Hi Everyone,
I am in the process of building a CNC milling machine. I have been very lucky in getting high precision linear rails. These are all in one, lead screw, linear THK rails with 2 pillow blocks each, and THK screw lead inside, NEMA 23 mount for stepper or servo. These used to be in semiconductor inspection equipment, all very heavy duty. Some of the units are listed at 0.000125" resolution. Many of these rail systems are 50+ lbs.
I'm planning on having the Y & Z Axis on a fixed gantry while the X-axis moves. just like a router table. I was planning on using some 8020 stuff to build up the gantry, double bracing all the points. Once I get that up and running, I plan on milling out some parts for the mill afterwards. Milling parts for the mill.
Everything will be running on some Gecko 320 and servo motors. I have all that and electronics up and running on the bench. All the servo motors can turn the lead screws and i can't stop it by hand.

My main question is the spindle. Since I plan on milling AL, plastics, maybe light steel, I'm a bit stumped in what I should do.

1. Use a Dewalt Router or some brand, put a metal milling bit and go to town on it. 10k-20k RPM, haven't bought it yet (Home Depot extended warranty?)
2. Build a servo to spindle, moderate to low speeds
3. Take apart a Sherline metal lathe, just for the head and mount that on my Z-axis. mount a servo motor to the spindle. also not high speed. I'm not using the Sherline since I have my 9x20.

When I do "mill", can I use a Router 10k-20k with a milling bit, 1/4" shank, end mill or ball mill, mist controller, will that work? This is in AL.


Thanks,

Curtis

You typically use high speed for small end-mill, v's ball noses etc. The thicker you go the lower the RPM. I think 10-20 RPM is too fast. At least thats What I thought?...


Also, Look into a Taig spindle head , there cheap, easy too mount and would be pretty nice.


-Jason

Sherline Industrial Products has some spindle cartridges available. I am not sure of the speed ratings for these though.

http://www.sherlinedirect.com/index.cfm?fuseaction=category.display&category_id=32&CFID=3287983&CFTOKEN=91522079

some of the HAAS machines have 10k-30k spindle speeds. Do they cut at those speeds? I was thinking of modeling after them. What would happen if I used a 15k spindle speed and cutting AL, small cuts? it should work right?

The Sherline spindle can go up to 10K RPM if the preload is backed off a bit.

They are light duty spindles though. Only two bearings (one on top, one below) that are pretty small. Quality but not intended for heavy cuts. Milling steel...probably not a good idea.

The blurb up top of the industrial products page says that some components have been "beefed up". I wonder if this applies to some of the spindles over the ones offered on the machines?

As for the router, most(all?) of those have too much runout for metal work. A Zoner made a precision small mill with a high quality router as the spindle but he re-surfaced the taper on a CNC lathe and made his own collets. Also, if you ever want to do steel, the RPM ranges are just too high. Plus you are restricted to very small cutters.

Serge

The rpm you need is dependent on the cutting speed, cutter diameter and cutter configuration. On my machine, with a 6mm dia (~0.25") cutter in aluminium I cut at 1200mm/min (~48IPM) at about 5000 rpm. I can't however cut aluminium at all at above ~2000 rpm without a shed load of coolant - it just welds to the cutter otherwise.

In theory, you could run a 1/4" at 20,000 rpm if you can get your machine to run at 192IPM and it has enough rigidity/spindle power to take the cut. You would need a lot of coolant though.

Quite a lot of DIY builds use a Sieg X2 head which is good for ~4-5k rpm, which you could easily manage with a belt drive and a servo.

A Haas or other modern VMC will go 10-20k and beyond. It will also cost more than your car and maybe your house and maybe weigh more than them too.

You haven't said what you want to want the machine to be capable of. Gantry designs are not as inherently rigid as a bed- or knee-mill configuration. It doesn't matter how good your leadscrews are if the frame bends under cutting forces. There is a reason why metalworking equipment frames are usually made out of cast iron.

All my axis are a "all in 1" design. Rails and leadscrew on a 3/4" to 1" thick AL blocks. All my framing is using 1/2" to 1" thick AL. Most of my material is free from dumpster diving, scrap material I find anywhere. Since my planned cutting area is 8"x16"x5", it's quite beefy.

The X axis is the bed, Y and Z are on the fixed gantry in the middle of the X-axis.

I'm mainly cutting AL and possible steel.

