experiment report

[Fred_Farkle]

For those interested...

In a past thread I had indicated that I would experiment with common jade and synthetic corundum for use as lathe bits and scraping blades. The following is the results so far.

Common jade - Lathe bits. Pretty much a wash out. Micro-fractures on all materials I tested except for Al. On Al, the jade leaves a visably better finish than HSS and does not weld. I do not have carbide bits that are specifically recommended for Al. so I did not compare carbide.

On Al, the jade holds its edge, with no visible wear for about 12-13 minutes of cutting, then suddenly, the edge wears completely out in a few seconds. Thermal stress?

Common jade - Scraping blade. Refuses to cut. Edge fails on first attempt. Tried several angles between 85 and 95 degrees.

Corundum - Lathe bits. Interesting results. Started the experiment with the corundum ground to the same angles as a common c-2 carbide insert. The corundum would micro-fracture on contact with the work material, then would continue to cut with no measurable wear for as long as my patience held out (about an hour). Work surface finish was terrible, all feeds/speeds. Reduced the relief to 1-2 degrees, tried again. Isolated micro-fractures, similar to what you see with a new carbide bit. Surface finish on work was as bad as I have seen. Kinda of a hacked appearance. NFO In its single crystal state, corundum exhibits directional hardness with respect to its crystal lattice. Usually 9.0 on the MOHS scale, and 9.4 MOHS cross-grain to the lattice. So with difficulty, 9.4 is hard, even for diamond, I cut and polished a new test bit against the grain. Cuts like a champ, very few micro-fractures (same angles as carbide). Surface finish really, really bad on all metals. Now here's the suprise - Plastics... best finish, smoothest cut. Didn't do wood Thrud, but based on your tip, I will.

Corundum - Scraping blade. Ground and polished to 93 degrees. Wears much better than carbide. In fact, I'm still using one of my test blades after 6 hours of scraping. No measurable wear at the macro or micro scale. Very fragile. Made three test blades, lost two to fractures in the first few minutes.

Discussion - The jade was fun, end of experiment. Don't know whats going on with the corundum. Much better than the carbide I have, but still not suitable?? Going to try some more experiments, open to suggestions folks.

Mike

[This message has been edited by Fred_Farkle (edited 02-09-2003).]

[Spin Doctor]

I'm not really suprised with the performance of jade in cutting Aluminum. However it probably does not perform as well as diamond on the same matereial.
The corundum breakage in scrapping use really doesn't suprise either. Think of scrapping as being similar to shaping. Lots of impact events over a short time.

[Fred_Farkle]

Spin Doctor and all - There are several reasons why I pursue the experiments that I reported above. Several folks have stated the obvious with respect to the brittleness of corundum. But let's take it a bit further...

1. First, PCD does not bond to the material that is the substate for a bit, usually carbide. Diamond is inert for all practical purposes, so to form the bit the carbide surface must be roughed to allow the diamond to attach. There is no molecular bond. The diamond will fail at the temperature that the carbide fails. Al Oxide (corundum) can form a molecular bond with a suitable substrate (ceramic materials are best, also Al oxide), and can be deposited with a polycrystaline structure. Hi-temp ceramics are available with working temp to 4,000 F. Single crystal corundun is good to 3500-3700 F.

2. Al oxide (corundum) is about a thousand times cheaper than PCD, in any form.

3. A scraping blade made from corundum, cut against the lattice, will not break from the forces available from a humans arm. The corundum blade that I continue to use is cut against grain. If I bang it on things, it will fracture.

4. My experiment is really toward reducing or eliminating micro-fractures. Quick example-- The old people made their tools from stones. When they could, they preferred to make their edged tools from several kinds of Quartz minerals. They formed the edges by pressing an antler (or bone) against the material at just the right angle. This spalled material off and allowed the formation of a kind of serrated edge. They did not stike or hit the material. Quartz is from 7 to 7.3 on the mohs scale, very hard. The force used induced slippage in the lattice of the material. My experiment is to find a form, angle, where this slippage can be reduced or eliminated in corundum and still have it cut metal.

5. Diamond is slightly more brittle than corundum and cleaves much more readlily. PCD does not have cleavage. With corundum, the trick is to find a way to predict in the raw material the direction of the lattice that provides the best resistance to the cutting forces.

6. In the lathe bit experiment, I found what I thought might work in terms of angle, but still had a problem with "tearing" the metal surface. Not so on plastic. The surface finish from the corundum scraper blade is the same as carbide or HSS.

7. Corundum is also much cheaper than carbide. Wears much better, but unfortunately, leaves a surface that looks like crap.???

8. Most of the time when folks say they can predict the results of an experiment, they are usually right. I like playing the odds, even if they are against me.

9. Where I work, we are working on coating missile radomes with PCD. All have failed because the PCD conducts heat super good and the non-bonded substrate fails. Next experiments will be with corundum in the polycrystaline form with doping to create a thermal gradient, so I may have some scrap that has been surplused to play with in my experiment.

10. As to the jade. In lapidary work, jade is one of the toughest materials to work. So when I was thinking about playing with the corundum, I thought I would give it a try. If it did actually turn out to be useful for metal work, a new problem would arise. Jade is a highly variable mineral.

[This message has been edited by Fred_Farkle (edited 02-09-2003).]