I need to bump knurl this part on my CNC lathe and try to hold a tolerance of +/-.001. I am using a straight knurl but am having a problem staying consistent. I have been dwelling when I reach depth for 5 seconds and have a feeling that is not the best way to go about it. If anybody has any suggestions that would be great. I realize knurling is a trial and error type of thing but I don't even know if my speeds/feeds are anywhere close to where they need to be. Thanks.

OOPS! That file I attached was horrible. This one should be better.

I was hoping to see some intelligent answers here. Don't know if it is because we seldom get a knurling job, but knurling has always been a PIA for me. Most are done using X-axis to push the knurl into the part. I assume this is what you call "bump knurling". We did straddle knurl one job. Broke the knurling tool twice. (chair)

Picked up a used Hardinge not long ago that came with a knurling tool I hadn't seen before. Looks like you cut with the edge of the tool, and feed in from Z-axis.

I know that pitch and diameter are tied together, but that knowledge doesn't seem to help me much. Must admit that knurling is a subject I never researched.

Surely there must be someone on this forum that could enlighten us!

BTW, what do I need on my computer to open a .dxf file? Autocad? Know a .dxf file can be imported into MasterCam.

would probably help if you posted picture version of your dxf, since its kind of extra pain to download it and open in acad or other software, specialy on older computers, it takes quite awhile to load the software. :)

.

I was hoping to see some intelligent answers here. Don't know if it is because we seldom get a knurling job, but knurling has always been a PIA for me. Most are done using X-axis to push the knurl into the part. I assume this is what you call "bump knurling". We did straddle knurl one job. Broke the knurling tool twice. (chair)

Picked up a used Hardinge not long ago that came with a knurling tool I hadn't seen before. Looks like you cut with the edge of the tool, and feed in from Z-axis.

I know that pitch and diameter are tied together, but that knowledge doesn't seem to help me much. Must admit that knurling is a subject I never researched.

Surely there must be someone on this forum that could enlighten us!

BTW, what do I need on my computer to open a .dxf file? Autocad? Know a .dxf file can be imported into MasterCam.


AutoCAD will open dxf's, but you can use this to open it:

http://www.solidworks.com/pages/products/edrawings/viewer.html

But I must ask the question.... Any advise on this part. It is driving me and my boss crazy. HELP A NEWBIE IN NEED!!!!!

AutoCAD will open dxf's, but you can use this to open it:

http://www.solidworks.com/pages/products/edrawings/viewer.html

Thanks for the link. It did the job!

Crashmaster: I hope someone comes along soon to give us a lesson on knurling.

.....Picked up a used Hardinge not long ago that came with a knurling tool I hadn't seen before. Looks like you cut with the edge of the tool, and feed in from Z-axis.

I know that pitch and diameter are tied together, but that knowledge doesn't seem to help me much....

I have not knurled on a CNC but in the past have done plenty on Turret and Center lathes.

To comment on your second point yes there is a relationship between the knurl spacing and the diameter; you can consider the knurl and the workpiece to be a pair of gears so you have to fit an integral number of bumps around the workpiece. But at the same time your workpiece is changing diameter because the bumps come up as the hollows go down. I suppose it would be possible to calculate a 'pitch diameter' for the knurling and make that your starting diameter; I nearly always did it by trial and error.

Regarding feeding in from Z that is how we did it on turrets, both straight and diamond knurl; it wasn't really feasible to do anything else. It also seemed to be easier and I always figured it was because the knurl was working on a smaller area of metal. If you come in on X the knurl touches across the entire face and is distorting the workpiece simultaneously on the full width of the knurl; coming in from Z means it is plowing through into the workpiece only on the leading edge.

On a center lathe I would come in at the end of the workpiece with the cross slide and have the knurl overlapping just a small amount. This would start the knurl and get everything in synch and then I would traverse along with the carriage.

Sometimes I found the knurl would start to cut a double start with the knurl rollers spinning twice as fast as they should be; you could see the different appearance on the work when this happened. I would back out and start again. This would leave a poorly knurled section at the end but generally after knurling I would re-chamfer the start end to remove the sharp burrs that the knurl pushes out on the end.

Incidentally starting from the needs a bit of chamfer before knurling to allow the knurl to ride into the part.

