Just had to share! I ran my first test movements with a 9" Thomson Superslide linear rail, it's using ball bushings and a .2"/turn ball screw. Running a Xylotex 269oz motor with an Intelligent Motion Systems IM483 driver, right now hooked up to a cheap 24v power supply, with a hacked printer cable and some alligator clip wires.. Mach3 running on my IBM desktop PC, and I've consistently moved the thing at 270 inches / minute! Dang these slides and stepper controllers are nice! Same motor on my Taig only gets me 50ipm max, and that's with a bunch of backlash. So far this thing has less than .0005" backlash (no load anyway), which is all my cheap digital caliper will go down to. I have it set to 10,000 steps / rev, so I can theoretically move .0001" per step, if I can get a decent enough caliper to measure that. :) Good stuff! Now I just need to actually build the machine..
While I'm at it, anyone have a longer SuperSlide (or similar?) linear stage for sale / trade?? The current plan is to use one 9" one to move the table, and another to move the Z axis (on a gantry) in the opposite direction to get 18" of movement, but if I can get it in one stage it'll be easier and more stable.. I have a really nice set ( two 4" linear + one rotary) of Newport servo / encoder controlled stages I could sell / trade.. I think the linear stages are marked .0001" and the rotary is marked .001 degrees, I think they're pretty accurate. :) (UTM100 and URM100) I also have an MM3000 driver box that's supposed to make them move, but not sure how to interface it with a PC. Looks like it's either serial or GPIB..

Do you have a pic of the superslide?

I am curious how it compares to the NSK slides I am planning to use. I like the .2" pitch you have better than the 10mm pitch on my slides, but then again, I can get 1000IPM :-)

After playing with the tuning, dropping to .0005" / microstep, playing with the kernel speed and acceleration, I got it up to 400ipm! Yeehaa.. It'll stall out if a fly lands on the carriage, but wow.. Back to .0001" / microstep, I can get a solid 200ipm out of it, and I can't stop the carriage with my hand, without fear of getting it crushed. I need to find a scale I can slip in there to measure the driving force. As it is, it'll completly crush a 20oz plastic bottle cap @ 15ipm. Oh - you run out of room on a 9" slide really fast at 400ipm. LOL

LOL 1000ipm could be fun. :) Are the NSK slides a complete "setup" like the Thomson ones? (rails / ballscrew / bearings are all factory assembled into a unit) I like the typical NSK rails with their side profile.. I'd be afraid to set up a round rail system like this myself, but where they're already pre-built it makes it easier. Edit: Here's a small picture, without a stepper mounted: (from the danahermotion website) http://www.danahermotion.com/website/com/eng/img/product/LMS_BSDBG_2DB.jpg

Yes, they are a complete setup, very similar to yours I think (ballscrew between 2 linear rails with 4 carriages). They have a lid and two belts that keep dust debris out of the innards. I have had them for awhile, but only recently found a suitable drive for their 300 watt 4 pole brushless motors (4 pole brushless motors run horrible on trapezoidal amps). Just a few minutes ago I was doing some load testing on them, moving a car battery around at 800IPM (the only heavy thing I had nearby :)

I am planning my router/mill similar to yours I think. One slide will move the table and the other slide will be mounted to a fixed ganty (with z mounted to it). For general wood/plastic routing, it should be fine to cut anywhere on the 20" x 20" 1/2" thick mic6 cast aluminum plate I plan to use for the table. When I want to cut metal, I will just stick to the center 8"x8" section of the table which is fully supported by the slide's 4 bearings underneath. Best of both worlds! :-)