I sent an e-mail to Sherline about the industrial spindles. Specifically about the 6507 cartige spindle with ER-16 and if this was any beefier than standard Sherline spindles. The body is steel so it should be a little more rigid at least. I'll report back with the response.

Edit to add: If you are even going to approach 10K RPM, you need balanced tooling to get a good surface finish. This can be expensive. Unless you will be using 1/8" and smaller endmills all the time, why the need for such high RPM?

Also: It better be one solid gantry to even think about milling steel. The amount of force required to mill steel (even 1020 or free cutting) vs aluminum is a factor of three or more to one.

10 to 15,000 rpm with a 1/4" carbide cutter in 6061 aluminium alloy should be OK. The problem with routers is that they are not normally designed/built for the kind of continuous use they will see on a CNC machine. The cheaper ones have deep grove ball bearings with little or no preload so axial and radial play exists and they burn up quickly. Also the collet taper in the spindle can be "shall we say" a little rough and ready. The best place to go is on one of the CNC router forums and find out what the guys with the experience are using. I finally got a reasonable result with an Italian made ELU router, expensive though. There are good and bad routers for this application. I wouldn't recommend a "buy it and see" approach, I have a box full of bad ones.

Phil


Hi Everyone,
I am in the process of building a CNC milling machine. I have been very lucky in getting high precision linear rails. These are all in one, lead screw, linear THK rails with 2 pillow blocks each, and THK screw lead inside, NEMA 23 mount for stepper or servo. These used to be in semiconductor inspection equipment, all very heavy duty. Some of the units are listed at 0.000125" resolution. Many of these rail systems are 50+ lbs.
I'm planning on having the Y & Z Axis on a fixed gantry while the X-axis moves. just like a router table. I was planning on using some 8020 stuff to build up the gantry, double bracing all the points. Once I get that up and running, I plan on milling out some parts for the mill afterwards. Milling parts for the mill.
Everything will be running on some Gecko 320 and servo motors. I have all that and electronics up and running on the bench. All the servo motors can turn the lead screws and i can't stop it by hand.

My main question is the spindle. Since I plan on milling AL, plastics, maybe light steel, I'm a bit stumped in what I should do.

1. Use a Dewalt Router or some brand, put a metal milling bit and go to town on it. 10k-20k RPM, haven't bought it yet (Home Depot extended warranty?)
2. Build a servo to spindle, moderate to low speeds
3. Take apart a Sherline metal lathe, just for the head and mount that on my Z-axis. mount a servo motor to the spindle. also not high speed. I'm not using the Sherline since I have my 9x20.

When I do "mill", can I use a Router 10k-20k with a milling bit, 1/4" shank, end mill or ball mill, mist controller, will that work? This is in AL.


Thanks,

Curtis

I think the quality of the spindle and tooling will largely determine your results, and matching spindle RPM to feedrate to get the right chip load is important. If you're using 8020 for the structure, you will probably get the best results with a faster spindle, lighter cut and faster feeds.

Lately, I've been cutting 6061 using a 1/4" 2 flute carbide EM at 40,000 rpm & 160 ipm, just light cuts. It gives quite a nice surface finish and - some will reel in horror - I haven't been using any coolant at all.

Perhaps you could build it with a router or die grinder and do some work with it, then look at upgrading the spindle when you know what your needs are, or when you can't stand the noise any more :)

All the best for your build!

Regards,

Jason

I think the quality of the spindle and tooling will largely determine your results, and matching spindle RPM to feedrate to get the right chip load is important. If you're using 8020 for the structure, you will probably get the best results with a faster spindle, lighter cut and faster feeds.

Lately, I've been cutting 6061 using a 1/4" 2 flute carbide EM at 40,000 rpm & 160 ipm, just light cuts. It gives quite a nice surface finish and - some will reel in horror - I haven't been using any coolant at all.

Perhaps you could build it with a router or die grinder and do some work with it, then look at upgrading the spindle when you know what your needs are, or when you can't stand the noise any more :)

All the best for your build!

Regards,

Jason

Are your endmills coated at all - I would love to be able to run dry, but even at 5,000 rpm I am melting the chips and welding them to my TiN coated HSS tools!

Digits - no, they're uncoated, 2 flute Atrax solid carbide on special from Enco. I tried a TiN coated one but it wasn't as good. The best I think I've found was Tialn coated, a 1/4" niagara 3 flute. Unfortunately it was an ebay one off and I can't get more.

