I'm wondering if there is an even simpler (but more expensive) way to take care of the backlash (if it's even an issue). This monster may be heavy enough that it really does not have much backlash.:
Buy two of the nuts and mount them flange to flange. Put some springs in the middle of the resulting "sandwich" and you have a pretty tight nut. If nothing else, it would double the length of the nut on the shaft and reduce the tendency of the assembly to tip back and forth.
I’ll try the printed shoulder approach from the original first and see how things work out. Since it’s a printed part it’s essentially free. (Real good reason to get a kit printer going first …).
One concern I have about going to crazy with the anti-backlash stuff is that everything probably will wear faster.
--------------------------------------------------------------------
I pinged Wantai about the max speed of the stepper controller. The person on the quick answer line didn’t know the answer. Since it was the middle of the night over there I was impressed there was a real person on the other end. Hopefully I will get an email with more data in the next day or so. The quick answer was that they probably would be good to > 600 rpm. That’s fast enough for me to print at a reasonable speed with a 0.5” thread. Hopefully the real answer will confirm at least that speed with some level of micro stepping.
--------------------------------------------------------------------
There is another side to this. After digging into the stepper code in Marlin, it will only step just so fast. It may not be happy at all (it looks like it throws an error message ...) at the kind of speeds and steps I'd love to be able to run. The ARM based solution may come into play sooner than I'd thought. I'm pretty sure it will be needed for the "ultimate" >500K steps/second solution. The Mega isn't all that fast... Of course the controller is only a 200KHz PWM so it's unlikely to be happy at that rate either. So much for 128 microsteps ....
Buy two of the nuts and mount them flange to flange. Put some springs in the middle of the resulting "sandwich" and you have a pretty tight nut. If nothing else, it would double the length of the nut on the shaft and reduce the tendency of the assembly to tip back and forth.
I’ll try the printed shoulder approach from the original first and see how things work out. Since it’s a printed part it’s essentially free. (Real good reason to get a kit printer going first …).
One concern I have about going to crazy with the anti-backlash stuff is that everything probably will wear faster.
--------------------------------------------------------------------
I pinged Wantai about the max speed of the stepper controller. The person on the quick answer line didn’t know the answer. Since it was the middle of the night over there I was impressed there was a real person on the other end. Hopefully I will get an email with more data in the next day or so. The quick answer was that they probably would be good to > 600 rpm. That’s fast enough for me to print at a reasonable speed with a 0.5” thread. Hopefully the real answer will confirm at least that speed with some level of micro stepping.
--------------------------------------------------------------------
There is another side to this. After digging into the stepper code in Marlin, it will only step just so fast. It may not be happy at all (it looks like it throws an error message ...) at the kind of speeds and steps I'd love to be able to run. The ARM based solution may come into play sooner than I'd thought. I'm pretty sure it will be needed for the "ultimate" >500K steps/second solution. The Mega isn't all that fast... Of course the controller is only a 200KHz PWM so it's unlikely to be happy at that rate either. So much for 128 microsteps ....