[acannell]

i just got a dyna myte DM2900 mini mill. its travel is something like 10" x 7" x 10" (specs are hard to come by and i cant measure the travel yet)

i would like a 4th axis, but im not sure exactly how MUCH 4th axis I need. I think i would get the biggest bang for the buck, for the kinds of parts I make, using it as an indexer to allow multiple planes of machining without requiring multiple fixtures and/or manual refixturing. I doubt I would ever actually have true 4 axis parts being made. so it can be slow.

I'd like to look into making my own 4th axis. I have access to an accurate manual-lathe with DRO to make parts for it.

Some "big picture" things I am not sure about though, and would really appreciate the advice of more experienced folks who have used 4th axis, are:

-how big does the 4th axis have to be to not significantly detract from rigidity? i know this is a complex question, but at the moment I have no idea how much rigidity im going to lose. (thats what he said)

-my mill bed is pretty small, maybe 9" x 18". im not sure how much room that gives me for a 4th axis. maybe this is all moot.

-what basic parts am I going to need? i'm assuming ill be buying a precision bearing, but i probably dont need a chuck as most likely ill be bolting fixtures directly to the 4th axis. i.e. assuming a textbook design, how many bearings and how are they supported at both ends? what kind of shaft diameter are we talking about? i.e. whats the bare bones "body plan" of a simple 4th axis?

-does the motor holding torque become a significant requirement or does the transmission somehow lock things in place?

-since i dont need it to be fast, and i can make some of its parts myself on my mill or lathe, im assuming i can save a significant amount of money here (yes i want to do it CHEAP. there I admit it.) heres a design i think would work, or at least a starting point

[awerby]

Are you kidding? This looks like an old-fashioned water wheel set up to drive a grist mill (those are entirely different from the milling machines we talk about here). Do you really plan on making a medieval-style 4th axis out of wood and stone? It might be fun to watch...

Here's a link to a "ghost view" of the Sherline 4" rotary table, which I've used with some success: 4" Rotary Table With that gear ratio, the motor's holding torque has no problem keeping it from shifting its position. The limiting factor isn't really the XY dimensions of your mill's bed, but the Z axis clearance has to be sufficient to allow your longest tool to clear the fattest stock you plan to cut.

Andrew Werby
www.computersculpture.com

[acannell]

Are you kidding? This looks like an old-fashioned water wheel set up to drive a grist mill (those are entirely different from the milling machines we talk about here). Do you really plan on making a medieval-style 4th axis out of wood and stone? It might be fun to watch...

Here's a link to a "ghost view" of the Sherline 4" rotary table, which I've used with some success: 4" Rotary Table With that gear ratio, the motor's holding torque has no problem keeping it from shifting its position. The limiting factor isn't really the XY dimensions of your mill's bed, but the Z axis clearance has to be sufficient to allow your longest tool to clear the fattest stock you plan to cut.

Andrew Werby
ComputerSculpture.com ? Home Page for Discount Hardware & Software

lol yes im kidding..trying to weed out anyone without a sense of humor..ive had it up to my f8***ing eyeballs with smug jerks on forums

anyways, thats a great idea with the sherline 4th axis. ironically this mill is my upgrade from a sherline, and i wouldnt try using the sherline 4th axis on the sherline..there just wasnt enough room when i tried it and the rigidity of the whole mill was too flimsy to use it seriously. i may try it again! I got the cnc one with stepper for $320 on ebay, that seems hard to beat even DIY, and it may be as "much" 4th axis as my new mill can handle, although i bet i could go a little bigger.

good idea!

[handlewanker]

Hi, I expect you know about backlash and also stepper holding power.

I read that the 4th axis selling on Ebay that goes with the 3040, 6040 etc routers for around $350 has little holding power.

For what it's worth if you have access to a lathe you could put one together using just deep row ballraces and drive it the same way with a toothed belt and stepper motor.

Your main outlay would be for the 2 bearings and the stepper motor with a toothed belt and pulleys to suit.

If you are up to doing some welding and turning, then I would get some thick wall square tubing and weld two thick steel covers on the ends, bored out to take the bearings.

Being a 4th axis it won't be subject to high RPM's or huge thrust loads, so the deep row ballraces held in with cover plates will suffice.

Even more simpler, you could also go for two 25mm bore plumber block type bearing mounts that just bolt down onto an alluminium base plate, and being self aligning will not be a problem to align on the base plate.

They cost about $15 each on EBAY, so plenty scope for design.

