Distance rings?

[zephyr9900]

Does anyone have experience using 1/8" shaft carbide endmills with distance rings (i.e. PC board mills) with the Proxxon tool? If so, how repeatable is tool insertion, both tool-to-tool and successive inserts of a given tool?

I realize that the Proxxon collet has a tapered OD , but finger-tightening it, even with the little lozenge wrench, should not be too variable.

If the tools were repeatable (or maybe heat-shrunk aluminum rings...) I could populate a higher reach of the tool table with lengths for the little endmills (separate, of course, from the primary spindle tools).

Thanks,

Randy

[justgary]

Randy -

You hinted several days ago that you would ask this question. I am very close to needing to make a PC board on my mill, and will be able to answer better after that. However, I have the surplus box o' carbide drills with rings on them, and all of the ones that I have played with are very tight on the shaft. I plan to do the same thing in the tool table if it works.

Not having my Proxxon here at work, I don't remember if the collet sticks proud of the nut or not. I suspect, though, that even if it ends up indexing off of the nut itself it will be very repeatable.

An emperical test would be easy to set up: Face an aluminum plate and then switch to the Proxxon. Install a mill with a ring into the Proxxon, and bump Z down and then move X or Y a little bit. When you make a faint mark, note the Z value. Replace the mill and try again. After a few trials, you should have the answer (remember, three is infinity for testing, especially if the numbers are within 0.0005" or so). That is a decent way to measure offsets for your tool table, too.

You could do this with a touch plate, but I wonder about the endplay in the Proxxon bearings when it is running. That may be more than the placement error of the mill. Your locomotive drivers show that it is not much, but I still wonder if it isn't the larger error.

Where do you get your mills? I get them from the dentist (for free), but they don't have rings on them. My hygenist knows to hand me a bag of them whenever I come in. I don't think they are carbide, either.

Regards,

- Just Gary

[thackman]

I have the upgraded primary spindle, and I've tried using some 2mm PCB at 5,000 rpm on 6061 and haven't had much luck. They tend to break at the top of the flutes after about 20 minutes of cutting. I wasn't sure if it was because of the low sfm or because the flutes were just too long. Could either of you post some of the feeds, speeds, doc and stepover that you've been using. I've been considering a HSS but didn't know if that would make a PCB endmill last any longer. Their flute length is far longer than what you would find on a endmill designed for metal but they are so inexpensive that they are cost effective even w/ the constant breakage.

[Herbertkabi]

Perhaps it is not what you need - I use 1/8" two flute carbide endmills when making ultra thin distance rings for angular contact bearings (for example) from adequate thickness bronze sheet (or Kapton, or Invar ... what ever) - 10...20 ... 30 ... pieces of sheet between two ca 2mm aluminium disk, then centre hole, then screwed on to round bar and using rotary table I cut OD and ID, RPM ca 30.000 and of course coolant.
Veeeeeery nice distance rings!
cheers,
herbert

[zephyr9900]

Just Gary, thanks for the feedback. The collet is proud of the nut when finger-tightened with the "lozenge wrench"), and also seems to be proud when just finger-tight, so that is encouraging. I just found a few PCB mills with distance rings I bought years ago at a model railroad show, so I can do some testing.

I have been buying my tiny endmills from Bits & Bits, and they're not quite as free as your dentist-originated bits.

I recently replaced the nose bearing on my Proxxon with a hybrid ceramic bearing from Proxxon World in Australia, which seems to have a better feel than the stock metal bearing did. It is a single bearing, but spring-loaded upwards against a stop on the spindle. With the light milling passes I'm hoping (and it seems so) that the mill doesn't pull down against the spring enough to deflect it (well, I realize that everything has deflection, but relatively not here...)

thackman, the Machinist Mate software I use to calculate feeds and speeds (not being knowledgeable enough to do it in my head ) says for a 2mm 2-flute endmill in aluminum 29krpm and 16ipm feed. So at 5krpm the feed would be 2.75 ipm--I've had good success ratioing the feedrate from the rpm, assuming I'm using coolant. The .020" endmills I've been using want 76krpm and 10.7ipm in brass, and I'm not breaking any lately going 2.8ipm at the Proxxon's (assumed) top speed of 20k.

herbert, the distance rings I'm talking about are the ones pressed onto the endmill itself. I know the term is used for various purposes (I've seen it used for lens spacer rings too). But your bearing distance rings (for preloaded pairs, right?) sound very nice. I'd like to hear about your experience doing that, if you don't mind telling.

Randy

[justgary]

Randy -

Very nice photos.

I'm thinking that indexing off the collet will work just fine, but I guess we really need to test it. My Proxxon doesn't sound anything like 20K RPM, by the way. My old Dremel has a much higher pitched whine. I don't have an optical tachometer, but by the sound, I don't think the Proxxon is even running 10K. Maybe it's just mine, and yours is faster.

I'm working on whipping up one of those Porter-Cable router heads as an HSS, and putting the Proxxon on the bench. Having the 1/4" and 1/2" collets available for high-speed is just too nice to not head that direction.

Speaking of Machinist's Mate (you mentioned it months ago when I was still waaaay to new to know to care), do you find that it gets the numbers right and that the Tormach can keep up? Or do you break tooling when using their numbers? I don't have any problem starting with 50% feed and working up, but do you trust what it tells you? I'm getting more interested in saving time now.