Exactly! It seems like a good idea on "paper" and in my head... At least I'm not the only one thinking that way! Only thing I'm a little concerned about is rigidity of the "gantry" setup. I was planning on using 2 uprights made of 3060 80/20 (3" x 6" columns!), but after reading some of the 80/20 posts I'm not so sure.. The deflection calculator shows something like .001" deflection @ 24" on a single 3060 stick, (maybe more, I forget..) so 2 together should be more than enough. I get way more movement than that on the Taig, but it doesn't matter since there's so much backlash in the table anyway. With the .0005" or less backlash on these things, I want to get the Z as stable as possible.. I considered welded steel for the frame, but worry about not having any adjustment to make up for warping. At least the 80/20 is easy enough to adjust, and can always be TIG tack-welded here and there once it's all set. Those NSK slides sound insane! I forgot to try standing on the slide and moving myself; I had done that earlier with it connected to my Xylotex board but could only get 100ipm or so out of it. The new IMS drivers are quite a bit more powerful (for steppers only though!), but as a result they run quite a bit hotter than the Xylotex. I have 6 of them stacked in my enclosure right now, but re-thinking that after running 1 for a couple hours. I like to keep the enclosure as small as possible, but these things really need to breathe. I'm probably going to use a mini-ATX case instead.

I was also thinking of using 8020 for the fixed gantry, but only using 1530 (1.5" x 3"). Since it is a fixed gantry, I was planning on having 2 struts also made of 1530 at a 45 degree angle to the gantry on each side. I think you could drive a truck over something like that...

I had drawn up my plans with some 45* 80/20 struts attached as well, but sometimes I'll cut a hole in something that's much bigger than my mill, like the top of a PC case or something.. I was worried that the side "struts" would get in the way of cutting something long like that.. I have on occasion used the Taig to cut off managable sections of a 2' x 4' sheet of plastic, it's nice not having anything in the way. Although they can only be 5"-6" deep with the Taig.. I was thinking of maybe making the strut things removable, if I ever did need to cut longer things.. Would definitely add some strength. First I need to get past the cost of the 3060 material! Even on their eBay store it's a few $$$... Cheaper to buy and weld steel. Ugh. :)

I guess I am not seeing how the struts would really be in the way of anything? Both ends of the fixed gantry mill would still be open, just the sides woudl have the struts. Since the table has to move back and forth in order to be able to cut something, even if you didn't have struts you woudl still not be able to put on an item wider than the distance between your two fixed gantry uprights...i think???

Think about a moving gantry...it pretty much has to have the sides clear of obstructions, otherwise the gantry would hit it as it travels.

Basically, my table would be 9" x 18", and would move around inside the envelope of the frame without hitting anything. So I could cut something that was 9" x 18" (or actually more than 9" front to back, since front-back would be open). But with the 45* angle things on the sides, if I wanted to cut something that was 36" long, it wouldn't fit between the uprights on the sides. My "fixed gantry" would be mounted at the *back* of the base, so something that was longer than the table can go *almost* all the way to the back of the mill. I wouldn't be able to cut a hole in the *middle* of that really huge piece, but if it were just milling holes within 6" of the edge of it, or in a long bar or something I could do it.. In all reality I'd probably be fine with the angled supports there, as it is I rarely cut anything that big.

For a quick update... Using the same IM483 stepper drivers, I switched from 24v to 48v supply and WOW did it make a difference.. Before I could get max. 400ipm rapids, but without much torque. Grabbing the slide would stall the motor. With the 48v supply and all else the same, I can stand on the slide and move myself (~180lbs) back and forth at 400ipm! My fastest rapids have been around 600ipm, but again with not much torque. End-to-end on a 9" stage @ 400ipm comes quick enough! I think I'll probably keep it around 300ipm or so, still insanely fast compared to the Taig. Oh - also tried the same controller on my Taig instead of the Xylotex board; with the Xylotex I can go max. of 16-18ipm on my Y axis before stalling when I grab the table, with the IM483 drivers @ 24v I'm around 30-40ipm and I can't stop it by hand. I like the Xylotex board for its simplicity and relatively cheap cost, but wow, nothing like using some professional drivers. I also lucked out at work and found a 2 axis driver box, opened it up to find 2 commercial stepper drivers mounted inside, each with its own honkin' 53v power supply. Can't wait to try those!! Still changing the design of my machine every other day.. Do I use steel, 80/20, make the whole Z axis stage go up and down like a real CNC machine (more clearance underneath) or just the headstock (lighter, but deeper parts will hit the stationary stage parts like the Z column on a mill..). Ugh.

