Just wanted to show off my spindle upgrade. I had bought a Bosch Colt palm router about the same time I bought my Taig from Nick Carter. I tried to make an erzatz mounting bracket for it, but didn't really like the results. I also didn't like being limited to 1/4 shank tooling. I shelved the project, and learned to cut with the conventional ER-16 spindle.

Still, I wanted something more in the performance department from my mill. So I looked around, saw that K2 sold a bracket for the Colt, and ordered it. Simultaneously, I found a source for precision collets and a closer nut from precisebits.com. I ordered the parts, fabbed up a mounting plate for the bracket, and here are the results. A Taig CNC mill with a 1 HP spindle, capable of up to 35K rpm. We'll see how the bearings hold up to extended duty, as my tool paths can take hours, if not days, though the Bosch is reputed to be quite solid.

And if I need to switch back, it's only a matter of 4 10-32 bolts.

Here's the pics...

http://johnbearross.com/oldspindle.jpg
http://johnbearross.com/newspindlemount.jpg
http://johnbearross.com/newspindlecolletandmount.jpg
http://johnbearross.com/newspindle1.jpg
http://johnbearross.com/newspindle2.jpg
http://johnbearross.com/newspindle3.jpg

Best,
John Bear Ross

Interesting, John as I considered doing this same exact thing at one point.

Unfortunately I saw the inconsistencies in the Bosch Colt laminate trimmers as I own one and the runout is just too great for the work I do.
http://www.precisebits.com/lab_reports/bosch_colt_TIR.htm

How much runout are you getting and, if you don't mind me asking, what kind of work to you do? What's the smallest cutter you use?

Thanks for sharing-
Dave

Dave, that link you posted says they've only seen two Colts that have not worked with their precision collets.

Dave, that link you posted says they've only seen two Colts that have not worked with their precision collets.

Hi Gerry, yes but that's what "they" have found ;)

And mine would be a 3rd-

Dave

Hi Gerry, yes but that's what "they" have found ;)

And mine would be a 3rd-

Dave

Have you measured the runout on your Colt without the standard Bosch collet and nut? Was it new when you got it? Any play in the bearings?

-James

Have you measured the runout on your Colt without the standard Bosch collet and nut? Was it new when you got it? Any play in the bearings?

-James


Hi James,

Don't get me wrong, I love the colt by bosch as I use it for numerous other jobs, but it's just not near as accurate as my taig spindle. I do small engravings and I just can't afford to have any runout at all but I would love to have the versatility of the Laminate trimmer speeds and the mounting.

The colt taper is not "true" as the runout is indeed .003-.004" with a bit installed and to purchase the precisebits.com collets and nut would be useless at this point or at least a crap-shoot (for me!). I think that's why they (precisebits.com) say to try and buy the collets first and THEN go buy a colt and try the collet in the spindle to check for slop. Well that's going to be a needle in a haystack kind of thing and I'm sure the store would just love the customer opening up all the boxes just to find a good one, lol!

This is why I asked John, to see what his runout was and what kind of work he was doing. Not trying to stir anything up here at all.

Thanks,
Dave

EDIT: Yes the colt was indeed new and no, no play in the bearings.

John, that mounting plate looks really nice. Did you fab it on the Taig? Also, what material(s) would you cut with a laminate trimmer at 35K that the 10K standard mill doesn't do so well?

Not to worry, Dave.

I'm doing sci fi models in wax and delrin, maybe epoxy, too. Nothing too high-tolerance. I'll check the runout with a dial indicator tonight.

Best,
JBR

John, that mounting plate looks really nice. Did you fab it on the Taig? Also, what material(s) would you cut with a laminate trimmer at 35K that the 10K standard mill doesn't do so well?

noisillator,

I did fab it on the Taig. If the spindle head knew it was assisting in creating its replacement/supplement, I'm sure it would have gone on strike. Heh.

I'm really on a quest to reduce my tooling marks, more than anything. Surface finish is pretty important to my clientele, and even the tiny milling marks of a 10K rpm cut at 5 ipm brings questions of "can that be buffed off?" I'm just trying to avoid that. I work mostly in wax and delrin, sometimes epoxy, too.

Best,
JBR

The colt taper is not "true" as the runout is indeed .003-.004" with a bit installed and to purchase the precisebits.com collets and nut would be useless at this point or at least a crap-shoot (for me!).

