[bschwand]

Hi everybody,

I am thinking of building an automatic tool changer for my Sieg mini-mill. I read the thread about the Hoss ATC, seems sweet. But I do not like the massive bulk of the piston, levers, etc. on top of the mill head, plus requiring a compressor.

I was thinking, why not use ER collets ? Each tool in its collet and nut, and the tool changer just needs a mechanism to engage the nut and unscrew it. A set ring could be used on each mill to set the depth in the collet. On ER 25 the torque spec is 77ft/lb, which is not that huge. Using ball bearing ER nut would also help. So the changer would take the mill, collet, nut together, and exchange for a new set on each tool change. ER collet and nuts are cheaper than separate tool holder like the tormach ones.

Has anybody thought or tried this ? Any comments ? Would it work ? Do you see any problems with this scheme ?

[pippin88]

I'm not sure I can it see being accurately repeatable.
Would need to be precisely designed to get the threads picking up right - one cross-thread and you'd be in trouble.

[SCzEngrgGroup]

I've now designed two complete power drawbar/toolchanger systems, and I can tell you that what you describe will be far more complex, and difficult, to get working reliably than you think, and will have absolutely no functional advantage. Why is the "massive bulk" of the power drawbar even a concern? There's usually nothing else going on up there, and the mechanism need not be terribly heavy, so that should not be an issue. And, if it is, why not simply design it differently? A power drawbar can be made VERY compact - I have one with a "foot-print" of only 2" x 3.5", and they can be made smaller still, by going to a hydraulic, rather than pneumatic, system. 77 ft-lbs is achievable, but far from trivial to achieve, especially on a budget. And it will require the addition of a (very strong) spindle lock (a lock, not just a brake). Then there's the time, and quite complex mechanism, it will take to lock the spindle, grab the collet nut, unscrew it, remove the tool, insert the new too, re-install, and torque, the collet nut, then set the length. Toolchanges will be slooooow, and there will be many, many opportunities for things to go wrong. What would be the practical/functional advantages of the system you describe over a much simpler, conventional design? What problem are you really solving, other than, apparently, and aesthetic one?

Regards,
Ray L.

[bschwand]

I have a longer term project of a large parallel kinematics router/milling machine using a high speed spindle. Such an ATC like I describe would allow to fit this system and also all gantry router tables using high speed spindles, which generally use ER type collets.

A sensor can be used for setting the tool height after change so I am not worried about repeatability.

pippin88: good point about picking the threads correctly. I think it would not be too difficult, when inserting the new tool, to press the nut (spring loading ? ) and rotate counter clockwise until it clicks, which means the thread has just been passed. Then rotate clockwise. Detecting when it clicks could be done with an accelerometer, but I do not know how reliable that would be.

HimyKabbible : functional advantage one is to use normal high speed spindles with ER collets. Advantage 2 is lower weight on spindle and head (less inertia, faster moves, less strain on motors, lower power motors required) . Advantage 3 is reduced size of spindle and head (better reach in some situations...)

Complex mechanism to lock the spindle: We just need a key or wrench to engage the flats of the ER collet holder. A wrench is pressed against the shaft, the nut is rotated and the wrench engages once it reaches the correct position. Am I missing something ?

Engage the nut: again a wrench (well like a tube or flat) is pressed against the nut and rotated, when it reaches the correct position it engages, drives the whole shaft until the spindle lock wrench engages, continues and unscrews the nut.

It can all be spring loaded and have one strong rotating mechanism to unscrew the nut and engage the spindle lock.

When the nut is unscrewed, it is lowered with the tool and collet, put away, then a new collet, tool and nut is grabbed and raised to the spindle/collet holder.


Thanks for the replies, all good points, that is why I posted !
more comments and suggestions welcome !

[handlewanker]

Nope, too complicated when you have an existing holder replacement already set in a chuck type of method in one form or another.

If you want to go simple, just have an ER32 chuck in your spindle with a 3/4"collet and get some ER chucks with short 19mm straight shanks and do the tool changing with preloaded straight shank ER32 chucks, one for each tool.

If you have a 3/4" collet in an ER32 chuck on the mini mill, and you go to a 19mmshort shank on your tool chucks, and knowing that there is about 2 thou difference in the 3/4"and 19mm diams, when you just slacken the nut on the mill chuck half a turn the straight shank tool chucks will come out freely.......you just need to devise a method to slacken and tighten the mill spindle nut for ATC mode when removing the tooling.......the spindle nut and collet does NOT have to be removed as you would normally for each tool or cutter change.

You also don't need to bang the end of the spindle to shock the tool out of a taper mount, and the ER system will cater for up to 1mm difference in sizes when you insert the new tool shank.

All of the tool shanks will need to go to the same depth in the spindle chuck so all you need to do is mount your cutters in each straight shank chuck to a depth with a gauge.

