View Full Version : knurling:Chart/Formula

11-17-2014, 06:56 PM
Good afternoon guys,well I'm a beginner at machining and at knurling.I watch a Utube of ghostes with examples how to figure the diameter of stock to get the best knurl.The chart had different diameters for different TPI, so I was wondering if someone here could point me in the right direction to a chart that would have examples so that I could learn how to figure the closest diameter for what I need.Say I wanted to knurl 3/4 dia rod,using 20tpi,how should I go about figuring this.If any of this don't make any sense,I understand cause the more I read the more I get confuse.I readed that the diameter makes a difference and readed that it don't.Helpppppp-----kroll
I may need to add,I'm alittle simple minded so please take that into consideration

Optics Curmudgeon
11-17-2014, 07:03 PM
Oooh, I love it when this comes up, I'll make the popcorn....

Seriously, I knurled things for years without worrying about the diameter before someone told me I was doing it wrong. I still don't worry about the diameter, and it still works.

11-17-2014, 07:16 PM
In order to get the "best" knurl, the circumference of the work must be such that it will contain an integral number of "teeth" so if you have a 10 TPI knurl wheel That is .1 inch per tooth. The circumference of the work must be a multiple of that pitch.

The circumference of a 1 inch diameter piece of stock is 3.14 inches. This will result in an inconsistent knurl given a 10 TPI wheel. You would need to turn down the stock to a diameter of .955 to get it to fit an integral number of teeth.

If for example you have an 18 tpi wheel, that pitch is (1/18) or .055 inches. So your stock circumference should be an integral number of .055 increments. So if you again take a 1 inch dia piece of stock, the circumference would be pi*D or 3.14 inches. Divide 3.14 by .055 and you get 57.09 Not exactly 57 so you would have to reduce the diameter of your stock to .998 to get it to be perfect. ((57*.055)/pi).

Hope this helps.

11-17-2014, 07:18 PM
Seriously, you don't need to worry about it. Just plunge the tool in and it will probably work out.

If you want to do the calculations, it's very straightforward. Make the circumference of the work an integer multiple of the pitch of the chosen knurl.

If that's too much math for you, there's a program on my page.

11-17-2014, 07:39 PM
I've found the formulas to be about as useful as a spork on spaghetti day (bad memories from middle school, spaghetti wednesday's)...

What I have found out... And what has worked really well for me.

When feeding it in on the X, SLAM it in. You want to get to full depth in less than one rev. And SLAM it HARD. with a low
rpm, then you can pick up the revs (far beyond what the charts say) and then feed it off, (far faster than the charts say).

Ignore the theoretical "depth's" and just slam it home...

I've had 96-98% success with this method.

11-17-2014, 07:48 PM
This thread might be helpful


11-17-2014, 09:51 PM
Some years ago John Stevenson turned a shaft so that it had a stepped taper. Then he knurled it. It looked like this:



Mike Amick
11-17-2014, 10:16 PM
ahh come on !!! .. don't leave me hangin like that. I see the results and to my inexperienced eye
it looks like all the diameters look good. Is that right ? .. or .. does it show one (the right) diameter
looking better than the rest?

Mike A

11-17-2014, 10:42 PM
I'm another fellow who has for over 30 years unwittingly knurled all kinds of things at different diameters and types of knurls without regard to any formula for the process. And they seemed to come out fine every time.


11-17-2014, 11:49 PM
Get a scissor type knurling tool, throw the charts in the garbage and make perfect knurls the first and every time.

Deus Machina
11-18-2014, 01:51 AM
Even my $10 tool works great. Take that as you will.
If it's a production environment, things may work a hair faster or cleaner, but at home? Forget it.
If you want them sharp but to an exact dimension, take a skim of the OD.
Otherwise, knurling tools are self-spacing. Since they deform the material, they essentially just do so until it's the size it needs.

Frank Ford
11-18-2014, 03:02 AM
The advice I got when I first wanted to try knurling was from the guy who sold me my first lathe. He said, "Knurling is magic. Believe in the magic, plunge the tool and knurling will happen."

Since that day I have continued to believe in the magic. I've knurled all manner of sizes, pitches and styles, never once calculated or measured before plunging, and the magic still works. . .

11-18-2014, 07:17 AM
I agree with Frank, and have demonstrated how NO formula is required. It just plain works.

The parts in the photos I posted are all medical or fiberoptic related parts in either 303 or 316 stainless steel. Each one of them is knurled UP TO a required diameter with a +/-.002" tolerance. I quickly learned that for a 30 pitch knurl I should turn about .016" under nominal, for a 35 pitch it was a little less than that. I could have the tool set up and knurl running in a few minutes, all on CNC Swiss screw machines.

Lew Hartswick
11-18-2014, 08:21 AM
I basically did the John thing with a diamond scissor knurler at school back when that was posted with
the same results. So the old crap about the circumference having to be some majic relationship to the
knurl is pure BS.
edit: Plus over the years I've had a hundred high school students knurl (diamond, medium and coarse)
the aluminum hammer handle that the dwg. calls out .687 with over (seldom) and under by anywhere
from .005 to .030 or more under and still have the knurls come out fine. The only "trick" is use both
hands,: Turn on the spindle and quickly crank the pressure as far as comfortable immediately stop the
spindle and check the depth/tracking, repeat till it's right. next start engage the carriage drive at fairly
high rate on the start of the spindle. ie don't let things sit there going round and round without having
the other things doing their bit. :-) We do those at 118 RPM and fastest carriage available ( I forget
the travel per rev.)

11-18-2014, 10:31 AM
Disregarding the circumference on steel might work as you make lots of chips and dull the knurls but on brass, it looks like crap!

Stu Miller
11-18-2014, 11:02 AM
Let's take Joe_B's formula example. In addition, assume the knurl depth is .02 inch. The diameter at the root of the knurl is then .96 inch, the circumference is 3.01, and the number of spaces is 3.01/.055 = 54.72. The formula gave 57. How did the knurl get from the 57 spacing to the 54.72 spacing as the knurl was fed in?

I am with the slam it in crowd.

11-18-2014, 11:56 AM
kroll: Joe_b got it right. It's like setting up to cut gear teeth, you need for the knurl to have the right pitch for the circumference for it to work right.


11-18-2014, 12:10 PM
See, that's the trouble with asking these kinds of questions, you're going to get conflicting answers, all from people who claim to know with certainty. So in the final analysis you still have no clue as to what's right! :D

Lump me in with the "Diameter doesn't matter crowd."

Paul Alciatore
11-18-2014, 02:32 PM
I need some melted butter for my popcorn. Anyone got a formula for that?

11-19-2014, 01:25 PM
For the calculating types, Dorian offers a knurling calculation worksheet: http://www.innovativetoolsales.com/Dorian_technical.htm
Scroll down, on the bottom left.


11-19-2014, 02:02 PM
I need some melted butter for my popcorn. Anyone got a formula for that?


You can never go wrong by melting butter in a container held in boiling water :)

Just reminded myself - when you melt butter for popcorn add a bit of coconut butter or oil to it and mix it well. Best popcorn ever. You going to share?

11-22-2014, 03:41 PM

You can never go wrong by melting butter in a container held in boiling water :)

Just reminded myself - when you melt butter for popcorn add a bit of coconut butter or oil to it and mix it well. Best popcorn ever. You going to share?

When Knurling the circumference increases with the depth of the cut, unlike gear cutting where the circumference is constant. I have been knurling for over fifty years and have used a formula twice. Just go for it.