You only need to make a simple spindle to have either a thread for a chuck backplate or just a plain face plate......a centre is a must have to hold longish shafts and a tailstock too....very basic stuff.

To adjust the tension of the toothed belt you will need a jockey pulley to one side against the back of the belt or have the stepper on a pivot.

That design is simple compared to a worm drive gearbox type that cost quite a bit just for the worm drive box alone, and they have backlash you can't completely remove.
Ian.

[acannell]

Hi, I expect you know about backlash and also stepper holding power.

I read that the 4th axis selling on Ebay that goes with the 3040, 6040 etc routers for around $350 has little holding power.

For what it's worth if you have access to a lathe you could put one together using just deep row ballraces and drive it the same way with a toothed belt and stepper motor.

Your main outlay would be for the 2 bearings and the stepper motor with a toothed belt and pulleys to suit.

If you are up to doing some welding and turning, then I would get some thick wall square tubing and weld two thick steel covers on the ends, bored out to take the bearings.

Being a 4th axis it won't be subject to high RPM's or huge thrust loads, so the deep row ballraces held in with cover plates will suffice.

Even more simpler, you could also go for two 25mm bore plumber block type bearing mounts that just bolt down onto an alluminium base plate, and being self aligning will not be a problem to align on the base plate.

They cost about $15 each on EBAY, so plenty scope for design.

You only need to make a simple spindle to have either a thread for a chuck backplate or just a plain face plate......a centre is a must have to hold longish shafts and a tailstock too....very basic stuff.

To adjust the tension of the toothed belt you will need a jockey pulley to one side against the back of the belt or have the stepper on a pivot.

That design is simple compared to a worm drive gearbox type that cost quite a bit just for the worm drive box alone, and they have backlash you can't completely remove.
Ian.

thanks this is great info!! have you made your own? it sounds like you have, or are planning to.

[handlewanker]

Hi, nope not yet, I'm seriously thinking about getting a 6040 router, but till then it's just forward planning.

The "design" concept is pretty basic and you can go overboard with the specs and still end up with the same working model.

If you look at the 6040 routers on EBAY especially the ones with a 4th axis you can get an idea of what it looks like and how you would DIY build one using stock aluminium sections........coming in at $350 means you can save a lot if you have the ability to do turning and maybe milling.

They come with an 80mm 3 jaw self centring chuck, so that will cost you about $75 for a start (highly desirable) if you want to incorporate one in a design.

By the time you source all the other parts like the stepper motor, bearings, two toothed pulleys, a toothed belt, metal, screws etc, you might just save $50 which is not a lot when you have to do all the building work.
Ian.

[acannell]

Hi, nope not yet, I'm seriously thinking about getting a 6040 router, but till then it's just forward planning.

The "design" concept is pretty basic and you can go overboard with the specs and still end up with the same working model.

If you look at the 6040 routers on EBAY especially the ones with a 4th axis you can get an idea of what it looks like and how you would DIY build one using stock aluminium sections........coming in at $350 means you can save a lot if you have the ability to do turning and maybe milling.

They come with an 80mm 3 jaw self centring chuck, so that will cost you about $75 for a start (highly desirable) if you want to incorporate one in a design.

By the time you source all the other parts like the stepper motor, bearings, two toothed pulleys, a toothed belt, metal, screws etc, you might just save $50 which is not a lot when you have to do all the building work.
Ian.

i think i need to step back for a minute and really consider what i need the 4th axis for. one of the parts I make needs to be machined from 4 different sides, and not at angles which are at right angles to a feature that can be gripped by a vice, so i have to manually hold them in the vice and use a protractor to get the right angle. that obviously has to stop.

i could just make angled fixtures which allow the piece to be held in the vice at the correct angles, but id like to replace doing even that with a 4th axis. so i just have to visualize how that is going to happen in space, and then how you would realistically connect a 4th axis to the part to do it. if i remember correctly, when i machine the part, i just rotate it around one of its axis, so that should be easy to do on a 4th axis, i would just bolt the part to a fixture plate that is connected to the 4th axis, and also supported on the other side with a tailstock of some kind. so i think i need to look at the design from that perspective.

i have an extra sherline headstock, i bet i could use it for this. i also have quite a few stepper motors, ill try to figure out which has the highest holding torque.

heres the part im talking about, and a picture of it at various angles as it gets machined. the vice always clamps the same sides of the part, it just gets rotated at different angles. so technically i just rotate it around one axis to get the different planes.

