Perfectly ground Twist Drills every time.

Gray,

Sorry, I can't point you to an authoritative reference; but your recollection seems entirely logical.

At the microscopic level, even the best grinding marks are like the furrows in a field [first illustrated by Robert Hooke, in Micrographia] so, obviously, grooves parallel to the cutting edge will weaken it dramatically: At the opposite extreme therefore, grooves perpendicular to the edge will weaken it least, and every angle between these extremes will be intermediate in effect.

MichaelG.

Edited By Michael Gilligan on 04/05/2014 10:11:02

Michael,

The cuttting edge of a twist drill isn't truly radial. Is the best angle at 90-degrees to the cutting edge or tangential to a circle at the point where it crosses the edge. And if the latter, at what point on the cutting edge? Half diameter? Half way along? Outer end? Inner end?

I expect a detailed finite element analysis of the forces on the cutting edge by 4:00pm

Neil

(If I know you lot someone WILL go and do this! - so here's a starting point: lisafea.com/)

Edited By Neil Wyatt on 04/05/2014 10:22:34

I believe it is mentioned here Graham

**LINK**

Regards,

IanT

Never underestimate how grinding affects the outcome. I can see in Grays case why he's thinking about getting the marks at 90 degrees to the cutting edge.

My 4 facet grinder does that and it also has the facility to do it in either direction. It makes a big different, probably not so much on the cutting but more on the life of the tool.

Sorry to harp on about the piano company but because we were using all non standard tools and had been for over a 100 years there was a lot of experience on how to do the job and get the best out of the tools.

All the slotting saws were special widths. We tried having them commercially made but they were expensive and didn't last because like all commercial saws they were hollow ground and after the first couple of grinds they were under size.

If you just surface grind a saw to the correct width all it does is burn the wood because it has no clearance. The trick was to grind them a thou undersize then sharpen the teeth making sure the wheel ground off the cutting edge, not onto it. Every alternate tooth was ground one way and then either the saw was turned over or the grinder reversed to do the other teeth.

Basically the burr acted as clearance. Now you would think that burr would soon be worn away but truth was they lasted quite a time and a simple face grind and they would be good to go again.

We can prove the same thing in this world as well. Get a brazed tipped lathe tool and grind one face on a green grit wheel with the direction of grind going down the face and look at it under a decent glass. The edges will be all small craters because the carbide is so friable. Then reverse the operation and the edge damage is reduced considerably.

Same applies to HSS but because it's not so friable the results are not as obvious to the naked eye, but they are still there hence the advise to hone the edge afterwards.

.

Nice one, Neil

Of course, in an ideal world the grinding marks would be radial to every point on the curved cutting-edge [limited only by the pitch of the furrows made by the grit of the grindstone] ... don't think I'll bother.

MichaelG.

.

P.S.  Finite Element analysis is [by definition] a digital approximation.

Edited By Michael Gilligan on 04/05/2014 14:51:02

Hi JohnS,

"Modern technology has proved time and again what we were taught by rote and everyone since has repeated parrot fashion is not always the correct way."

Ain't that the truth.

Chris, the other John Stephens' son

.

... He also describes it beautifully

... There's a good transcript here.

MichaelG.

Had to smile at this one Gray, another teaching by rote.

Some years ago I was in the gear cutters and they were machining some big blanks from some form of alloy steel. Chpipings coming off dull red and 3' deep on the floor. I asked why they didn't use coolant. The reply was similar to yours unless in this case you can absolutely flood the work which on this old unguarded manual lathe would have been impossible or messy best not bother as part cooling just cracked the inserts.

The bit that made me smile even more was knowing this I still dunk my drills like you. Old habits die hard.

Edited By John Stevenson on 04/05/2014 15:43:18

Hi Gray,

As to how to quench, I always do it from the cold end, so there is far less thermal shock than plunging the hot bit first, and as I don't like the smell of burnt fingers I shall still have a bucket of water handy, besides flood cooling might be a bit messy on my B&D 8 inch .

ATB

chriStephens

Gray,

Query please from the statement above.

Using the jig you show and having the edge trailing you run the grinder in reverse ?

Won't this also depend on which end of a double ended grinder you use?

Sorry Graham you have lost me.

Regardless of whether it's the edge or heel making contact first the wheel is always going in one direction, usually down and I don't see how the cutting edge can trail unless it goes up.

Am I missing something ?

Ok see your explanation but in my book the only way a cutting edge would trail is if the wheel is throwing the burr off the edge hence the question about direction.

Another question.

What is wrong in using the periphery of a wheel ?

For one thing wheel manufacturers do not condone using the side unless it's a wheel specifically made for this like a cup wheel.

Many cutters are hollow ground for clearance and some profession tool and cutter grinders hat keep the spindle horizontal at all times use the periphery and distances above centre to get the required angles. The Boxford G200 even includes a table giving wheel sizes and at what distance above or below centre to achieve a certain angle.

If we get down to the nitty gritty [ is that a pun ? ] on a 6" wheel with a land of 0.100" wide the curve is 0,00042"

Hardly earth shattering.

See your point but it's cosmetic anyway.

The heel does nothing only provide clearance you could put that on with an angle grinder if necessary.

In fact on some of the bigger stuff I have resurrected from scrap yards like 2 1/2" diameter drills to save time and prevent too much undue heat I have sliced back an inch or so on damaged drills with an angle grinder.

OK I can see where you are coming from, I was thinking more along the lines of the drill lying across the wheel so the lines were at 90 degrees to the edge and the curvature would be minimal.

However even though on a 1" drill this is 0.37 of a mm curvature does it matter ?

Where is it written in stone that a drill land has to be a straight line? In fact drills made especially for very soft aircraft alloys are ground more like a corner rounding endmill than a drill, they have a real radius on the land.

No not meant to cover special purpose drills only asking if a straight line was absolutely needed.

However this has spiked my interest in that these jigs have been around for literally century's but for obvious reasons industry cannot use them so what do they use ?

Even semi industrial units like the Darex still swirl the drills round a wheel.

Do a patent search on the latest Swiss drill grinders, the big industrial ones and you will get a pleasant surprise in that although they are CNC the actual mechanism once the servo motors are removed hasn't changed form the 1950's and is absolute simplicity itself.

Gray, et al absolutely fascinating thank you.

I must ask, why bother with the original base at all? if you are going to the bother of machining an extension wedge why not ditch the original low quality casting and knock up a replacement steel base with a 14* hole and perhaps even a nice brass bush for it to rotate in.

Paul

Gray,

Sorry think I have unwittingly confused you by mentioning CNC.

The modern Swiss made ones basically have had the mechanical bits changed to servo's, smartened up a bit, bunged in a big box with fancy logo's on but underneath they are still the same design as the 1950's models.

They are hopper fed and thru feed the drills after grinding so high speed, can't see how many per hour but that obviously depends on size. The brilliant thing is they are so simple and no weird geometry to have to understand.

I'm simply amazed than none of the smaller industrial players have latched on to this, can't be patents because the idea is so old.