Actually, I thought TiN coatings weren't recommended for aluminium? Something about the coating reacting and causing BUE (built up edge). I've been told any coating will round the cutting edges very slightly, since the cutters are sharpened before the coating is applied. Someone with more experience might offer some comment on this?

I think I get away with it because of the high speed and light cuts - the chips get thrown clear and generally the cutter path is pretty unobstructed. I made a video clip that hopefully will show what I mean:

YouTube - CNC milling 6061 aluminium

Regards,

Jason

Cheers Jason - that's a nice vid. What sort of DOC are you using?

The one thing I did manage to make without coolant was a new coolant manifold for my machine - I think I had to run about 0.25mm DOC at 600mm/min/2500 rpm to avoid galling though. With coolant I usually run about 1mm DOC @ 1200 mm/min as life's too short for 80 roughing pases to get through 20mm!

I did try both uncoated and TiN coated 2 and 4-fluters dry without much luck, and also some high-helix 3-fluters that have a black coating, which made me think they might be TiAlN which is supposed to be less sticky in Aluminium.

If I can get my machine fixed, I might buy some solid-carbide end mills as my indexable face mill cuts brilliantly dry.

Fantastic finish! It looks superb! With such low DOC and good speed as yours, the cutting is perfect. Chips also get blown away from air generated from the tooling.

Love the video.


-Jason

One thing that you may not be able too do is use coolant. But I love running my CNC with coolant. I find it too have that "cool" factor...and it also really does help my cuts out. I have welded some aluminum before and it's very ugly.

Anyway, love the vid though.


-Jason

Very nice video. Nice machine too. What is it? Other than a mill, I mean. ;)

I use an X2 spindle on mine run off a 220 VAC controller for a 2 HP DC motor. It runs top speed @ 6600 with the stock bearings and motor and spindle temps are always below 95 degrees.
It's a real nice setup and would fair well on a gantry type machine too I think. If you stick with the stock motor and use a belt drive kit, you can get around 4300 RPM and it is pretty self contained that way.

It would, however stick out pretty far from the gantry if mounted in the standard configuration, but you might be able to fab up a sideways mount. Getting the tool closer to the center of gravity on the axis will help reduce the leverage it would have on anything less than a very rigid system.


A router might be your best bet on such a small machine. Without coolant though, cutting aluminum will be extremely hot on the router and reduce it's life span.
That said, I have often cut aluminum with a PC 690 router on my router table. I douse it with liberal amounts of WD40.


I want to get away from cutting it on my router.
Right now I am installing a Ridgid trim router next to my mill head on my mill.
I will be using coolant.
This is for faster cuts in aluminum. I have to cut sheet aluminum and have to take two passes with the mill to get decent cuts on it. The router will be a high speed type spindle.
It blows air out the bottom very well, so really don't have to worry about my coolant getting in there when it's running. I am thinking about putting a splatter shield on it so when I blow off chips between part changes, it doesn't easily get in there. Just speculation so far. I did get my mount made though.
I used some 3/8" steel plate and built a box.

I will have the info and pics soon on my mill build thread. (http://www.cnczone.com/forums/showthread.php?t=32609)

I want to get away from cutting it on my router.
Right now I am installing a Ridgid trim router next to my mill head on my mill.
I will be using coolant.
This is for faster cuts in aluminum. I have to cut sheet aluminum and have to take two passes with the mill to get decent cuts on it. The router will be a high speed type spindle.
It blows air out the bottom very well, so really don't have to worry about my coolant getting in there when it's running. I am thinking about putting a splatter shield on it so when I blow off chips between part changes, it doesn't easily get in there. Just speculation so far. I did get my mount made though.
I used some 3/8" steel plate and built a box.

I will have the info and pics soon on my mill build thread. (http://www.cnczone.com/forums/showthread.php?t=32609)

Lee, I would be worried about getting streams of coolant in your router without adequate splash-guarding - I find it goes everywhere when a coolant jet finds some odd hole on the part to deflect from. I would really like a fog buster system once I have a compressor - it looks like a much cleaner option!

I used to use WD40 on my X-1 before I had flood coolant - it works very well, but only because my X-1 was such a dog - with 20-40,000 rpm cutters and feed rates and rapids in the 100's of IPM, you really don't want any body parts anywhere near the moving parts of the machine.