[acannell]

Wait..this might be dimb..but what about buying a used but nice manual rotary table for a couple hundred and then hooking up a stepper with encoder to operate it? How bad are those yuasa rotary tables for 250 on ebay?

[handlewanker]

Hi, .....short answer....worm and worm wheel = backlash.

I take it you would be using the rotary table mounted with the rotary table at right angles to the mill table?

There would be no provision to lock the table from turning against backlash once indexed to position, unless you do that manually which defeats having a stepper drive the worm.......even a new table has to have a degree of slackness to turn freely.

I think other people have worked with these tables, so one of them might join the discussion if they see it.

Even the 4th axis on EBAY that goes with the 6040 routers probably won't suit your needs as it relies on the tension on the stepper/belt to maintain position, and I think with some drilling that you show the belt would flex too much.

Possibly one like that would work better with a chain or gear drive also depending on the strength of the stepper motor to hold position.

You could get very adventurous and mount a small 80mm 3 jaw chuck directly on the powerfullest stepper motor you have, but it would have to be able to resist the forces of drilling etc......if it worked you'd have infinite, within reason, positioning ability for 360 degrees, but in one plane only.

BTW, what's your volume and how many of these items do you want to do over a week or so?
Ian.

[COLOR="silver"]- - - Updated - - -[/COLOR

[handlewanker]

Hi again, looking at your part, firstly the stock piece appears to be overlength needing the final end machined off before you round the corners around the bolt hole area......why is this necessary to be over length, it's being held by the side material flats?

I think you can mill this piece in one set-up (3 moves) against a face plate on a 4th axis or a dividing head/rotary table on it's side.......in manual mode if a dividing head/rotary table.

A dividing head normally has a chuck, but you can make a small face plate with a shank to hold in the chuck......a rotary table only needs a face plate bolted to it.

The face plate has a couple of holes drilled and tapped into it to hold the job, and also a couple of small locating pegs in the face to locate the aluminium blank.

The same applies to the 4th axis, face plate held in the chuck.

The job is positioned on the faceplate and clamped with two clamps across the face of the job, leaving the other side faces clear to be machined.

At zero starting position, the block is faced off and the side is milled to the width required.....you will need a packer between the job and the plate to allow the cutter to clear the plate.

If you're using a 10mm diam cutter you'll need a packer at least 12mm thick to allow the cutter to pass across the face.

It's much easier to machine the face plate away to clear the cutter where necessary.

The small hole that goes down to the big hole is drilled and chamfered ( countersink?).

The 4th axis is now turned by the stepper to do the angled face, or the dividing head/rotary table is cranked round manually to the angle required.

Mill the block to give the angle face.

Bore the big hole.......the finish in the photo of the bore looks very rough.......it can be given a final clean up pass with the boring bar if you want it smooth.

4th axis (or dividing head) is turned again to drill the two holes and chamfers.

The radii on the corners are cosmetic, and due to the clamps being across the face need to be put in by hand with a linisher or bench mounted disc sander having a table at right angles to the disc, but as 95% of the work has already been done by the machine, the radii are just 5 min clean-up finishing ops.

The disc sander can also give the other faces a quick clean up to remove small machining burrs and make it look good.

You save material too by not being overlength.

If the volume was high enough this job can be almost fully automated with the aid of a 4th axis and an ATC under CNC control, needing just the corner radii for finishing.

I expect it's easier said than done.
Ian.

[acannell]

Hi again, looking at your part, firstly the stock piece appears to be overlength needing the final end machined off before you round the corners around the bolt hole area......why is this necessary to be over length, it's being held by the side material flats?

I think you can mill this piece in one set-up (3 moves) against a face plate on a 4th axis or a dividing head/rotary table on it's side.......in manual mode if a dividing head/rotary table.

A dividing head normally has a chuck, but you can make a small face plate with a shank to hold in the chuck......a rotary table only needs a face plate bolted to it.

The face plate has a couple of holes drilled and tapped into it to hold the job, and also a couple of small locating pegs in the face to locate the aluminium blank.

The same applies to the 4th axis, face plate held in the chuck.

The job is positioned on the faceplate and clamped with two clamps across the face of the job, leaving the other side faces clear to be machined.

At zero starting position, the block is faced off and the side is milled to the width required.....you will need a packer between the job and the plate to allow the cutter to clear the plate.