You have .004" runout? Would you mind telling me where you got it? We are trying to narrow this down since it is the vast minority of people that have had issues we think it might be from a single source that these are coming from.


I think that's why they (precisebits.com) say to try and buy the collets first and THEN go buy a colt and try the collet in the spindle to check for slop.

I don't want to say where the 2 we know of came from (you can imagine why) but I don't know of anyone so far that has got one from a retail store that has had a problem.



One last thing if you do ever want to try them and they don't work for you you are alway free to return them per our standard policy:

At PreciseBits our goal is 100% satisfaction. We believe in our products and are willing to stand behind them.

If you are not satisfied with the tools that you have received, for any reason, you can return any of our products back to us, in the original packaging, within 30 days of the invoice date, and we will refund your money*. All you have to pay for is the shipping and for any tools you do not return. It is as simple as that.

We have built our reputation on quality products and services that are delivered in a timely and cost effective manner. That's not going to change.


Ron Reed
President

Inevitable fine print:

* If your payment was made with a credit card or PayPal, the 3% non-refundable processing fee that is charged to us by the credit card company will be deducted from your refund. Carbide bits that have been used in any way will not be covered by this warranty, so please do not return dull or broken tools.

This is an interesting idea for a Taig. For certain types of work it should do a good job. Given that the spindle speed only goes down to 15k it would not be good for use with machining steel. For use on wood, wax, PCBs, and aluminum (with smaller cutters) it would work fine.

Mr. John Torrez,

I sent a memo via the website, asking for a 3/16th collet to be made in the future for my new Colt rig, so I can use my tooling with that shank diameter. Any chances of that happening?

Best,
JBR

This discussion is in the TAIG section of the forum, and my comments were aimed towards the applicability of using the Bosh Colt with a TAIG mill. The TAIG cannot do high speed machining as I guess was the point in the PDF you attached. The Taig is not rigid enough nor is a Bosh colt powerful enough.

Please refer to a good feed/speed reference for various steels and you will see my point. You won't be cutting steel with a 3/8 end-mill at 15K RPM on a TAIG.

Bosch specs the speed rang of the router motor from 15K to 35K as that is the range where the motor can generate usable power. So for certain types of work the spindle speed would work great for other work it is not appropriate.

Hi John,

You have .004" runout? Would you mind telling me where you got it? We are trying to narrow this down since it is the vast minority of people that have had issues we think it might be from a single source that these are coming from.

I don't want to say where the 2 we know of came from (you can imagine why) but I don't know of anyone so far that has got one from a retail store that has had a problem.

Well I'm not sure it would be fair to say who it is just as you would prefer to keep your info quiet. Let's just say it's a major retail chain store.




One last thing if you do ever want to try them and they don't work for you you are alway free to return them per our standard policy:


I understand that and that's why I posted the link for all to see. Not trying to ruin anyone's credibility here, just wanted to get John's (Bear Ross) feedback that's all. It would be fantastic to have a retail store that would be willing to do that, I just don't think it's going to happen that way. It may benefit you, since you've only found 2 (and now 3) that maybe you could look into purchasing a batch of these that you know the runout is reasonable, and sell them right off your site as a package. I'll tell ya, if you did, I'd be first in line.

Thanks,
Dave

Hi Dave,

I also am not trying to be contentious, but the problem with the Bosch, Porter Cable and other routers has been found to be the STOCK COLLET / NUT combination. When you put a router bit into that and measure runout you are NOT measuring the actual capability of the router itself.

-James

I know that the donut guys comments were more to troll a fight than actually discuss anything rational, but I thought I'd put another point into the mix;

I do high speed machining of camera parts, often on steel, and with small enough bits (under 1/16") to require speeds that high. On paper it looks like it would work. It doesn't work in real life however.

You can't go over 5% of a bits diameter in runout without trashing it. Even that is pushing it. Runout of .001 won't bother a larger bit, but with small enough bits to require these high speeds it simply destroys the bits.
For bits small enough to require 20k+ RPM, the runout has to be in the tenths. Just not router territory.

Even if a routers runout was only .001 (and it usually isn't) it still won't work for high speed machining of steel as it can't handle tools small enough.