You can also use this method for a manual set-up when it become a pain to keep removing the chuck nut to change collets for a number of tools etc........all you lose is about 50mm of spindle height with the chuck in chuck.

I can't see any disadvantage in having a mini mill spindle with an ER40 collet taper (for maximum tool shank size) built into the end of the spindle and use parallel shank chucks to mount the tools.

This would make a very compact tool mounting method compared to the ISO30, 40, R8 or MT system, and no draw bar to worry about, as the spindle is the chuck.

Now that I've thought about it I might go on Ebay and cost up a number of straight shank ER chucks for a system for my self, all in ER32 size as I've already got a set of ER collets, and any extra ones are now so cheap.
Ian.

[bschwand]

sound like a cool idea, but I am not sure that stacking collet holders would be so great for rigidity and runout, wouldn't it ? I really like your idea though, a mechanism to just slacken the ER32 nut would be much easier to do than something that can take the nut off and rethread it back. hmmm more food for thought.

[DLA_Dezign]

I'd like to know how this person did his ER air cylinder setup:

[DLA_Dezign]

Sorry, apparently the forum had to approve my video post so it ended up being a double post.

[handlewanker]

Hi, up to a point you are quite right.......yo would also lose some height if you have a collet chuck in a collet chuck, but as I envisage a 12 mm cutter or thereabouts, the cutting forces would not be all that great.

The advantage of being able to have premounted chucks with cutters of different shaps and diams would make it a quick change idea even for manual set-ups.......ATC is a design challenge.

I did a test on my own mill with an ISO 40/ER32 chuck in the spindle and an ER32 chuck with a 19mm straight shank in the chuck and using a plain piece of 10mm round HSS to simulate the cutter.

Runout was not more than a thou at most and improved somewhat by removing and replacing the chuck.

I would agree that, for maximum rigidity, the cutter close to the spindle end and in a collet like the R8 system in the Bridgeport I used to have, is the ideal, but as R8 collets mean unscrewing the drawbar every time it would be slow.

If the mini mill had the spindle with an ER taper in the bore and threaded on the end for a nut for the closer, the stick out would be the same as a regular mill spindle with ISO tapering systems and a collet chuck......the advantage being that now you could use all threaded or unthreaded milling cutters directly in the spindle end and the plain spindle chucks without loss of rigidity for the quick change method.

Changing tools for a repetition job in the ER system is quicker if you have the cutter, collet & nut as an assembly.....(spare nuts and collets are available on EBAY at minimal cost), only drawback is you lose cutter height position when you remove it, whereas with the parallel shank cutter/chuck assembly the chuck always goes down to the bottom of the collet each time.
Ian.

Just as a matter of interest and cost feasibility, I browsed Ebay for ER32 chucks with 19mm X 50mm shanks for a cost and got $A24 + $10 for mailing from Hong Kong....probably cheaper mailing for more than one chuck.

This method would only really appeal if you were doing a bit of repetition work where a number of peripherals were being used, like a 9mm end mill + a fly cutter + a ball nose cutter + a keyless drill chuck with parallel shank and so forth. making toolchanging a quick process as opposed to undoing the draw bar to change chucks or the chuck nut to change collets and cutters etc.
Ian.

[handlewanker]

Sorry, post duplicated.

[SCzEngrgGroup]

How is that any better than TTS, other than being more complex?

Regards,
Ray L.

[handlewanker]

Hi Ray, simplicity, nothing complex about it, you just stick the end up the hole and tighten the nut... even my son who doesn't even have a drill press agrees......if you have the dosh for a couple of parallel shank chucks....you can use smaller chucks too...ER11's etc... for the very small end mills if you can get them in parallel shanks all the same size......it's a matter of choice what you do, and you will get concentricity each time.

I have one that is ER32 and has a 100mm shank length to get you down when the side clearance is obstructed.

What ever is TTS?
Ian.

[SCzEngrgGroup]

Hi Ray, simplicity, nothing complex about it, you just stick the end up the hole and tighten the nut... even my son who doesn't even have a drill press agrees......if you have the dosh for a couple of parallel shank chucks....you can use smaller chucks too...ER11's etc... for the very small end mills if you can get them in parallel shanks all the same size......it's a matter of choice what you do, and you will get concentricity each time.

I have one that is ER32 and has a 100mm shank length to get you down when the side clearance is obstructed.

What ever is TTS?
Ian.

TTS = Tormach Tooling System, which is the de-facto standard used for toolchangers on small machines. Virtually every toolchanger you'll find in this forum uses TTS: http://www.tormach.com/product_tts2.html

Uses a simple, modified 3/4" R8 collet, and straight-shank tooling to provide quick, easy toolchanges, with highly repeatable Z length.

Regards,
Ray L.

[bschwand]

I like the tormach system, it seems nice. But 30$ for an endmill holder ? I can get ER25 collet and nut for 5$.

