Best Method to Hold 3/32 Drill Bits; Spinner Suggestions for Grinding Pilot-Point

[car2]

We have been grinding a LOT of pilot-points on standard 3/32" HSS drill bits (Viking drills). The pilot tip is .25" long, ground down to 0.052" (straight ground shoulder, nothing fancy).
We have a KO Lee tool and cutter grinder with an added on auto-feed (guess that would be "y", perpendicular to the table), K60 wheel. We're using a converted a hand-spinner (they type with no bearings, just the spindle in case-iron sleeve) with a motor, and are gripping the bits near the tip with a standard 5-c collet.


This grinding aspect seems to be working pretty well, but the work-holding equipment is causing some issues. I was inquiring if some of the grinding experts here had any suggestions for better work-holding and spinning to grind the pilots. It seems that clamping on the flutes near the tip with a 3-tine standard 5-C collet produces some runout issues, particularly on the random bit that is chucked "just wrong" in orientation. I'd like to use a better spinner than the converted indexer, like a Harig, but these are chunky and very difficult to get near the wheel with a bit chucked near the tip.

Any suggestions on holding these 3/32 bits for pilot-grinding would be appreciated. For example, would a custom 2-tine extened-nose collet for the bits, in a ball-bearing spinner work better, or does anyone know of a compact spinner with an extended nose, etc. Or should bits be held by a different method altogether. (Also, if anyone is interested, or knows a source for doing this sort of work, feel free to PM me, we're using at least a couple hundred of these a month, and are looking to outsource it).

 

[stephen thomas]

How are you finishing the lands?

If it truly is work-holding and not the import spinner or cheap collets causing eccentricity, then 5ST collets are 4-split and extended length (internal) engagement, more or less originally designed for holding drills and mills by the flutes for sharpening. You say some drills don't go in staight??? Are you using quality collet? Actual size?

One of my spinners uses Jacobs rubberflex, but at that size it probably only has 4 jaw collet & i think you are better off with the 5ST's.

If you get a ball bearing spinner, reseach between the Harig and Suburban. New, i think Suburban is a better design and easier to rebuild in future. Used, I think Suburban is a better design, and less likely to arrive on your bench needing a rebuild. But maybe that is just me.

Plain bearing spinner is probably fine for your app, unless you go motorized. Suburban has motorization kits.
Set up a drip oiler in the screw hole for the spindle screw aluminum clamp screw, since you say your plain bearing unit is motorized. Use light spindle oil. It should last a very long time, if it was tight to begin with. Did you fix the fore-aft stop rings so the spindle does not cam back & forth?

FWIW, there are a half dozen import spinners here, acquired in auction box lots with other equipment. Some are perfectly OK and i use one for most spinning to save the Suburban, and because the "accurized" import is more convenient than some others. Also have Suburban, PC101, KOLEE airbearing,, Hardinge indexers, RA Machine spinner, & All-Tool spin/index with multi-angle position for comparison. I use the RA Machine spinner for lands on step drills, because it is essentially "automatic" & amount of advance (helical clearance on the lip) is adjustable. However, these units seem to be vanishingly rare. I grind (spin) the pilot with one of the import spinners. Then do the lands as a secod op.

All that said, if you are doing hundreds/months, maybe you need something more modern, and automated? At least maybe air collet instead of screw in/out?

smt

 

[car2]

Thanks very much Stephen.

The bits are used in acrylic for a limited number of holes, in acrylic, in a dental lab application. A #55 pilot hole is drilled first, so the "pilot" on the 3/32 bit is non-cutting and follows the hole. The 3/32 pilot is straight ground down from the ID to .052 dia; the shoulder of the bit bottoms out on a metal part so the non-cutting square shoulder is desirable, no lands or relief are used.

Thanks for the headsup on the 5st collets, think I'll order a couple from Hardinge.

Attahed is a picture of the setup. I added that dog-clutch mechanism for the feed which works pretty well; aslo rebuilt all the bearings, bushings, seals, roller oilers, rack on the KO lee grinder, so it works pretty well. The spindle was good on the grinder, and it's run with a VFD. Feeding the bit into the wheel by hand was extremely tedious and variable; so with that dog-clutch, you dress the wheel, touch off the rotating bit, "mark" the starting position on the handwheel with a green-magnet, mark the end position .02 away with a red magnet, push in the spring-loaded clutch, let it feed (takes about 10 sec), let go of the clutch knob when the red magnet is reached. The wheel is dressed every 12 pieces (you have to keep track of the wheel wear a bit by moving the magnets a mil or so every couple of bits).