Cheers Jason - that's a nice vid. What sort of DOC are you using?



Thanks Digits!

I used .25 mm DOC... It's not as slow as you might think, though would still be slower than you cutting 1.0 mm @ 1200 mm/min I guess. I think I was running at 2625 mm/min in that video. 4000 mm/min works fine too.

I should mention that the uncoated carbide bits I use are really sharp. I cut myself on them regularly... :(

Jason - Thanks for your kind words too! Why do you suggest coolant might not work with my setup? The spindle has an air supply for cooling and sealing. I was planning to shift that spindle to my next machine (pic below) soon, and that one was going to be enclosed for coolant... I was rather hoping it would give me an even better cut :)

Lee - Thanks for your kind words also. I've read your thread, like your mill. I used the HRW rails too, they're great. To answer your question, the mill is my first CNC effort. It leaves a lot to be desired and I learned a lot from it, but I'm not embarrassed by it - first try and all that :) I'll rebuild it once I have the next machine I'm working on now finished...

I think that coolant or not, when the speeds get to 20K & above the chips go everywhere anyway, they blow around and bounce off everything. They're easier to clean up without coolant though, and they taste better (don't ask :))... Plan for way protection from the start - trying to add it later really sucks.

Regards,

Jason

wow! Awesome machine. I only said that because I didn't know for sure what your set-up was. I didn't know if you were going too build an enclosure or what, or how your mill looked and was set-up!

Use coolant! It will be awesome :)

Jason, is the base of that a Granite surface plate?

wow! Awesome machine. I only said that because I didn't know for sure what your set-up was. I didn't know if you were going too build an enclosure or what, or how your mill looked and was set-up!

Use coolant! It will be awesome :)

Thanks :) I only really want the option to use coolant, I'll give it a try using the coolant from the manual mill & evaluate the benefits. If they don't stack up I won't go to the trouble of setting it up... That machine is intended for my brother, he will mostly cut wax, but would have the ability to cut pretty much anything.

Jason, is the base of that a Granite surface plate?

It's 2" 6061 aluminium plate. I considered a granite plate, but my design calls for a rebate where the Y axis ball screw and servo mount is, which would be difficult for me in granite with the machines I have. The aluminium is pretty solid anyway... :)

Regards,

Jason

Leotech - really sorry for the thread hijack, in retrospect my machine is not that relevant to your plans. I hope all this is useful to you somehow...

Best regards,

Jason

Thanks Digits!

I used .25 mm DOC... It's not as slow as you might think, though would still be slower than you cutting 1.0 mm @ 1200 mm/min I guess. I think I was running at 2625 mm/min in that video. 4000 mm/min works fine too.

I should mention that the uncoated carbide bits I use are really sharp. I cut myself on them regularly... :(


Sorry, I didn't mean your machine was too slow - and I didn't realise it was cutting quite that deep or fast! I just nearly lost the will to live watching my old X-1 cut at 0.25mm DOC at 600mm/min!

I don't think that you've hijacked this thread too badly either - I would have thought that this sort of high-speed, low force cutting is exactly what the OP's slide modules and gantry would be good for. With high speed though, you really do have to consider the magnitude of the motor forces - they'll probably be greater than the cutting force if you have the sort of acceleration needed to cut usefully at high feed rates.

I am also pleasantly surprised to learn just how well you can cut 6061 dry with appropriate cutters. I've just 'accidentally' increased my carbide collection thanks to Ebay ;)

Thanks everyone for all this good information. Since I don't have a manual milling machine to fab parts out, I have make baby steps.

At this time, I'm going to use my Sherline spindle with motor. That should work for a while until I can think of another way. Has anyone though of using a router motor, 1/4" shaft to coupling, shaft to a high speed spindle? Since the router bearings are not designed for loads.

Next topic, making parts for a milling machine using a hammer, hand drill, AJAX.

Thanks everyone for all this good information. Since I don't have a manual milling machine to fab parts out, I have make baby steps.

At this time, I'm going to use my Sherline spindle with motor. That should work for a while until I can think of another way. Has anyone though of using a router motor, 1/4" shaft to coupling, shaft to a high speed spindle? Since the router bearings are not designed for loads.

Next topic, making parts for a milling machine using a hammer, hand drill, AJAX.