[Winnfield]

So the goal is to save a couple hundred dollars on tooling and spend countless hours trying to make it work when a simpler, proven solution already exists?

[bschwand]

like I said
"I have a longer term project of a large parallel kinematics router/milling machine using a high speed spindle. Such an ATC like I describe would allow to fit this system and also all gantry router tables using high speed spindles, which generally use ER type collets."

Is there an ATC for those high speed spindle ? They have an ER holder integrated with the shaft so I see no other solution than exchanging nut, ER collet and mill.

I am open to ideas and suggestions, that is why I posted in the first place.

[victorofga]

you migth take a look on the fact, that automatic toolchangers are using approximately 20 deg angle tapers

while er collets are almost selflocking.. what you can open easyly with wrench.. however simply pulling out a a locked er collet would require tremendous force..

also a point a very minor difference in the tool diameter and collet will be deeper or less deeper in the colletseat.. so regardless of rings, with collets you wont get the repeating accuracy like you get with iso tapers..

[wizard]

I have a longer term project of a large parallel kinematics router/milling machine using a high speed spindle. Such an ATC like I describe would allow to fit this system and also all gantry router tables using high speed spindles, which generally use ER type collets.

Router type spindles have also been known to use ISO 20 and similar smallish taper solutions. These are used where automatic tool changing is needed.

A sensor can be used for setting the tool height after change so I am not worried about repeatability.

pippin88: good point about picking the threads correctly. I think it would not be too difficult, when inserting the new tool, to press the nut (spring loading ? ) and rotate counter clockwise until it clicks, which means the thread has just been passed. Then rotate clockwise. Detecting when it clicks could be done with an accelerometer, but I do not know how reliable that would be.

Why waste the engineering time doing something like this? There are well established tapers for just about any machine you could imagine. Beyond that the nut would be pointing up like a saucer waiting to collect chips and junk. I really don't get the concern about mass on the spindle either, use light weight materials if you have to but all we are talking about here is a very short stroke air cylinder suitable for the energy required to compress the draw bar springs.

HimyKabbible : functional advantage one is to use normal high speed spindles with ER collets. Advantage 2 is lower weight on spindle and head (less inertia, faster moves, less strain on motors, lower power motors required) . Advantage 3 is reduced size of spindle and head (better reach in some situations...)

I don buy most of the above either, the draw bar assembly sits on top of the spindle assembly and shouldn't get in the way that often.

Complex mechanism to lock the spindle: We just need a key or wrench to engage the flats of the ER collet holder. A wrench is pressed against the shaft, the nut is rotated and the wrench engages once it reaches the correct position. Am I missing something ?

It sounds easy doesn't it?

Engage the nut: again a wrench (well like a tube or flat) is pressed against the nut and rotated, when it reaches the correct position it engages, drives the whole shaft until the spindle lock wrench engages, continues and unscrews the nut.

I really believe if this was worth the trouble we would see commercial systems with such mechanisms. Or look at the other way, automated assembly systems still cross thread nuts onto bolts.

It can all be spring loaded and have one strong rotating mechanism to unscrew the nut and engage the spindle lock.

When the nut is unscrewed, it is lowered with the tool and collet, put away, then a new collet, tool and nut is grabbed and raised to the spindle/collet holder.


Thanks for the replies, all good points, that is why I posted !
more comments and suggestions welcome !

Is the word headache in your vocabulary? Really I ask because that is what you would end up with.

[wizard]

like I said
"I have a longer term project of a large parallel kinematics router/milling machine using a high speed spindle. Such an ATC like I describe would allow to fit this system and also all gantry router tables using high speed spindles, which generally use ER type collets."

Except most gantry systems supporting automatic tool changing use some sort of ISO standard taper. I don't buy the idea that the ATC hardware above a spindle needs to be a problem. Design the spindle and hardware so it isn't a problem.

Is there an ATC for those high speed spindle ? They have an ER holder integrated with the shaft so I see no other solution than exchanging nut, ER collet and mill.

I am open to ideas and suggestions, that is why I posted in the first place.

I'd look up some of the larger gantry router builders and see how they handle ATC. I'd be willing to bet the vast majority are using some sort of ISO or other standardized tool system. For example these guys: http://www.thermwood.com/cnc_router_three_axis_main.htm use HSK style tapers. Here is what PDS has to say: http://www.pdscolombo.com/engineering_clamping.php. You will find that the router industry seems to be moving to HSK or HSD type spindles. ER collets are often used with those tapers for actual tool holding.

If you are serious about building a router type machine to the point of implementing a tool changer, looking at the various electro spindles on the market seems like a smart move.

[victorofga]

wizard youre rigth.. just because we have lathe or mill.. even they re cnc equipment doesnt mean we could produce competitive products comp-ared with a factory..

regardless of brands if a product like this atc not produced in a massproduct manner it will becomes very very expenses..

while its not impossible technically to build an atc i cant imagine to make it really affordable.. i mean in small quantities..