That spinner is a tiawan-made job, with the cast-iron journal, and scabbed-on motor and mount; the spindle fits well and we do oil it before each use thru the lock-nut. I swapped the spindle around and removed the index-plate to get the longer nose stickout. It runs very true with a dowell pin (using lyndex collets), and not bad with a drill-bit, so maybe the (new Hardinge) ST collets will be a notable improvement in consistency. I don't know how much better a Haring type spinner would be, and the main issue is that it looks very difficult to get that type of chunky spinner with the collet flush with the casting close enough to the wheel (without some sort of extension). Also, the table on that small KO-Lee grinder isn't too large, and some of those motorized spinners are very large. I have not seen any motorized compact spinners with a somewhat extended small dia nose.

The issue we have seems to be related to inconsistency of the 5c collets clamping on the flutes, so maybe the 5st will push that over the edge.

I haven't seen other simple type grinding machine that has a slow auto-feed function, short of some elaborate CNC grinder (not interested in getting into that).

We do plan to outsource this job, but surprisingly have had a difficult time finding anyone to do it.

Thanks again for the input and help. Charles

 

[michiganbuck]

A precision work head is best, perhaps a TC or OD grinder one. A good quality collet holder would be Ok.
Might run 250-300 rpm and try down grinding, wheel going down and part spinning downward on the wheel.

Low-speed spinning the part is best.. you might try even lower than 250-300.

Grind one part to size and then bring a solid diamond to .003-.005 (even zero) away from the finished part diameter, with having the diamond very close to the work area. This is so you might dress quickly ...and know that part finish is .003 more infeed.

Dress is only a minimum skim..to about what is needed average to maintain the intersect corner you desire..

Touch and travel might be used for length..you let your table slide along to wheel touch part end, and lock your table lock..then pull the length gauge set at your table lock to know/get the needed step length.

 

[michiganbuck]

(guess that would be "y", perpendicular to the table),

Letter feeds are not for manual grinders. Long and cross are best said.

 

[stephen thomas]

That's a pretty slick set up!
Hopefully, the collets will improve the issue that is giving trouble.

smt

 

[CarbideBob]

How long is the bit and how long is the fluted area?
Collet clamping on flutes is always a runout problem because of the slots and back taper in the flute area.
We do this a couple of ways but basically have it stickout a lot and run the fluted area in a drill bushing, vee or fixed supports. (Backwards so as to not harm the cutting edges)
Pic of overall finished tool? Runout spec across the length of pin area?
Really short fluted length is a problem with this as your flutes don't wrap around enough to always be supported by the vee.

Extended nose 5C collets can get you more clear for the head but have reduced clamping force and a touch more runout.

Other options include a really small centerless in plunge mode and of course a Dedtru type. Both $$$$
What is golden here is the straight flat cut with no reliefs or second drill tip so a simple pilot front grind.

This guy fits on most small grinders and has a optional air clamp but is not cheap.
HARIG 120-100 5C Collet Indexing Collet Fixture |Travers Tool

You are running hundreds with all this dress, lube, and load. I do so love it as a hack and it great and respectful but what is the output per hour of good parts?
On the other hand if an employee has nothing else to do and the machine empty, always setup and works it is just about free so do not discount that as you look to outsource.
Unsure why you can not find small shops that want to at the least quote this.
Bob

 

[car2]

Thanks again very much for everyone's time and input.

(I didn't realize it had been a year of grinding bits on the K.O.Goldberg (KOG) contraption, probably a thousand or two bits. If interested in the post below are some pictures of the KOG "autofeed" retrofit
Update: Pilot Grinding 3/32 Bits); KO Lee Autofeed Addition)

I don't want to belabor everyone's time, so I think we'll try to improve the workholding somewhat (ordered a Hardinge ST collet), the KOG is working just "adequately" with regard to surface-finish and tolerances, but we will outsource this due to the quantities and required more "refined" results, and lack of proper equipment for volume.