Why put up with all that noise? There are much quieter ways to get the rpms.

What's your suggestions then?

You can get a 10,000 RPM pulley kit from Sherline.

http://www.sherlinedirect.com/index.cfm?fuseaction=product.display&Product_ID=164

Leotech - I'd be interested to hear how the Sherline spindle goes for you. I'm sure lots of others on the zone have them, but I don't seem to read about them much.

I have the Sherline ER16 one with no motor sitting on my desk - I haven't used it yet but I was planning to mount a brushless servo inline on top of it - then I could easily control the spindle through Mach3. It would give me the option to use up to 10mm cutters, and the lower speed would allow longer cutters without chatter. More flutes would increase the feed speed, and I'm hopeful that with a greater DOC it would be faster than the HF spindle for 2d work. My perception is that you would get a better surface finish and nicer cut with the sherline than with a router - perhaps someone with experience of both could comment?

Sorry, I didn't mean your machine was too slow - and I didn't realise it was cutting quite that deep or fast! I just nearly lost the will to live watching my old X-1 cut at 0.25mm DOC at 600mm/min!

I don't think that you've hijacked this thread too badly either - I would have thought that this sort of high-speed, low force cutting is exactly what the OP's slide modules and gantry would be good for. With high speed though, you really do have to consider the magnitude of the motor forces - they'll probably be greater than the cutting force if you have the sort of acceleration needed to cut usefully at high feed rates.

I am also pleasantly surprised to learn just how well you can cut 6061 dry with appropriate cutters. I've just 'accidentally' increased my carbide collection thanks to Ebay ;)

Don't worry, I didn't take it as a criticism - fact is it is slow for roughing compared to a heavy machine, but for finishing the HF spindle really comes into it's own. Your comment regarding the motor forces is true too - I snapped a 1/4" carbide cutter in the blink of an eye when I tried to take a deeper cut. The spindle stalled, and BANG. Scary. Never did find all the pieces.

Oh, I have that ebay problem too. Oops, I slipped and bid, won another auction... :) It'd be good to hear how the new cutters go for you.

Regards,

Jason

Has anyone though of using a router motor, 1/4" shaft to coupling, shaft to a high speed spindle? Since the router bearings are not designed for loads.


Do you have a lathe? There's an interesting discussion on a variation of this idea from page 5 of this thread: http://www.cnczone.com/forums/showthread.php?t=12184&page=5

There Herbertkabi is using brushless RC motors to drive high speed spindles. Much quieter than a router, and it seems the motor and controller wouldn't necessarily be much more expensive.

Regards,

Jason

Here's my X and Y axis. connected and powered by Gecko G320. I'm getting a pulse surge noise that I traced out to my parallel port. It's putting out 3.4v. Now I just need to build the Z axis and spindle.

http://www.leo-tech.com/images/IMG00081.jpg

Leo, as you can see, you are on the right track with high rpm, high feedrate 100IPM and low DOC, when working with AL.

We used a Hitachi router (M12VC) with a .25" 3FL non-coated carbide endmill and .01 DOC and 20K RPM, 11 hours nonstop.

Here's the results. The AL plate was 1" thick.

No detectable play in this router after double sided machining of both plates. 11Hours X4=44 Hours.

Looking good - you sure are handy with a Hammer, hand drill & Ajax :)

Is it upside down in the pic? Just wondered... Did you decide which way you're headed with the spindle?

Jason

I have 6 sherline spindles. I would need to get some collets. E-X? I plan on mounting a servo motor, with belt to the spindle. This should be ok for now.

I need a T-slot table to mount on top of the x-y axis. In building the Z axis, I was thinking of building a 8020 fixed gantry. I also need to find a AL slab to mount every thing to.

Are your spindles like these: http://www.sherlineipd.com/spindles.htm ?

They have either a morse taper or an ER16 nose, so if you have the ER series one, you want ER16 collets. You'll probably need a collet nut as well - I don't think they come with one?

This place is good for good quality collets: www.maritool.com

Regards,

Jason

Yes, spindles are just like those. 1 of them. the others are from another manufacture. I need to clean them up. Some have some play in the bearings.