Since we last had the discussion about 8020 and rigidity, and about using struts to brace the gantry, I bolted my unit together. I am very pleased with the 8020, and I only used 1530 (1.5" x 3") instead of the much stouter 3060 that you have.

I don't understand what you mean about the z axis. I took the common? aproach and made a plate that slides up and down, with the headstock or router/whatever attached to that. How do you plan to do your z axis? Do you mean the whole thing would move, including the z axis motor?

I have some pics of my current build here if you want to see how I did my z axis using a slide:

http://www.cnczone.com/forums/showthread.php?t=57673

Exactly, basically instead of attaching the spindle / motor / etc to the moving plate and having the stage fixed, I would bolt the stage's moving plate to the gantry / top X axis stage, and attach the spindle / etc to the back of the stage. Stage and all would move up and down, so I wouldn't be limited by the distance from the cutter to the stage, or "Z column" as it were.. Down-side is that I'm lifting and lowering probably 15 pounds of stage, plus the motor / spindle setup, all cantilevered out a few inches. The stages are rated at 650lbs, so hopefully they're solid enough? Plus I'll have gas struts to counteract the weight... I dunno. Probably not even worth doing, but I wanna see if I can.

I just checked out your build, impressive!!! Those stages look great. And you're right, it's all about the rapids even if you can't actually cut at 600ipm. :) I just want to see the thing whiz back and forth between cuts. Should go a long way to speeding jobs up. I've since scaled back my drawings to using 2020 and 1030 parts instead of the 3060, but if I stick the Z up there like that I may need the 3060 after all. Unfortunately everything I have (the stages, the Taig headstock...) is pre-drilled with 1" apart mounting holes, so it doesn't go too well with the 1.5" 80/20 stuff.. More stuff to think about. :)

I should probably just go with something like you have it, with the spindle hanging down on a plate from the stage.. There's a good 1.5" from the edge of my moving plate to the edge of the stage though, so it'd be hanging down 3-4" to be useful. I had drawn it up like that in one revision actually. Would probably be fine. Just over-analyzing everything as usual! Oh - (if you don't mind..?) How much / where did you get the mic6 plate? Was considering it for some of the parts but hadn't actually seen it anywhere.. For a table, I was thinking .25" thick steel plate bolted to the X stage plate, with a bunch of 1030 extrusions on top of that to form a T-slot table, just deck it flat with a flycutter or something..

Yes, I was wishing the t-slots would align up with some of the NSK robot slide mounting holes too, but they were spaced in metric, and not even spacing that would work with the metric 8020 line. I ended up just making metal plates out of 1/2" aluminum plate to attach the slides to the extrusion.

That is why I didn't do the Z as you discussed. I didn't want the smaller Z motor to have to move the weight of the whole slide + spindle + motor. Besides, if it can fit under the fixed gantry it can fit under my z axis assembly, so I don't see an issue...

I bought the Mic6 on ebay. I started to search cnczone to figure out where people were getting their plate, then decided to check ebay. There are several metal suppliers who seem to list precision ground cast aluminum plate in various sizes, and I emailed one of them asking about a 20" x 20" x 1/2" piece. He got back to me and said he could provide it for $120, and if I wanted it, he would create the auction (which I did and he did). Shipped out the next day, very prompt. The ebay ID is ssshapiro

