looking forward to the results. :)
 
I'm not sure I'd be the best one to try and test this. I'm pretty much putting convex or convexing most of the knives I carry because my sharpening method is free hand on a bench hone. I could try and find my old Lansky to get a flat bevel edge but I'm not sure where it is. Not real sure how a person would run a test to compare the two different bevels anyways. I'm still trying to really understand how to compare flat and convexed.
 
db said:
Not real sure how a person would run a test to compare the two different bevels anyways. I'm still trying to really understand how to compare flat and convexed.
Click to expand...

You could always measure the width of the edges with a micrometer and use Pythagorean's theorem for the equivalent flat bevel. You could also put various materials which you cut on a cutting board on a scale or forceplate. Phil Wilson recently did that to compare some steels he normally uses with a steel he normally doesn't use. You could compare the amount of force needed for cutting various media with both sorts of edges and how much they change over time (edge retention more or less).
 
Thom it's not the cutting part I don't understand. It's the Pythagorean's theorem part I'm having trouble with. :)
 
I know you guys are really serious about all this, but don't you think it's getting a little ridiculous?
Bill
 
Of course it is! Sharpening is mostly about getting the edge thin enough to cut what you want without causing unreasonable* damage to the knife. All of the rest is swapping tips, recipes, and myths. If we can use calipers, strain-guages, and real-life versions of Patrick McManus stories to have an argument, we will.

* = this part will vary from person to person. :)
 
You must log in or register to reply here.
Back
Top