If interested below are responses to folks' generous input:
a) the bit is a standard jobbers length 3/32 HSS uncoated Viking bit (so about 2.24 oal and 1" flute length (we require at least 3/4" flute length). The pilot is simply ground straight in (no relief or cutting flutes) to a dia of .052+/-.004 for a length of .27+/-.01 a straight shoulder.
b) when a 3/32 dowel-pin is held in the collet in the converted spinner, the runout is about 0.0008 (at ~1/2"), so as noted, maybe the additional runout and variabity is due to the clamping the flutes in the 5c collet.
c) I'm concerned those Harig spinners are too chunky to get close to the wheel without some sort of extended nose collet, or as Bob indicated some more complicated work-holding with bushings or support and clamping the shank end with the drill projecting (not really interested in getting deeper into grinding intricacies and equipment). (I am interested if anyone knows the part-number for those Baldor variable-speed integral motors--I've looked all over for those and cannot find a catalog number, maybe they're just an OEM available item).
d) I was also surprised what a pain it has been getting this job outsourced--got a couple of quotes out of probably 10 requests, both were high, one was sort of ridiculous. We're not pinching pennies to get these made, but don't understand $8-15 quotes per part when the KOG setup can load, grind and measure a part in one minute; the part-time HS kids have been grinding them.

We also have some custom HSS bits with 2.35 MM shanks (for example HSS #55 drill with a 2.35MM shank, 0.75"min flute length), which we have not yet found a source for. So if anyone's interested or has suggestions for either this pilot bit or custom 2.35mm shank HSS bits, feel free to PM me. Quantities would probably be at least a couple hundred a month, and could be ordered in larger batches. So yes, we don't want to be in the grinding business, and the KOG setup was only intended to do low-volume/prototype work.

Thanks again for everyone's time and input, greatly appreciated. Cheers, Charles

 

[stephen thomas]

Do you need jobber length for the app?
IOW, using either single end or double end stubbies would practically allow you to grip (mostly) on the shank. & Maybe the DE version could be a useful economizing point if you are using thousands per year?

Screw Machine (Stub) Drill Bit Chart | DrillsandCutters.com

smt

 

[michiganbuck]

Some such parts can be run through a bushing
Yes, a steady of sorts need be there to hold the bushing.
Often TIR is held to very close that way..with the but end in a collet or something and the out end running in a bushing.

All American Drill Bushing 3/32" ID x 3/16" OD x 1/4" L; SF Slip/Fixed Renewable | eBay

likely stubs may be too short for this method.

*eBay bushing shown is not the best choice because a 1" OD diameter guide bushing would allow many more size parts/jobs to be run...and a thin wall can easily distort under a set screw or other pushing lock/hold.

also, a bushing jig can be handy for odd tooling..a CRS bushing can be made using the actual odd size drill, reamer, step tool, cutter to finish ID size. With step size drilling, a bore will often be .0002 over the drill/cutter used, so will run very close... You make perhaps 12 potential bushings that fit your holding device and make ID sizes (special or common) as you need them.

 

[CarbideBob]

.. (I am interested if anyone knows the part-number for those Baldor variable-speed integral motors--I've looked all over for those and cannot find a catalog number, maybe they're just an OEM available item).

If you want one PM me. Have one barley used as I convert this fixture to servo drive so that I can index for tips and reliefs on my homemade 5 axis.
It is basically a paperweight to me. I do not know if I still have the original O-ring pulley.
Bob
(your PM box is full, sent email I hope)

 

[hanermo]

S:
uggest.

A better collet setup is probably going be very repeatable and fast.
Any of the usual name brand suppliers, and types, pick your poison.

Option.
You can grip the new collet holder in the spindex, and id grind the toolholder to 1 micron.
You cant take it out ever again, but it will be dead nuts accurate - including the inaccuracies from your spindexer bearing as they grind out.

I think 5C collets are a poor choice for this type of work.
Short range, etc etc.

2.
Suggest an air chuck.
Really fast, open, slip part in, press lever, done.

3.
Reason You don´t get quotes.
Imho.
200 x 8$ / part is 1600$ / month.
Who wants to risk their rep for 1600$ per month, especially when You said 8$ is a lot ...
So are likely to skip vendors for a cheaper offer.