Well, several months later, back with an update! I (naturally) haven't finished the machine, actually haven't quite started it yet. All sorts of stuff came up as it always does... Every time I design it, I find something else (on eBay!) that will work better. Now I'm up to 18"X by 15"Y by 9"Z axis, with a moving Y plate and an X-Z gantry made of 80-20 3060 extrusions. The 3060 is like $140 a stick (used!), but it should be rock solid.
Anyway... Just wanted to say that the Thomson 2DB12JUBQL24 slide I got for the Y axis is working incredibly well! (15" travel after you subtract the 3" for the ballscrew bearing mounts and 6" for the moving plate, 2 turns per inch ballscrew with preloaded ball nut) I've got Mach2 maxed out for velocity, 4 microsteps / step (1600steps/unit), and I'm getting 1311 inches per minute! I had to crank the acceleration down a bit to get the 1311, but it's doing it! That's 21.85 inches per second! (555mm/sec or 33,000mm/min if my calcs. are correct..) Unfortunately you run out of the 15" of real estate very quickly.. Wish I had a 60" table that could do that! Realistically I'll probably set it at 200 or 300 inches / minute for rapids, but damn this thing is fast! What's a typical industrial CNC machine run for max rapids? Considering the servo motors each weigh as much as my entire mill...
I noticed some play in the ballnut so I took the slide apart and figured out how to adjust the spring washers, but I may have put a bit too much preload on them. Need to check that, it may actually move a bit faster if I get that dialed in just right. According to Thomson, .001" of movement between the 2 ballnuts = 500-1000 pounds of preload..

Ah crap, I also just found the "Max" rating in the Thomson document -- 11.7 inches/sec.. Whoops! Better slow it down before I break something. Maybe it has to do with slinging the lubrication off the ballscrew if you're whipping it that fast. Then again, the same ratings apply to the 10turns/inch and 2turns/inch slides, which doesn't make sense.. Maybe it's the rating of the linear bearings and not the ballnut they're concerned with..


Ok, last experiments of the night... Set the driver (IM483) to 25microsteps / step, so 1 microstep = .0001". I didn't think the motor or the driver could do it... But they do! I can step .0001" at a time, indicated on a tenths dial. 5 steps move exactly .0005". Backlash is currently .0001"-.0003" depending on conditions. With Mach2 maxed out (Kernel speed 35,000HZ on my laptop) I get 209 inch/min rapids. The strange thing is, since the motor is physically turning 10 times slower than on my Taig (20turns/inch) and the slide is so much longer (15" vs 5") 200" per minute doesn't look fast at all! It looks like I'm doing 10" per minute... Crazy. I'm going to crank the Kernel speed to 45,000 Hz, but I think the last time I did that it freaked out the 1.8Ghz laptop. Need to get the 3Ghz machine set up.. Here goes.....

(dramatic pause while I change some parameters and hope I don't crush my tenths dial by accident!)

SWEET!!!! The computer doesn't like it (slow responding on the MDI screen), but it's doing 269 inches per minute on the slide, and I can't stop it with my hands! I'm manually entering "G0 Y10", "G0 Y0", "G0 Y10", etc.. and it's getting there almost before I can finish typing it! Great, now I'm not going to be able to sleep tonight!!! :)

No really, I'm going to bed soon!! LOL Realizing that .0001" resolution is probably WAY overkill for the machine (my Taig ER16 spindle probably has more runout than that anyway!), I've set it to 5 microsteps / step, so .0005" / step which is more than good enough. That gets me a solid 400 inches / minute, and I have to struggle to stop it with my hands. I think that's going to be good enough! Not bad for a 269oz stepper motor... Hopefully the X and Z axis will behave the same, and I'll be cutting pockets at 400 inches per minute! Albiet with a .005" depth of cut, but damn if it works I'll figure out how to take and post a video of it! The X and Z are 5 turns per inch ballscrews I believe so I'll need to adjust the steps/inch to match, but those are the slides I was moving at 600"/min before so I don't think I'll have too much of a problem. I'll be using 2 of the 9" slides to lift the gantry, and probably some gas struts to help out... This thing's going to be a beast!