[BobWarfield]

I don't know if you've noticed, but Aaron has pix of a substantially improved mill up on the site. Some of the changes:

- Longer Z-Slide to reduce cantilevering

- Ground ways = no more lapping

- Big Z-Axis collar

Not sure what all else, but it looks like a nice package of improvements to create a new generation machine. Price is listed as $2000 for this mill. Still a lot of bang for the buck.

Best,

BW

[Ron111]

I guess no more debates about lapping the ways, of well. HeHe!! I guess I should hold my tongue!!!!

Your right, I looked at the web page a couple of days back and have been wondering how to justify snatching one up!!

Bob, I need a little education about "- Longer Z-Slide to reduce cantilevering"
what does this mean?

Now, I thought that this mill was already pretty rigid, so I guess that it's very rigid now.

Some of the RF-45 guys saying that their machine came with 2 speed electric motors from the factory and I guess they top out around 3000 rpm. May be a good mod just to replace the motor. But, I guess it would make more sence to go with VFD motor. What are you planning to do?


Ron

[BobWarfield]

The cantilevering issue is due to having a 240-285lb head that cantilevers out from the Z-axis slide. By increasing the length of the slide, the goal is to better accomodate that force. It's quite straightforward to reduce it using counterweights with either lead (needs to be a pretty big weight!) or gas springs. In any event, it will run more smoothly, require less servo/stepper force, and presumably sit more true on those slideways with this new design. I also like the way the head tram nuts are easier to get at on that collar rather than being buried.

It is tempting to look into getting my ways ground instead of lapping. I'll have to think about it.

On the motor front, there are several things to consider. First, you can up the motor's rpm. Stock is a 1700 rpm motor and you can upgrade it to an American made motor running at 3425 rpm with a VFD. That is the path I'm on, though I have not made the swap yet. Given the right gear oil in the box and a 3425 rpm motor, Aaron says you'll be able to run 3000 rpm, which is nice when using smaller cutters, carbide, and when cutting aluminum.

The second possibility is a belt drive conversion. Aaron is supposed to be working on one. This is tempting and would allow much higher speeds, say 6000 rpm. Given the slop everyone has found in the quill, and the fact that most CNC'ers use the column for Z-Axis instead of quill, it is tempting to try building a belt drive head for the machine. I would think it would be straightforward if you are eliminating the quill, though by no means a small project. An ambitious person could make that new head in aluminum, with no quill, and a longer z slide area and have a very nice upgrade.

Last thing on spindles is that 6000 rpm is still not enough for really getting jiggy with aluminum and small cutters. For this purpose you've got folks lashing things like the Proxxon spindles on that run at 20K - 30K rpm. You'll need to use them with 1/8" or less cutters, but they should do quite a number on aluminum. I'll probably set up one of these that I can clamp to my spindle when I want real high speeds.

Ron, these look like nice machines. I still think one needs to consider whether all the trouble of converting an IH Mill to CNC is worth it, or whether you're better off to just buy a Tormach. I suspect it is a function of how much a hobbyist you are, but with these new improvements, the IH is looking less and less a poor cousin to me.

If I can ever finish my lathe, I'll look forward to jumping into the mill conversion. I'm putting together the PC for the lathe today in between breaks to jump on the 'puter, tend to the kids, eat meals and what not. It's a nice little Athlon 64 system on a micro-ATX mobo that fits in a neat little 19" rack mount chassis I bought off eBay. Just about ready to do a smoke test on it. After that, it'll be time to build the stepper power supply and wire up the Gecko drives, at which point I can start spinning motors. Still vacilating between making it all work with a G-Rex or parallel breakout board. Pity the G-Rex is not coming along faster. They may just push me to the parallel route!

Best,

BW

[Ron111]

I guess the g-rex would be nice, My understanding it frees us some of the computer processing and uses the usb port currently. In the future, it will allow the steppers to be used in a closed loop setting using encoders, which will eliminate possible missed steps.

While I was looking at the IH site, I noticed that Aaron is placing the encoders on the ballscrews instead of on the motors. Have they always been done this way? or is this a change. I realize that this would eliminate gear and belt backlash from entering into the picture. As far as your lathe, you could also make provisions for encoders on the bearing blocks/ball screw mounts for retrofit when this feature becomes available (if it's not already).

Will, this G-Rex upgrade be a software upgrade or firmware?

Different subject, have you ever seen download plans (PDF for around $15) for constructing preloaded ball nuts from single ball nuts. I think I remember seeing some on a conversion plans website, but now can't find it.

Thanks,
Ron

[Randall]

I dont know if its been changed but AAron puts encoders on servo and on
ball screw. I was wrong just on the ballscrew.
Randy

[BobWarfield]

No encoder on the servo that I can see. Looks to me from his installation instructions on the site that the encoder is on the ballscrew directly. Some people have said this can lead to dithering, but I would think with a timing belt drive if it is properly engineered, it would be fine.

Ron, my understanding on GRex is you will need a firmware upgrade to get the steppers to run as servos. I believe this will be part of the upgrade currently being worked on. The same upgrade will be needed to do threading on a lathe, so at the moment, the GRex is not a very good choice for lathes (no threading!).

Once all the features are present, it should be a nice device. It is definitely taking a LOT longer than expected, however.

I think the biggest advantage of the GRex is its potential to radically simplify things, especially if you want a control panel. To make my relatively simple lathe panel work without GRex requires a combo breakout card, a ModIO card, and an iPac keyboard emulator. That's a pretty ridiculous amount of interface electronics to hook up a few switches, and it traces back to the severe limitations of a parallel port. From this perspective, the GRex is in many ways too clever by half. It has a very sophisticated motion control capability that seems to require a lot of firmware work to work well with Mach 3. A simpler design that wasn't trying to be a Galil-style unit would've been working a lot sooner. Oh well.

Best,

BW