Thanks for all the links. It helps quite a bit. :)

I have one of those IKO slides on my Z axis. Heavy mothers aren't they? :)
I eventually found mine had a little slop in it. Not that I could move by hand, but when the Z had pressure on it from traveling down and touching the material. I never took it apart after I bought it on Ebay.
Once I took it apart and cleaned it all real well, I learned how it works. One side acts like gibs on the ball bearings. It adjusts the pressure on the balls. I was able to remove the play by tightening the set screws sequentially like you would on a standard dovetail way.

I would consider using solid sheet way covers on these. They are designed so that this would be very easy to incorporate at least on your X axis.

I was thinking overnight in my sleep. the IKO slides will be the Y and Z axis. I have a large X slide table that I can use. It has a stepper motor that I need to change to a servo.

How did you hold up your IKO? I need enough material to hold up two IKO slides. 1 for Y and 1 for Z, plus the Spindle/servo. That section, I estimate will be around 100lbs.

That is a great idea! That's two solid axis right off the bat. What is the travel on those? Probably a bit short for X, but great for Y and Z.

If I remember correctly Lee used an 8020 extrusion (3" by 6") supported by an angle plate from the back, triangular plates on the sides and a 3/4" or 1" plate bolted to the front of the 8020 that was holding the IKO slide as his column.

I'm looking forward to how this will turn out!

Serge

Serge is right about what my column is made of. Here is the build log for mine.
http://www.cnczone.com/forums/showthread.php?t=32609

My IKO slide has 9" travel. These two appear to be identical.
My Y axis travel now with my new way covers in place is just over 9". I get 14" now on my X. I modified the table a little better the other day to get the extra it should have originally had.

I'm looking for a T-slot table top to mount on my slide tables. All the ones i've seen are X-Y axis prebulit tables. I just need the top. I'll drill and mount it to my slides.

List of things to purchase
1. Angle brackets
2. 80/20
3. T-slot table top
4. low profile vice
5. collet/draw bar
6. spindle (tried to reuse older sherline spindles however they are worn out)

Finally got back to this project. I received my 80/20, got the 3060, 3" x 6" x 24" for the Z axis frame. Angle brackets came in. bolted mostly everything together.

For the spindle, I took my Sherline 4000 lathe, took off the spindle and mounted it to the Z axis. This hopefully should be good enough to mill out some parts manually by hand. Those parts will go into the motor mounts.

Question I have is the Z axis weight is about 45 lbs. What's recommended to counterbalance this weight? Gas shocks? springs? pulley system wrap over the top and down the back side?

A shock would be simplest. Two bolts and your done.

I'd go with the gas shock too, they look nice and are super easy! Try to either get one without the damping at the end of the travel, or configure it so you don't need to use the last inch or 2 travel - the damping may slow your machine down. As I recall, you're using servos so you shouldn't have any trouble with position errors due to the damping, but better safe than sorry.

A spring could also work, but they're harder to implement, and harder to keep it looking tidy.

I would avoid a counterweight system - you would double your inertia for the Z axis, and although this probably won't be an issue while you're using the Sherline lathe headstock for your spindle, you may want to change to a high speed spindle in the future and the higher feedrates may give you some grief then. 3D contouring and parallel finishing need a good Z.

Best regards,

Jason

Thanks Jason.

What spindle would you recommend? VFD? Direct Servo drive?

Thanks Jason.

What spindle would you recommend? VFD? Direct Servo drive?

I'm sure I've expended more mental resources on that question than any other single aspect of these machines!

I would start with the lathe headstock, especially since you already have it. Actual metal removal rate is likely to be as good as a high speed spindle taking lighter cuts at higher feedrates. Have you read this: http://www.5bears.com/cnc.htm ? Particularly this page: http://www.5bears.com/cnc27.htm - he uses basically the same spindle you have, and gets an excellent result.

Best,

Jason

For this size of machine, it really is a tricky problem.

5 Bears ponders this issue here:
http://www.5bears.com/news.htm

I think his approach of tuning the X2 spindle with better bearings and a belt drive kit may be the best long term solution. Reasonable strength and decent precision with new bearings installed (some get lucky with the stock bearings). Not too heavy and easily installed on a flat mounting plate (no precision bored clamp required). Quality, reasonably priced quick change tooling avaialble from Tormach.

Using the Sherline spindle to get things rolling is a great idea. It will however be the limiting factor in the mill considering the strength of the IKO components and servos.