Looking at the tools you found and the 0.020" DOC listed they are showing 90.55 IPM at 28.5K RPM. I don't think the Taig will be stiff enough to pull that off. Also as you move up in bit size if you figure out the actual HP required to make the cuts at the feeds/speeds suggested you'll find that it is more than the Colt can provide and would no doubt put more side load on the bearings than they were designed to handle.

Now, if making circuit boards were your thing you could feed about 35 IPM with the router motor at 35K and it would work great.

Thanks cameraguy! It is nice to see another source of real facts and good conversation.

Jeff Birt,

Here is the high speed, high rpm cutting data you requested.

JoeyB

I have to chip in and say that Taig can indeed do 25k+ RPM machining of steel, but only with a stupidly expensive Japanese spindle costing significantly more than the Taig, and with a runout listed in nanometers.

A router just doesn't work, as much as I would have liked it to.


The bits themselves will snap with only a few ounces side pressure, they can't load the machine. When you get to tiny bits and that high speed, side loading isn't the limiting factor.
Either the chips are clearing or the bit just snaps. Rigidity isn't as big of a deal as you might think. Its all about runout, chip clearing, and damping oscillations.
Its just a completely different form of machining entirely.

This is an interesting project, thanks for posting the pics. I like your clamping bracket, it looks very cool. Will those pockets in the clamp fill up with chips? I might have put tapped holes in the clamp, then countersinks on the mounting plate for screws to come in the other direction to fasten the parts.

Interesting stuff about high speed steel machining with small bits. It makes sense how runout would be the limiting factor in that style of milling. I definitely don't have the knowledge to add anything to the discussion, but I like the approach of just trying something and proving it works. The OP is definitely proving that this mod works really good for his type of application.

JBR,

That is a really nice mount you made.

How much time is the Colt saving you?

Semper Fi,

Jeff...

Jeff,

Thanks, and Semper Fi to you, too. The adaptor plate was just a piece of scrap 6061 I had laying around. It works.

We'll see how much time it saves me when I get back from ComicCon, and fire it up.

Minibeast,
The bracket should serve nicely. Delrin dust and wax chips will probably settle in the pockets, but they're harmless, for the most part. As to the bracket, it's not my design, but it's a solid piece of kit. K2CNC makes a nice product.

Best,
JBR

I know that the donut guys comments were more to troll a fight than actually discuss anything rational, but I thought I'd put another point into the mix;

I do high speed machining of camera parts, often on steel, and with small enough bits (under 1/16") to require speeds that high. On paper it looks like it would work. It doesn't work in real life however.

You can't go over 5% of a bits diameter in runout without trashing it. Even that is pushing it. Runout of .001 won't bother a larger bit, but with small enough bits to require these high speeds it simply destroys the bits.
For bits small enough to require 20k+ RPM, the runout has to be in the tenths. Just not router territory.

Even if a routers runout was only .001 (and it usually isn't) it still won't work for high speed machining of steel as it can't handle tools small enough.

Cameraguy, Precisebits makes some of the finest tools in the industry. Growth into certain markets was being limited by the quality of the spindles. They did some tests and discovere that MANY routers (Porter Cable first) had essentially ZERO runout. It was badly made factory collets causing all the problems. Please look here:

http://www.precisebits.com/products/equipment/bosch_colt_collets_nuts.asp#Bosch_Colt_Collets

either .0006 or .0004 runout specs are available. Really great for mother of pearl and recon stone (Porter Cable - my applications).

-James

Mr. John Torrez,

I sent a memo via the website, asking for a 3/16th collet to be made in the future for my new Colt rig, so I can use my tooling with that shank diameter. Any chances of that happening?

Best,
JBR

Currently demand for 3/16" isn't enough for a production run. I'm not sure why but we have only had 3 people even ask for them :confused:.



You can't go over 5% of a bits diameter in runout without trashing it. Even that is pushing it. Runout of .001 won't bother a larger bit, but with small enough bits to require these high speeds it simply destroys the bits.
For bits small enough to require 20k+ RPM, the runout has to be in the tenths. Just not router territory.

I don't know in steel but we have had multiple customers using a .0156" cutter with at least .001" runout in ebony. As I said I don't know about steel but I would have bet a lot that that would have broken in ebony.


