Freehand knife reprofile/set new bevel using 120 grit ceramic belt

BluntCut MetalWorks

BluntCut MetalWorks

Knifemaker / Craftsman / Service Provider
Joined
Apr 28, 2012
Messages
3,462
It took 20 minutes to set new bevel and then sharpened 2 knives to 220 w/d sandpaper + stropped(balanced-strop) to newsprint slicing sharpness. Sorry, camera battery died around 8 min 25 secs (:grumpy: didn't noticed). So it only captured the new bevel setting for knife1. Good/long/bore enough, I guess :D

Knife 1: 14C28N steel, subzero ht, ~61rc
Knife 2: 1084 ht ~59rc
Belt: Norton Blaze 2x42 120grit ceramic

Recently, it took me less than 3 minutes to set new bevel for a new D2 knife with similar profile. K390 blade took twice as long; however it was a bigger knife.

Thanks for watching...

[video=youtube_share;uI4e5LDqlDo]http://youtu.be/uI4e5LDqlDo[/video]
 
Here are bevels of my 2 WIP knives I did in the video
p0ck.jpg

5o12.jpg
 
Well, I have another WIP knife, so here is a new video. Thanks for watching & comments.

Knife: 3"(75mm) 15N20 carbon
Belts: 120 Norton Blaze, Norax 1K structured abrasive
Strop: Yute rope + white compound (aka balanced-strop)
[video=youtube_share;HpWZ8NhuWco]http://youtu.be/HpWZ8NhuWco[/video]
Bevel pic
qjo1.jpg
 
Chris "Anagarika";12764714 said:
Bluntcut,

Nice knives there!
Click to expand...

Thanks Chris! These knives are part of my test set.

I am almost complete making a set of knives for sharpening & performance tests. Some are done, some are just sharpened WIP. There are so many variables at play in performance test: Steel+ht+profile+edge/bevel geometry+ease of sharpening+etc... So I end up make & break knives to learn, perhaps I can find out a reasonable trade-off range where performance is good for wide range of user and usage.

Here is my knives test candidates (rc are guestimated based on heat-treating charts):
Done: cpm-m4@65rc, k390@63rc, cpm-154@61-62rc, s35vn@61rc, 52100@62rc
Sharpened WIP: k110/d2@61rc, 1084@59-60rc, 15N20@60rc, 14C28N@61rc
Production reference: Gayle Bradley cpm-m4, BM940 s30v, Stretch cf zdp-189

Hopefully I've enough palm fronds to perform cutting tests. 5/8" or 3/4" hemp/manila rope would be better, so don't hesitate send me some rope ppl:cool:... There are so much we can learn/study about grain;carbides;bevel+apex geometries;force vectors;ergonomic;so on. Either that or I just wasted a lot of time:confused::o
 
It is a joy to watch you sharpen, Bluntcut! I have one question though, at 8 minutes it looks like you run the edge 90 degrees to belt. What are you doing there? Are you working on the burr or is it something else you are doing.
 
Really like those knives, nice work!

How long do the ceramic belts last under that sort of use, or how long till they noticeably start to loose some steam? Have made many a block for holding sanding belts - works great.
 
Squashfan said:
It is a joy to watch you sharpen, Bluntcut! I have one question though, at 8 minutes it looks like you run the edge 90 degrees to belt. What are you doing there? Are you working on the burr or is it something else you are doing.
Click to expand...

Thanks.

~8:00 in the first video:
I was refining+minor deburr the bevel+apex by using violin stroke (still edge-leading but walking forth slowly). Violin stroke where the direction is perpendicular to the belt/stone as oppose to normal stroke direction parallel to the belt/stone. Its effective grit is much finer because abrading vector (scratch) is longer on the bevel face, so if pull/push this violin forward, you could get edge result of 300-400grit finish in spite of 120grit sharpening surface. If coupling this stroke with less pressure then conceivably you can get an effective grit in 1k range.
 
HeavyHanded said:
Really like those knives, nice work!

How long do the ceramic belts last under that sort of use, or how long till they noticeably start to loose some steam? Have made many a block for holding sanding belts - works great.
Click to expand...

Thanks Martin/HH!

Here is my 80 grit ceramic stone (3 1x30 sections gorilla glued) I made 6 months ago. It's still cuts fast & deep, although it feels smooth to touch. I've used on a lot of knives, garden tools.
IMG_0800.jpg
 
Bluntcut,

A lot of work ahead, and the challenge would be keeping several factors constant while changing others and repeat... I imagine the combination (even one steel type) is daunting..

But I guess you've done some single steel test on 52100 :thumbup:

Good luck and have fun!
 
Last edited:
bluntcut said:
Thanks.

~8:00 in the first video:
I was refining+minor deburr the bevel+apex by using violin stroke (still edge-leading but walking forth slowly). Violin stroke where the direction is perpendicular to the belt/stone as oppose to normal stroke direction parallel to the belt/stone. Its effective grit is much finer because abrading vector (scratch) is longer on the bevel face, so if pull/push this violin forward, you could get edge result of 300-400grit finish in spite of 120grit sharpening surface. If coupling this stroke with less pressure then conceivably you can get an effective grit in 1k range.
Click to expand...

I appreciate the explanation but what do you mean by "walking forth slowly"? I think it means making a short, slow, edge leading violin stroke. I understand every thing else you said though. Thanks for the help!
 
Squashfan said:
I appreciate the explanation but what do you mean by "walking forth slowly"? I think it means making a short, slow, edge leading violin stroke. I understand every thing else you said though. Thanks for the help!
Click to expand...
You've got it. Just my poor choice of words in describing - edge lead violin stroke so the blade is slowly moving along the belt.
 