Regarding holding +/-0.001" when knurling I don't see how that is possible. Also if the picture that was posted is the part being done I don't see why it is necessary, the knurl is on the handle of a screw-in alignment pin. It would be a very sensitive hand to detect a couple of thou difference in diameter.

The part is actually an arbor that is pressed inside of a tube. It the knurl is to big, the tube will bend when being pressed. If the knurl is too small it wont stay in the tube.

The part is actually an arbor that is pressed inside of a tube. It the knurl is to big, the tube will bend when being pressed. If the knurl is too small it wont stay in the tube.

Okay, I take it the knurl is then supposed to be acting as a spline that cuts, or forms, it own mating surface inside the tube as it is pushed in. I wonder if anyone has tested this approach rather than a simple press fit. I doubt that it gives a better load carrying capacity and would not be surprised if it was less.

Partly trial and error and then making a note of the depth/ start diameter and feed that produced the desired result. Holding to +-.001 may only be possi-ble with a specialized machine that only does the knurling procedure and not with a multi function lathe. Holding a good knurl all the way to the shoulder is also something that needs a specialized machine (IMO) Some jobs are not worth bidding on due to the frustration and near impossible (with the machinery on hand) oprations required.

Geof, what is involved with a simple press fit. Would I just turn the OD of the arbor to the ID of the tube? Or is the arbor slightly bigger? The tube I am pressing the arbor into is chrome plated, would press fitting it cause enough stress on the tube to crack the chrome?

Geof, what is involved with a simple press fit. Would I just turn the OD of the arbor to the ID of the tube? Or is the arbor slightly bigger? The tube I am pressing the arbor into is chrome plated, would press fitting it cause enough stress on the tube to crack the chrome?

For a press fit the arbor would be slightly larger than the ID of the tube, how much I cannot say.

Are these your own product or a job for a customer; if it is a customer you do what they want. For an in-house product I would do some experimenting. It looks like the tube might be 1" OD with a 1/16" wall, I would expect that a press of 0.005" would not crack the chrome.

You mention the tube buckling if the knurl is too large. How long is the tube? It is, of course, difficult to put a decent press on a tube that is longer than around 10 times the diameter because tubes do buckle. For pressing into a long tube I would build a fixture that the tube would be supported in. This would be a few thou clearance, 5 or 10 thou maybe, for most of the length of the tube but the part surrounding where the piece was being pressed in would have a bit of extra clearance to allow for the tube expanding.

We actually make a product that needs an insert pressed inside a 24 inches long rectangular tube and down to a position about 12 inches from the end and we do use a fixture that supports the tube which is why I am rambling on here. I am thinking about how I would go about designing a tubular arbor with shafts pressed into the ends. I don't think I would use knurling but I would also probably test both knurled and straight shafts at different amounts of press.

Id say if you absolutely have to knurl this part because you don't have authority to make an engineering change in either mating part... then knurl with a clamp type tool to an oversize condition and just polish it down carefully with a fine file or sandpaper. You can easily raise it past your finish diameter a few thou. If you mess up and go undersize just run the tool again a little more(deeper). It will bring it back up by displacing more metal. Not too much though because you'll reach a point of no return where no matter how much you try, there's no more material to displace.

If I had my choice... drop the knurl idea and go for an interference fit of a few thou., of course depends on the length and diameter you're shooting for and the strength of the tube your working with... As Geof suggests you might need to make a support tube so it won't kink during the press. Iv'e had to do just that many times and it works good.

I do this stuff quite a lot with thin tubing accepting a solid plug. I don't press anything anymore it's just too damn easy to scrap things. I'd go for a thermal fit. Probably heating the tube to about 300 degrees F/ +-50 and simply drop the arbor in the tube letting it shrink onto it. DONE!!

I seriously doubt you will beat a plain interference fit by knurling because of two things:
#1- Chrome is so hard it's not going to 'Bite' or deform with the knurl.
#2- Mating contact will only occur at the crest of the knurl meaning much less surface for holding power.
That's all that matters with fits... length, diameter, and how much strength the weakest component has to resist deformation.

Another nice thing about 'shrink fits' over a press is that you generally get twice to three times the holding power since there is no shearing of the parent metal that accompanies a press. Not all the time but most. And that knurled part will lose out on the material thing as you shove it down.

