electronic lead screw

Thanks again for your reply, the math, - for some reason i didn't get email notification so missed your response, till now. (was waiting with bated breath!)
so to clarify, for me, & those who have just dropped in on this thread, are you saying that 590 lbf or 2,624N is the lateral force applied to the carriage via the 1/2nuts, with your 3:1 stepper arrangement, driving the leadscrew? (and not via the gearbox).
And is that enough to do more gentle, but not 'toy-ish', depth of cuts for an 8 TPI thread?
I measured 10Nm static torque from my 5:1 setup, and the equation estimates that should provide 590 pounds of force on the end of the screw (or on the half nuts). There is further friction loss in the half nuts, and more in the sliding friction of the carriage.
I'm now the owner of a 1000lb dial force gage, and will be trying to rig something up to measure the force on the carriage when it stalls. I was busy making a precision slitting saw arbor yesterday, and I have another project to finish today, though.
I've also mounted a 120-tooth timing pulley and am running an 8:1 ratio right now. I should shortly have a 12-tooth pulley for the motor and can go to 10:1. The 120T pulley just fits under the cover, and overlaps the encoder pulley (but clears). Pulley/belt availability has improved markedly since I began this project. There must be a lot of people doing this kind of thing.
 
yeah, hey sorry re that drive ratio error, been jumping across too many ELS stories!;)
am not in favour of hi pulley ratios tho as lead screw speed will drop off, without using a servo motor, so gentle cuts on long-ish threads may become tediously time consuming, yes?
BTW, I bought a reasonable taiwanese MT3 slitting arbour, cos time vs $$ didnt add up, but well done bro.
Go easy on ur machine! 1000lb on 1/2 nuts for 3/4" dia screw is nasty
 
Thanks again for your reply, the math, - for some reason i didn't get email notification so missed your response, till now. (was waiting with bated breath!)
so to clarify, for me, & those who have just dropped in on this thread, are you saying that 590 lbf or 2,624N is the lateral force applied to the carriage via the 1/2nuts, with your 3:1 stepper arrangement, driving the leadscrew? (and not via the gearbox).
And is that enough to do more gentle, but not 'toy-ish', depth of cuts for an 8 TPI thread?
I forgot to mention that the static torque is scaling with the change in ratio.
Also, I have the microstepper set to 800 steps per revolution, which is as low as it will go. I don't know if it's also the highest-torque setting, but it seems like it ought to be. I doubt that I would be able to measure the difference between 800 and 1000 (where I had it with a 4:1 ratio) though. I guess I should try it and see.
 
yeah, hey sorry re that drive ratio error, been jumping across too many ELS stories!;)
am not in favour of hi pulley ratios tho as lead screw speed will drop off, without using a servo motor, so gentle cuts on long-ish threads may become tediously time consuming, yes?
BTW, I bought a reasonable taiwanese MT3 slitting arbour, cos time vs $$ didnt add up, but well done bro.
Go easy on ur machine! 1000lb on 1/2 nuts for 3/4" dia screw is nasty
I haven't recalculated the lookup tables for the ratio change yet. The step rate will be scaled, and the microstepper is rated to handle up to 200khz, so I think I have enough margin in the Arduino.
I made an expanding mandrel for modifying timing pulleys the other day and my cheap arbor and saw were really bad, so I decided making a good arbor would be good training. I hit all my dimensions to a couple of tenths, so I must be learning something.
These machines can rip their own guts out in low gears, as I have some unfortunate experience with. The stepper motor's torque limit is something of a feature on a machine with no clutch, and I think I'm still far short of the forces change gears can apply (famous last words?).
 
I haven't recalculated the lookup tables for the ratio change yet. The step rate will be scaled, and the microstepper is rated to handle up to 200khz, so I think I have enough margin in the Arduino.
I made an expanding mandrel for modifying timing pulleys the other day and my cheap arbor and saw were really bad, so I decided making a good arbor would be good training. I hit all my dimensions to a couple of tenths, so I must be learning something.
These machines can rip their own guts out in low gears, as I have some unfortunate experience with. The stepper motor's torque limit is something of a feature on a machine with no clutch, and I think I'm still far short of the forces change gears can apply (famous last words?).
Me getting tenths is impossible,, carpenter wants to blame tools here.. wish I had ballscrews everywhere. Well done indeed.
The higher the drive microstepping rate, the lower the stepper torque.
Re torque, pls clarify cos I am losing the plot here, wouldnt even 590 lb force on a tool tip, say HSS, cutting a thread, at speed be a disaster?
Again, why not just do gentle cuts?
 