If you're using a 10mm diam cutter you'll need a packer at least 12mm thick to allow the cutter to pass across the face.

It's much easier to machine the face plate away to clear the cutter where necessary.

The small hole that goes down to the big hole is drilled and chamfered ( countersink?).

The 4th axis is now turned by the stepper to do the angled face, or the dividing head/rotary table is cranked round manually to the angle required.

Mill the block to give the angle face.

Bore the big hole.......the finish in the photo of the bore looks very rough.......it can be given a final clean up pass with the boring bar if you want it smooth.

4th axis (or dividing head) is turned again to drill the two holes and chamfers.

The radii on the corners are cosmetic, and due to the clamps being across the face need to be put in by hand with a linisher or bench mounted disc sander having a table at right angles to the disc, but as 95% of the work has already been done by the machine, the radii are just 5 min clean-up finishing ops.

The disc sander can also give the other faces a quick clean up to remove small machining burrs and make it look good.

You save material too by not being overlength.

If the volume was high enough this job can be almost fully automated with the aid of a 4th axis and an ATC under CNC control, needing just the corner radii for finishing.

I expect it's easier said than done.
Ian.

thanks for the detailed advice! i think you are right that its doable automated on the 4th axis. this was a prototype piece and the first time i machined it so i left the stock long just in case i needed extra material to hold it in the vice.

when i did it manually, i only rotated the stock along the x-axis. when i got to the part where i did the outer profile, i bolted it to a separate plate which made it possible for cutter clearance on the sides and depth, re-zeroed, and then profiled it.

i would imagine i could duplicate the same thing (rotating it along just one axis) with a 4th axis, but it may have to be a fairly stout setup and with enough spacing off the plate to allow for the tool and possibly tool holder to get in there. i think i need to do some mock up and measurements to see what im dealing with.

this is a good opportunity for me to incorporate a 4th axis into my skill set...i.e. its a part ive already made, and one ive made only rotating in one axis. in other words i have a perfect excuse to buy another tool! lol

its a no-no to try and support the fixture plate on some kind of relatively thin "bar" between the 4th axis and a tailstock right? so im going to try and see this as "fixture plate bolted to 4th axis face"..and figure out how thick it needs to be to give clearance.

the question i would have about rotary tables out there, is how is backlash dealt with in modern 4th axis and manual rotary tables? i cant imagine its an issue on a $5k+ 4th axis for a vmc..or is it? yes for the ebay belt drive one i think its out. i need high rigidity for this (thats what she said)

can most manual 4th axis be locked with a separate handle or something? maybe a second stepper could be added to do that function

one thing i do have is lots of stepper motors, and lots of easy-style stepper motor controllers (i.e. advanced controllers with encoder feedback and usb or serial ports for PC control)

[sudsy]

Just thinking outside the square but I cant see any reason why you couldn't use a car air conditioner clutch as a brake on a rotary table to hold it firm. Just have to set up the electronics to disengage brake before rotating and then brining it on when in position. Easily available at wreckers yards cheap and plentiful 12 or 24v. Just something to look at

[acannell]

Just thinking outside the square but I cant see any reason why you couldn't use a car air conditioner clutch as a brake on a rotary table to hold it firm. Just have to set up the electronics to disengage brake before rotating and then brining it on when in position. Easily available at wreckers yards cheap and plentiful 12 or 24v. Just something to look at

i like this...since i dont need a true 4th axis it should work. i really just need a CNC indexer. so if i use a locking system like a clutch, the demands on holding torque go down..maybe the demands on other stuff too..just use a wimpy motor to position things with encoder feedback, then lock everything..

assuming the clutch doesnt somehow cause alignment errors when it engages or who knows what

[sudsy]

No reason why it would cause any alignment issues as the just pull straight in and release straight out. As long as everything is setup with the clutch engaged it doesn't matter what it does when disengaged to rotate as long as it comes back in line when locked again

[Hytek]

Just a thought, but have you considered using a servo motor to drive your 4th axis? Most I have used come with a brake built in. As mentioned earlier, if you get your gearing right I think this would be more than ample. Also they nearly always have either an encoder or resolver attached to them. This may be a bonus to have for future upgrades and such.

Another thing to look into is the possibility of adding the 4th axis to your mill and controlling it through the Dyna software. I can't speak for your model but I know on my Dyna it's a case of adding a few components and enabling the axis. If something like this might be on the cards then you're gonna need to go servo anyways being as that's what the Dyna mills run.

Whichever way you go, good luck