Geof you are right on the specs. 1" OD 1/16" wall thickness. The tube is 40 inches long (Its for a microphone stand mast)

done lots of knurling, but not with cnc - whats a bump knurl btw? As Geof pointed out, there is no point in calculating a pitch dia because it changes as the knurls develop, but with a little trial and error a start dia of od-2*knurl depth, you might be able to hold a tolerance. I've heard of guys cutting knurls in one pass, you'd have to use a scissor type knurler, but its usually done in multiple passes with fairly fast feed rate.

If you want make all these knurling questions and problems go away, and the job is worthwhile, consider a cut knurling tool instead of a pressure one. In that case, you do do the calculations for the blank, like you would for a gear, you'll get an exact result. making one is on my to do list, but may never percolate up to the top, they're to dang expensive for a home shop.

here's a link

http://www.zeus-tooling.com/produkte/raendelfraeswerkzeug_details.php?gruppenid=15

agree with Pressfit, doesn't sound like very sound engineering; putting a 1 thou tolerance on a knurled surface. i'd tell em about loctite for this app :D

..... i'd tell em about loctite for this app :D

Read what the man writes :D although I have found with chrome Loctite is not very effective.

This is a non-critical joint, unless someone like Mick Jagger gets hold and starts flailing around, so knurling is acceptable. Press fit would be easier though I think, presumably more than one is going to be made so a press fixture is probably worth making.

Alright guys, I think I'm almost there with this. We did a thermal fit with our raw stock tubing and it worked great. Now the only problem left is when we actually get the tubing back coated with chrome. Does anybody know how chrome would react to the heat? We would not want discoloration or flaking. Otherwise we may have to investigate the cut knurling tool.

Alright guys, I think I'm almost there with this. We did a thermal fit with our raw stock tubing and it worked great. Now the only problem left is when we actually get the tubing back coated with chrome. Does anybody know how chrome would react to the heat? We would not want discoloration or flaking. Otherwise we may have to investigate the cut knurling tool.

Get sophisticated, cool the shaft part in liquid nitrogen and do an expansion fit :) .

If your chroming is good quality you should be able to go up to around 400 F without any problems, maybe higher. Send in some offcuts and get them chromed for experimenting.

Alright, well thanks everyone. I will get going on this and let you know how it turns out.

That discoloration part may happen if you get it too hot with a torch. I don't know if a few hundred degrees will do it or not, but I imagine temps on the order of 500 or more would.
Chrome exhaust pipes on motorcycles will indeed turn blue at the head bolt flanges but are subjected to far greater temps than what you're doing. Direct flame on the Chrome may be a problem. You may have to do a test. I use welders temp sticks to assure I don't go too high as I'm heating something. Basically they're a Crayon and come in different melting temps.
200, 250, 300... and so on. Soon as you think the heated part is reaching the temp you're shooting for, start touching the stick on the part. When it melts then it's there.

Get sophisticated, cool the shaft part in liquid nitrogen and do an expansion fit :) .

If your chroming is good quality you should be able to go up to around 400 F without any problems, maybe higher. Send in some offcuts and get them chromed for experimenting.

LOL Geof. Yeah that's pretty sophisticated. Not to mention expensive and potenially dangerous if you spill it on yourself without the proper safety gear. I had a job one time where I thought the only way to do it was with LN. But after checking into the cost of the DOT approved container(Dewar), I decided another avenue. For some darn reason they just won't let you carry it home in a Thermos.:devious:

LOL Geof. Yeah that's pretty sophisticated. Not to mention expensive and potenially dangerous if you spill it on yourself without the proper safety gear.....

Nah, just leaves a little white spot which is okay provided you don't hit it with a hammer until thaws again :) .

To indirectly heat the tube it might be possible to have a sand pot. Heat the pot with the flame and just poke the tube into the sand.

The best would be an induction heater such as are used for shrink fitting bearings.

We tested it and it worked, another question though. After heating the tube is it better to let it slowly cool or to shock it with cold water?

We tested it and it worked, another question though. After heating the tube is it better to let it slowly cool or to shock it with cold water?

I would cool slowly. If you shock it the exterior contracts first, i.e. the chrome plating contracts first, this could crack the chrome. Not noticeable now but maybe in a short while.

Thanks a million guys. Geof, I think you've helped me on 3 or 4 threads in the past week. I appreciate the help you guys are giving a newbie.