And as You have discovered, getting set up for any custom grinding, efficient, good, is going to cost a couple grand and plenty of hours, low end.

If You offer an 4000$ NRE fee to a grind shop, and 4$ per part thereon, You might have 8/10 takers.

If You cannot afford the 4k$, then the grind shops would do a business mistake/error doing it for You free, no ?

Industrial anything well done is expensive to set up.
But it can be really cheap to run.
You could get 10k pieces run for maybe 1-2$ each. After setup, NRE.

 

[car2]

Thanks again everyone. As mentioned, we're not trying to nickel and dime these parts at all, and have asked for quotes in 2500 and 5000 piece lots, which we would be prepared to order in bulk. We'd also consider higher quantities depending on the price. Never got any quotes with NRE/setup charges (which we would entertain), nor did we state a price-range required and still got mostly no-quotes, or people don't even return emails after speaking with them, and sending drawings. Also, most did/would not send any representative samples of their work. Apparently, my fishing expertise and contacts on that topic is not good, and I assumed that finding someone to do grinding or making standard reduced-shank drill bits would be relatively easy.

Gents, thanks very much for the technical tips as well, on the bushings and collets, much appreciated; the bushing method sounds a useful trick for other tasks. We have some "stub" drills, they're not much shorter than the standard 3/32 (we need the 3/4-ish min flute length, which the "stub" drill is ok for).

Bob, got your email and will respond, thanks.

Again thanks very much for everyone's time and advice.

 

[car2]

Just wanted to report that the 5ST 4-tine collet (3/32") improved things considerably in the KO Goldberg setup. Part-to-part runout is better, and the ST collet grips with much less tightening. We also started dressing the wheel every 12 parts. Hardinge had 3/32 ST collets in stock ($95) Thanks Stephen for that suggestion, and for everyone's input. (Still working on outsourcing these things). Cheers.

 

[michiganbuck]

.050 is pretty small but we have spun a temporary circular point on some small work and came in with a live center to stabilize the out end.. than after diameter was done sharpened the circular to a cutting end.

I think your setup is great.
Another good device/machine for this task is the Royal Oak TC grinder. It can OD a cutting tool diameter and then with little adjustment can come back to back off the OD heal. You uptick roll the angle and give about a half-degree angle bump, to come back to a blue in to see the land.

 

[stephen thomas]

Glad the 5ST's worked ok.

It occurs to me that if they become a wear item (@$95/pop)
Perhaps a DA collet holder and DA collet would work and be cheaper.
The DA holders have an adjustable depth screw in the shank.
The closer is on the nose (might be convenient if you are not using a lever or air closer)

A DA holder could be held in the V-block on a Suburban Master grind,
Or in a 5c collet in your current set up - subject to some fussing due to stack tolerance accumulation.
I'm sorry i did not mention that option, but my first suggestion did not fully comprehend your impressive set up and volume!

DA 200 or the smaller DA300 series should hold those drills, and they are 4-split at both ends for better parallel hold.

Most of mine are auction or eBay source Hardinge & Erickson/Kenametal, but this company keeps coming up and may be cheaper for the holders, if you have to buy new. The page also included a quick cut-away drawing so you can see the options.

Craftsman Ind., Inc. - Made in USA Collet Chucks - DA, ER Toolholders

PS found Hardinge page:
DA300, DA200, DA100, and DA180 Collets | Double-Angle Collets | T

smt

 

[car2]

Thanks again Stephen and everyone for all the helpful and informative info and tips on workholding and equipment; I've stored them away for future grinding project reference. Hopefully the KO Goldberg setup will suffice until this task is vended out; it seems to be working sufficently well, and I'm trying to avoid making the task more complicated or time-consuming, so will see if this holds up. I think that ST collet will hold up pretty well, we clean the collet and spindle taper when starting it up each batch, and not too concerned with the cost of additional collets. Otherwise will look into the DA/ER (I have a pretty good selection of acura-flex collets used in KS200 tooling). Thanks again everyone, hope everyone had a good Christmas and Best Wishes for a good 2021! Cheers, Charles