Chris "Anagarika";12767444 said:
Bluntcut,

A lot of work ahead, and the challenge would be keeping several factors constant while the changing others and repeat... I imagine the combination (even obe steel type) is daunting..

But I guess you've done some single steel test on 52100 :thumbup:

Good luck and have fun!
Click to expand...

Appreciated! Oh maybe you can volunteer to help with some tests, right :thumbup:
 
Chris "Anagarika";12772602 said:
Sure. Just need more detail on what kind of test will satisfy (proof or disproof) the hypothesis. Tell me more :)
Click to expand...
I sure don't have a fully-baked plan for this yet :o and now being side-track by performing mission critical server disaster recovery for a client.

Sort of side-tracking my own thread. I am open to idea to a test plan, so what do you/all think, we could/should prove/proof/learn/conjecture?
 
bluntcut said:
I sure don't have a fully-baked plan for this yet :o and now being side-track by performing mission critical server disaster recovery for a client.

Sort of side-tracking my own thread. I am open to idea to a test plan, so what do you/all think, we could/should prove/proof/learn/conjecture?
Click to expand...

Let's start with some assumptions or guesses:
- The knives are mostly done with the HRC stated above
- Purpose of test would be edge retention based on steel and hardness

Method of test (easier than Jim's)
- Sharpen all to same edge geometry and finish (i.e. 1000grit?) at low angle (acute edge)
- Cut/slice palm fronds (or rope) 10x, slice newsprint, repeat until it starts tearing
- Repeat 3 times for each knives .. after resharpening
- Change geometry for those rated poorly and retest

Challenge when inviting 3rd party test:
- Getting same sharpening angle (unless using same jig), and finish
- Getting same slicing method (force, speed, length of edge used)
- Getting same cutting board (it may influence dulling, depending on the material)
- Getting same rope and newsprint (I noticed in Singapore, the newspaper uses different kind of paper than down here)
- Getting same observation if the tearing has started

Hope this helps .. and your client's server is up and running again :)
 
Chris "Anagarika";12777956 said:
Let's start with some assumptions or guesses:
- The knives are mostly done with the HRC stated above
- Purpose of test would be edge retention based on steel and hardness

Method of test (easier than Jim's)
- Sharpen all to same edge geometry and finish (i.e. 1000grit?) at low angle (acute edge)
- Cut/slice palm fronds (or rope) 10x, slice newsprint, repeat until it starts tearing
- Repeat 3 times for each knives .. after resharpening
- Change geometry for those rated poorly and retest

Challenge when inviting 3rd party test:
- Getting same sharpening angle (unless using same jig), and finish
- Getting same slicing method (force, speed, length of edge used)
- Getting same cutting board (it may influence dulling, depending on the material)
- Getting same rope and newsprint (I noticed in Singapore, the newspaper uses different kind of paper than down here)
- Getting same observation if the tearing has started

Hope this helps .. and your client's server is up and running again :)
Click to expand...

Chris, excellent suggestions + thoughts - thank you:thumbup: I don't think, I have enough consistence-age palm frond for 3x sampling + delta params tests. I am thinking along the line (keep it simple) of wear resistance test where each blade steel configure for optimal cutting until fail to slice paper. It's not about efficiency test like Jim Ankerson rope cutting thread. Here is an engine(car) and load analogy. Load=tow a 1ton boat (cutting palm frond). Use best RPM per engine type (edge angle & maybe geometry). While in Ankerson test, all engine run using same fuel (30* inclusive) and at either 1K(coarse) or 8k rpm(super fine). Diesel engine probably will sucks most of the time - heheh my nitpicking again. I plan to zeroing the 'Optimal angle' per particular 'Load' for each steel.

The broken server is punishing me for the heck of it. Can't even get a new installation of Hyper-V2012 up and running, how sad is that :barf:
 
I would think you'll need a fixture to hold the knife and drive it with a similar or identical amount of draw per divisions of drop, and a consistent rate of travel/pounds of force. A simple rig would not be too bad, an adjustable one could be real challenging.
 
Hmmm .. I just did lookup on Hyper-V2012 .. never heard of it ..

It seems IT guys are into knives (Dagon, you and me/ex-IT as example :p)

The variables seemed to be so many, but your aim is to find most optimum edge geometry for each steel & HRC. Still many runs is needed, but the consistency requirement might just go simpler ;)
We're really OT here, but it's your thread .. your call.

@Martin, you're not sleeping? :eek:
 
Last edited:
HeavyHanded said:
I would think you'll need a fixture to hold the knife and drive it with a similar or identical amount of draw per divisions of drop, and a consistent rate of travel/pounds of force. A simple rig would not be too bad, an adjustable one could be real challenging.
Click to expand...

There is idea - yet another good one :thumbup: - Thanks Martin!

but implementation seems to get in the way. Even CATRA test performed poorly for blades with thick-behind-the-edge, especially when testing for wear resistance. Divisions & Forces consistency are good-to-have but not a must-have because IMO variations in D & F are insignificant compare to load-impact between edge & cutting media. Back to towing 1ton boat analogy, you are naturally seek optimal torque per velocity. Where under-powered vehicle will died quickly or easy cruising for 7.3L turbo diesel. That elusive 'optimal' edge geometry (ok, keep that simple to just micro-bevel angle) per load is highly sought after by users. With optimal geometry, we then like to find out the wear resistance per particular steel. Obviously, my greediness + newb (:foot:aka white belt tester) like/want to achieve both.
 
Last edited:
You must log in or register to reply here.
Back
Top