Even if a routers runout was only .001 (and it usually isn't) it still won't work for high speed machining of steel as it can't handle tools small enough.

Goto http://www.precisebits.com/gateways/ColletsNutsHome.htm and click on "Video" under Porter-Cable. Total runout in the router, collet, and tool was .0005"




John Bear Ross:

Let us know how it cuts and what the runout is. I'm quite interested in seeing how the colt does in this setup.

I'm also interested in what you think of the K2 mount. Any thoughts?

Sorry about the 3/16" until we see enough demand for them we can't justify making hundreds of collets.

Currently demand for 3/16" isn't enough for a production run. I'm not sure why but we have only had 3 people even ask for them :confused:.

Let us know how it cuts and what the runout is. I'm quite interested in seeing how the colt does in this setup.

I'm also interested in what you think of the K2 mount. Any thoughts?

Sorry about the 3/16" until we see enough demand for them we can't justify making hundreds of collets.

Not to worry about the 3/16th. It would be nice, but I understand the economics.

I only recently discovered your presence after extensive searching on this forum, though, so it might be a matter of lack of exposure.

The lockup of the bracket to the router body is quite solid. Metal to metal engagement, and with a pocket milled out so the spindle lock button can be pressed while changing bits.

I chucked up a ground .250 dowel pin in my new collet, and set up my old Starrett 196 plunge-back. It didn't waiver, nor did the naked spindle (though there was a thousandths dancing back and forth because of the threads). I'll start making cuts, and will report back.

Best,
JBR

Thanks for the info JBR. I found some interesting info that all of you will get a kick out of.

James (Leonard), I did indeed check the spindle taper itself when it was new and pretty much dismissed the idea of using it as spindle so it has been getting used as a hand tool for chamfering/ rounding over edges etc. Out of curiousity I went ahead and checked it today and low and behold, it's only out by .0015" and yet still no play in the spindle. None. So now I do not know what happened as the only thing I can come up with is I measured it wrong and I wanted all of you to know.

Looks like I may be doing this mod after all as it might come in handy. Again, thank you JBR for getting me to revisit this :)

Dave

JBR, any update?

JBR, any update?

I just did my first test cuts in machinist wax. Here are some notes...

>Cutter used: .125 dia, .750 flute length Atrax 4-flute TiN coated carbide end mill, centercutting (they're on sale right now... http://www.use-enco.com/CGI/INSRIT?PMAKA=ST326-0142 )
>Feed: 15 ipm
>Step-Over: d/7 (.0178)
>Run Time: 25 minutes
>Part size: 1.8 inches wide by 3 inches long by .250 deep
>Stock size: 2 inch wax rod


This cut was for a tank-hull bottom that I do most of my test cuts on. Overall, I've happy, and will move on to delrin next, as well as experiment with smaller bits at higher speeds.

Obviously, my stock isn't as large as my part, but I just wanted to see how it would cut. I initially set the Colt router spindle at 4 for the first half of the cut, and bumped it up to 5 for the second half, just to see a side-by-side comparison in surface finish. Surprisingly, the finish was much smoother at 4 than the higher 5. From the manual...

SETTING RPM
1 16000
2 18000
3 20000
4 22500
5 26500
6 35000

4 is 22.5K rpm, while 5 is 26.5K rpm. Something about that extra 4000 revolutions made the wax heat up and streak, causing a nastier finish compared to the lower setting. The segment of the part done on "4" was very smooth, comparable to a hand-sculpted piece, and I was able to leave fingerprints on it. Not so on the "5" side, which showed tooling marks and a spottier finish.

The pluses of the new arrangement are that the downdraft from the spindle blows off most of the wax chips before they have a chance to re-fuse to the work, which has been a problem before with my wax builds. Also, the Colt performed like a champ. The K2CNC bracket heated up, but that is to be expected on a 25 minute run. The true torture test will be on some of my finish passes that last most of the day. We'll see...

I saw no visible run-out or end-shake by eye, even with the extended length . The precisebits collet and nut held up well for its initial test, even with .250 plunge cuts into wax at high rpm. I'm liking what I'm seeing so far.