I haven't worked through the section in Machinery's Handbook on Estimating Machining Power yet, but 590 pounds of force is way down on the scale in this world. Consider that a Grade 8 bolt has a typical tensile strength of 150,000psi. I remember Tom Lipton (oxtools) mentioning the amount of clamping force provided by a #10 screw in his Mini Pallet project a couple of years ago and being surprised. A noob like me isn't used to thinking in those terms.
 
In theory, a stepper has an infinite torque at zero rpm. In practice, it is limited by the resistance and inductance of the stepper.The theoretical torque is inversely proportional to rpm, torque x rpm = constant. The constant is dependent upon the geometry of the stepper and its electrical characteristics and will vary within motors of the same frame size and detent torque. We used to use NEMA 17 motors from Oriental Motors and they have published torque/rpm curves for their motors. The curves vary quite a bit between similar motors.

I made a decision on a motor for the 602 based in part on James' success with his hybrid stepper and that my Tormach steppers are 500 oz.-in. I ordered a 600 oz.-in. hybrid stepper and driver kit, https://www.ebay.com/itm/233342928169 . I plan on using 1:1 pulleys to drive the lead screw. I'm hoping that the inproved high speed response of the hybrid stepper will be sufficient. My gut is telling me that the spindle motor will stall first. If necessary, I can use the 2:1 gearing in the gear box to boost the torque.

My next concern is the power supply. The stepper has three phase windings and a rated current of 5.8 amps/winding. Does this mean that I need a power supply with a 20 amp rating? At 48 volts, that's 1,000 watts. That seems a more than a bit large. The Tormach has a 400 watt linear supply and is driving two 500 oz.-in steppers and one 700 oz.-on. stepper.
 
Pleased with my progress today. Enclosure for the electrics / electronics sorted.
9A2D3FB5-1372-4181-8E61-A50A6DDF84EE.jpeg
EADE0E37-873F-425F-B038-2B880B9FA0CD.jpeg


Circular connectors for all the connections to the outside world. Smaller 6 pins will be for display, spindle encoder and servo encoder. The larger 4 pin will be for servo power.

705DD6DA-422F-42A6-A914-7FBC2A17B4ED.jpeg
C6F676C7-AB23-430B-81B4-99DE7C62B6FF.jpeg


Now I need to wire it all up, and fit a fan on the front cover to ensure a constant flow of air
 
So, I think I'm going to take the plunge too. I've had ELS on my mind for a couple of years now. I just bought a TI development board to start noodling with. I am impressed at the depth and thoroughness of his design process. It is great to see a focused, high-quality design process that is documented that well right out of the gate. It is also very good that he kept it simple by making it a single purpose implementation to start out. There is lots of headroom in the chosen processor for added features later. Well done all around AND he has a day job. Very inspiring.

My plan is to build up an interface board that has similar functionality as Clough42's but that has differential receivers for the encoder signals because I have some of those encoders and like the noise immunity these encoders may provide. I have a couple of big motor drives nearby and that concerns me noise-wise. I will be using Clearpath step/direction servos because I have a couple of them leftover from a prototype I did.

This is going to be fun! Thankfully I retire next year so I should actually have some time for it.
 
yeah, hey sorry re that drive ratio error, been jumping across too many ELS stories!;)
am not in favour of hi pulley ratios tho as lead screw speed will drop off, without using a servo motor, so gentle cuts on long-ish threads may become tediously time consuming, yes?
BTW, I bought a reasonable taiwanese MT3 slitting arbour, cos time vs $$ didnt add up, but well done bro.
Go easy on ur machine! 1000lb on 1/2 nuts for 3/4" dia screw is nasty
I said in my earlier posting that 800 steps per rev was as low as my microstepper could go, but that is wrong. The setting in the table that's labeled "default" is 400 steps per rev. I said in an even earlier post that I would soon have a 12T pulley and be able to go to a 10:1 ratio. That was wrong, too. I misread the product description, which is for an HTD5M pulley, not the HTD3M that I'm using. With the 8:1 ratio I have on the machine, at 400 steps per rev, without changing my code, I'm feeding 25% too fast (25,600 steps per inch vs 32,000) already, so there will be no problem with the step rate that I can generate. I don't see any difference in the static torque test at 400 steps per rev vs 800.
 
Back
Top