Regards,

- Just Gary

[Herbertkabi]

Originally Posted by zephyr9900 J herbert, the distance rings I'm talking about are the ones pressed onto the endmill itself. I know the term is used for various purposes (I've seen it used for lens spacer rings too). But your bearing distance rings (for preloaded pairs, right?) sound very nice. I'd like to hear about your experience doing that, if you don't mind telling. Randy

Oh dear! Sorry, very embarrassing ...
I was slumberous and did not read carefully ...
At that - I use 1/16 endmills when cutting, 1/8 shank ...
and I have many endmills and micro drills with distance rings you are speaking,
but I dont use thoses systematised
cheers,
Herbert,
now goes to bed

[zephyr9900]

Just Gary, I don't think my Proxxon is going that fast either, but I don't have any way to measure mine either.

I know Machinist Mate is based on good solid numbers and when I use its recommendations I get good looking chips. The variable that is in my control (and for which MM gives no guidance because there are too many factors to consider--lenth of flute, length of cutter cantilevered from the holder, machine stiffness etc.) is depth of cut, and for that I tend to be conservative because going too far on that will snap cutters and I don't like that. I do know that I can be much more agressive with stub-length endmills (pretty much use their whole flute) than the extra-long cutters I usually use to machine narrow cavities as deep as I need to go.

Randy

[justgary]

Randy -

I have slowly learned the same thing about stub mills and lever arm on long mills. It's amazing the first time you actually see the difference between the two.

I may give MM a try, because I'm ready for fast *and* good finish.

On the distance rings topic, I'm cutting my first Proxxon parts today. I'll post them for fun later. The dental 0.030 mill is working great in brass at 5 IPM, 50% depth and 25% stepover. It starts to chatter above that, but I couldn't get it deeper into the collet, so it's pretty "long" right now.

I'll play with the distance rings after I get these parts done.

I may rig up a photodetector and watch a paint stripe on the Proxxon shaft. An oscilloscope will tell me the RPM.

Regards,

- Just Gary

[MichaelHenry]

I may give MM a try, because I'm ready for fast *and* good finish.

You might also want to consider ME Consultant Standard or Pro:

http://closetolerancesoftware.com/Pr...alculator.html

I use the Pro version (didn't know they had a Standard) and find it quite helpful and pretty much right on so far as feeds and speeds go. It can also be editted to add new materials. There's a trial version available, too.

Mike

[justgary]

MichaelHenry -

Thanks. I'm slow when it comes to spending money, but I'll start checking these out.

Randy -

OK, I've done the test. I faced a piece of brass rod with the shell mill, then switched to the Proxxon. I used an old broken #80 carbide drill and stepped Z down until the bit just left a mark. I drew a few lines with it, did a rapid to Z4 and back, and then a few more lines. I did that a few times. The Tormach was repeatable to the 0.0005" that I was stepping.

Next, I removed and replaced the bit and tried again several times. The Z placement was off every time, sometimes as much as 0.012". I faced the collet nut so that I could let the ring touch the collet face, and the numbers came down to under 0.005" or so. I could see that I was getting the ring flush to the collet every time (using a strong backlight), so I went looking for the problem. Yup, the ring on that bit was cracked, and I was apparently moving it slightly every time.

I found another broken drill (with a good ring) and tried again. The result is the top row of circles. The left one is where I zeroed Z, and the next one I came up 0.0005 to check the pattern. For the rest, I did a rapid to Z4, replaced the bit, rapid to 0.1, G1Z0.0005, and stepped over and did it again.

As long as I could not see light between the collet and ring, it repeated just fine. So now we know (but check the rings before you start).

Regards,

- Just Gary

[zephyr9900]

Wow, thanks for doing that testing, Just Gary! I was occupied most of the day and didn't get around to my own testing. That looks very promising. I'll repeat on my end and that will give us a sample of two Proxxons.

LPKF sells distance rings and a tool to set them. Conceptually it isn't any different than a bicycle chain tool. but since they depend on all tools being the same projection, you set the ring until the tool tip lines up with a reticle on a built-in microscope.

We have the advantage of a tool table, so we just need to be able to push rings onto 1/8"-shank bits. The LPKF tool instructions (which I didn't link to because it requires registering with their website to access) say that the rings can only be pushed towards the shank end and not the tip. If the pusher screw had a small bore (larger than the tip but smaller than the shank), I don't see why the bit couldn't be reversed, the screw bearing on the taper of the bit. We could minimize the overhang of already-ringed bits that way if desired.

Unfortunately my chain tool's push pin is .130" diameter, so the whole shank of the tool would fit in the push screw recess when the pin is removed. But I have a funny feeling that sometime soon I'll have a similar tool specially built for the 1/8" bits.

The only complication will be 1/8" diameter flute bits. If we just push the ring over the flutes, they would proably broach the ring into uselessness, but maybe twisting the ring onto the flutes down to the beginning of the shank and then pressing might work.

That's a mighty good job of focusing on the depth of interest, Just Gary. Do you have a technique? I press the shutter button halfway to let the camera auto-focus and lock, notice where it focused, and move the camera by a guessed amount to transfer the focus to where I think it should be. And repeat 5 or 10 times to get a maybe good shot...

Randy