Best,
JBR

4 is 22.5K rpm, while 5 is 26.5K rpm. Something about that extra 4000 revolutions made the wax heat up and streak, causing a nastier finish compared to the lower setting. The segment of the part done on "4" was very smooth, comparable to a hand-sculpted piece, and I was able to leave fingerprints on it. Not so on the "5" side, which showed tooling marks and a spottier finish.


Used for aggressive machining of stainless steels, high alloy carbon steels, nickel-based high-temp alloys and titanium alloys.

The tool you listed is a metal cutter which means it's going to have a cutting angle that is less than ideal for wax or plastics. They are more likely to cause melting and have a poor surface finish in plastics or other soft media.

You should be able to run faster with a better finish if you switched to a 2 flute tool designed for soft media. As an example here are some test cuts in HDPE I did recently. The "sweet spot" in these test is where I had a very good surface finish no melting and good extraction of the material.

HDPE

Tool: .1250" x .50" 2 flute upcut
RPM: 20500
IPM range run: 40-180 IPM
Sweet Spot: 165-180+ IPM

Tool: .0625" x .31" 2 flute upcut
RPM 20500
IPM range run 40-180 IPM
Sweet Spot: 145-165 IPM

Tool: .0313" x .13" 2 flute upcut
RPM 20500
IPM range run 40-180 IPM
Sweet Spot: 135-140 IPM

That would cut down your machining time by quite a bit. The only thing that would concern me is that Jeff said he didn't think the Taig could be pushed that fast. I don't have any experience with those machines myself so I don't know if that is an issue. I would still consider it though. Especially since you have the RPM for it now :).

Delrin results are in!

I roughed with a .125 diameter 4-flute end mill at d/5 step/tool paths, 15 ipm, and results came out nice. I did the same with a .250 4-flute end mill, and it hogged out nicely as well, for roughing. Both were at speed setting 1, around 16K rpm.

I then did a finish pass with a .031 diameter ball mill, at speed setting 5 (around 26.5K rpm). Things were moving at 15 ipm. Step/tool paths were at d/9, or .003 inches. Finish and detail are beautiful. Length of cut is only .300 inches on the cutter, though, so I'm limited to that depth for now (per toolpath layer).

Overall, I'm very happy with the results of the spindle swap-out. I'm going to be switching back and forth for stock prep and such, but for fine detail, I like the Bosch.

Best,
JBR

How fast can your machine feed reliably?

How fast can your machine feed reliably?

I'm doing 15 ipm right now. I can try 20, but past experience says I'll start getting binding and missed steps. I have nothing but time, though, and the work I'm doing isn't production.

I'll try, and report. I suppose I should take a picture or two, while I'm at it.

Best,
JBR

TDA - properly configured a Taig can do 100 IPM reliably. There are a few Drive systems for Taigs out there that will barely manage 20 IPM due to a poor selection/matching of components. Take a look at post 14 in this thread: http://cnczone.com/forums/showthread.php?t=84700&page=2, I have my Taig doing 250 IPM, but it was NOT reliable at that speed. Doing 100 IPM was fine, but with a 10K spindle there is not much point.

TDA - properly configured a Taig can do 100 IPM reliably. There are a few Drive systems for Taigs out there that will barely manage 20 IPM due to a poor selection/matching of components. Take a look at post 14 in this thread: http://cnczone.com/forums/showthread.php?t=84700&page=2, I have my Taig doing 250 IPM, but it was NOT reliable at that speed. Doing 100 IPM was fine, but with a 10K spindle there is not much point.

Jeff,

I assume this is with Gecko 540s, right? What about my aging Xylotex? I'm hesitant to cause a situation that releases the magic smoke from my box or steppers.

Best,
JBR

The Xylotex is limited by is basic design (low voltage, type of driver chip). They are great for what they are but they just can't perform as well as the G540. I would suspect that you should be able to reach 15 to 20 IPM though.

Hello JBR,
How is the spindle going? Was wondering have you cut any aluminum with it yet? And how has it held?

John,
Do you have any type of dampening on the steppers? harmonic dampeners? If not then I suspect that your motors are growling something fierce when you are at speed.

Do a search on the zone for harmonic dampeners if you are not using them already. They make a world of difference with your setup.

No aluminum, yet. Just Delrin and Wax. The Bosch spindle is great on those materials.

And thanks for the tip on harmonic dampeners. I'll have to look at that.

Best,
JBR