Electronic leadscrew

[tweinke]

Any body here tried this out? I have been reading the Yahoo ELS group info and looked at the Automation Artisans web site. This sure seems like an interesting project for someone who does not want or need CNC. Comments and pinions from any one welcome. Does any body have a working example in there shop? Are there any other ELS kits available

 

[Wreck™Wreck]

The X axis or Z axis lead screw?

If the the Z you can't thread this way without slaving the spindle to the lead screw.
Spindle becomes the C axis

 

[wrmiller]

It (spindle) is electronically slaved to the leadscrew, and you can thread with it.

I looked at doing this some time ago for my smaller lathe and had several conversations with the designer, but didn't have the money at the time and eventually moved on to other things.

 

[bpratl]

I converted my G0602 lathe to ELS with the stock lead screw, last spring, but I found that it to have too much friction and back lash, so I install a ballscrew/nut which made it much better. After that I added a X axis ballscrew and stepper which allowed me to do metric/SAE threading and very precision tapers. Since then I added mach3 and Linuxcnc software for full CNC which allows me to have the best of both worlds. There was quite a learning curve with heaps of fun. I also added a wireless Xbox controller for manual operation which is a little touchy. Tonight I machined 4 large bronze bushings which came out great.

 

[Wreck™Wreck]

I would be very interested to learn how the spindle drive is controlled.

Can it index as well?

 

[wrmiller]

The spindle rpm isn't controlled, it's monitored. Threading at a specific pitch is nothing more than running the leadscrew at a specific rpm relative to the spindle rpm to derive the proper ratio. I have been through the firmware for the ELS and I found it acceptable to do its intended job. (I wrote motion/data control firmware for over 20 years)

To answer your question about indexing: No it can't because the spindle isn't under it's control. What would be really cool here would be to drive the spindle, X, and Z screws with servomotors, but then you are basically converting your machine to CNC at that point. The purpose of the ELS is simply to allow more flexible feed/threading capability that that found on typical hobby lathes.

 

[Wreck™Wreck]

The spindle rpm isn't controlled, it's monitored. Threading at a specific pitch is nothing more than running the leadscrew at a specific rpm relative to the spindle rpm to derive the proper ratio. I have been through the firmware for the ELS and I found it acceptable to do its intended job. (I wrote motion/data control firmware for over 20 years)

Thanks, controlling the speed is easy, how does it control the rotational position and maintain a relationship with the lead screw, I understand maintaining a speed relationship between the drives, where is the spindle position derived from?

I sometimes run old Romi/Bridgeport lathes that will thread but the spindle will not index, they do have dynamic braking however.

The spindles have rotary encoders of course, the induction motor control must be working hard to maintain position during a 20 pass threading operation.

 

[wrmiller]

Thanks, controlling the speed is easy, how does it control the rotational position and maintain a relationship with the lead screw, I understand maintaining a speed relationship between the drives, where is the spindle position derived from?

I sometimes run old Romi/Bridgeport lathes that will thread but the spindle will not index, they do have dynamic braking however.

The spindles have rotary encoders of course, the induction motor control must be working hard to maintain position during a 20 pass threading operation.

It's not a closed loop system, so inaccuracies are inferred of course, but you have to fab a encoder (optical, hall-effect, etc.) for the spindle and feed this to the ELS. The more accurate/higher frequency response your spindle rpm monitoring is, the more accurate the threading. But as you point out, there is no rotational relationship between the spindle and leadscrew like there is in a gear driven system.

Accuracy with this system can be close, but it won't be perfect. Couple the cost of this system with the additional cost of stepper, driver, and power supply and you can see where this system can get pricy real quick. To make this type of electronic threading drive as accurate as possible, I would use a true rotary encoder on both the spindle and leadscrew and write my own code to synchronize both to the desired ratio based on the real-time positional data of the spindle, not its rpm. But that's too much like my day job... )

Edit: Sorry, forgot about your positional question. ELS has to be programmed with the number of pulses per revolution you have on the spindle. This can be used internally to create a once per revolution 'index pulse' or event to be used in place of true positional input from the spindle. As I said, inaccuracies will happen. If you had servos on both the spindle and leadscrew, you could use the positional feedback to logically tie two points in their respective rotational axis' to better simulate a mechanical drive train.

 

[tweinke]

I already have two steppers, belts, pulleys, drivers and power supply that were removed from my machine. So I ave a lot of the expensive stuff. I am not a programmer but can follow directions. Is this for me if I have time and patience? Threading and a possible taper are why I am interested .

 

[wrmiller]

You don't need to be a programmer. The firmware is done.

You just need to buy the kit, solder it up, and add the external pieces e.g., motor/driver/power supply. There are quite a number of people who are using this system and appear to be happy with it.