Here's a quick picture of what my mill looks like now.

http://www.leo-tech.com/images/IMG00098.jpg
http://www.leo-tech.com/images/IMG00099.jpg
http://www.leo-tech.com/images/IMG00100.jpg
http://www.leo-tech.com/images/IMG00101.jpg

Looking pretty good. I think your spindle should work okay within it's limitation's. You can always upgrade later if you find it less than what you wanted.

Have you considered flipping the X and Y over? This is how I would mount those. Mount the Y carriage in the center and then use UHMW or even bronze to support the rest of the rails as they traverse. Could be done with some roller bearings too. It will take a little more work, but I think it would be more solid overall. The way you have it stacked would seem to limit the table size quite a bit.

It won't completely cover the screws if you clean up with compressed air, but would not act as a trough for the chips this way.

since the X & Y are both the same size, I get about 10" x 10" of cross slide movement. I mounted the spindle on the Z Axis right above the center point. I'm not too worried about having it too small. Most everything is bolt together so in the future, I can swap parts.

I plan on having large rubber sheets to cover up the exposed lead screws.

Nice! :) I'm personally using a bunch of the Thomson Superslide's you've got for the Z axis. I've got some 9" ones for lifting the Z gantry, a 15" one for the Y axis, and an 18" one for the X axis. (mounted to the gantry as well, with a Taig spindle / DC motor mounted on it's plate, similar to your Sherline) Do you have any noticeable deflection with the 80/20 3060 chunk on the Z?? I ran the deflection calculator and it says less than .001" deflection, but still unsure if I should go for it. I haven't held a chunk of 3060 in my hands, it's been a little too much $$ so far.. (I'd need about 9' of it for my plans!) Would be cheaper to go with welded steel, but I like the adjustability of the 80/20 stuff.. Love your servos too! :) Also in my plans are sheets of rubber to cover the Thomson's slides and ballscrews, but was going to try something like a window-shade on each side of the table, where the rubber will roll up (and be wiped off by a wiper) so that it won't be in the way.. Of course, it's all in the head right now. Still need to buy the 8020 and make the parts on the Taig. :D

24" of 3060, bolted to a an angle iron block, bolted to a 3/4" base, it's not moving anywhere. No deflection and i have it hanging off of the 3060 6" like a "T", i calculated about 75lbs of weight. 3060 cost a bit however if you're using steel, it's going to weight quite a bit more.

My Z axis is made by Parker. It has about 9" of travel, pretty beefy. got it real cheap at a surplus shop. Servo motors came included with the IKO slides. another ebay special. They weight about 50lbs each!

I need to cut some more metal out

Dang, could swear that was a Thomson, lol. Looks identical from this angle. Probably pretty close in construction from the looks of them, I like 'em! I can imagine those IKO's are heavy, I made the mistake of thinking I could lift 4 of the Thomson slides at a time, and those are aluminum! Do you know about what you'll get out of your setup for rapids? Can't wait to see that thing running...

My IKO is on the Z. It's a pretty good slide, but is better suited to X and Y.
The design would simply work much better.
I will be adding some standard THK 25 rails to my Z. This will do away with any defelction I have seen. I have tightened up the IKO and it now shows no deflection at all. Very solid. Long term though, I would either have to replace the balls, which I have or replace with something more durable. It won't be soon as it is working in a production environment, sort of. This machine runs several hours almost every day.
These will last even longer on X and Y as I think that is what they were designed for.
It's only a single row of ball on each side. They are however, cast iron and steel, so they will last.
The screws are pretty accurate as well.

Those ball-screws look a bit on the fast side.

Phil:)

Here's a quick picture of what my mill looks like now.

http://www.leo-tech.com/images/IMG00100.jpg

they are very fast. I might need to build a 2:1 reduction to get the resolution. I got all the servo motors installed and wired up last night. I'm able to jog all axis. I still need to order some gas shocks for the Z axis. Also still waiting for some blocks of AL for the Z axis. It's all held together with C clamps now.

Since I'm using the Sherline 4000 Lathe spindle head,
I believe this is the correct collet set?
http://www.sherline.com/3060inst.htm

What pitch are the ballscrews? I can run some numbers for you if you like.

Give me the details - pitch, dia., length of ballscrew, coupling type, moving axis weight including the part you're cutting, expected max. cutting force, then I can give you a good idea of the torque you will require. Also if you give me the servo details I should be able to tell you whether you might have problems with inertia.

Best,

Jason