Sineos
4h
Heureka! Good things come to those who wait (or check their stepper motors) 
Personally I would advise against using Nema 23 in 3D printers
- To my knowledge, those who tried reverted back
- Not needed / no benefit for regular consumer printers of relatively small size. What do you want to do with the additional torque? Wrap the print head 3 times around the frame in case of a crash? CNC is a different story, there you need the torque
- Higher inertia thus not really fit for the small and quick movements during 3D printing.
That has already crossed my mind. But closed loop in nema-17’s leaves us so far with one choice, and it isn’t as stable as it should be, I have 3 of those and its relatively easy to get them to a buzzing condition. So I have 6 of the longer .92nm motors coming from amazon for about $75. If they won’t run fast enough on 24 volts, I have boxes of 50 volt capable drivers to get the speed back. I’m not married to tmc2209’s, or even the outrageously priced ($24/copy) fystetc QVH5160’s I could plug into an octopus-pro. There’s room for a 48 volt supply and a pair of DM542’s in the bottom of that printer, and it has been my experience the speed goes up faster than the voltage. 28 volts on an early mill just barely got me 8mm/sec,
39 volts on the same motor/driver on the next machine got me almost 40mm/s.
I’m looking for ways to reduce the weight the Y motor has to throw around on my ender 5 +. That includes moving
the x home switch off the x bar which will be CF carrying a linear rail when I get the ability to print parts back. Its lost Y home lots of times but x only once in 2 years. The nema-23 is already mounted & driving the rear axle, now a 25mm OD CF tube, at 1/1, but will require multiplying the Y count by 1.5 as that’s a 1.2 degree motor. Rotor inertia might be a problem, OTOH a pulley pair to slow the bigger motor is only a week away. That also means homing is two LOC, y first, unless g28 can reverse the xyz order.
So when the longer (60mm core) motors get here, I’ll hide a 650 watt 48 volt supply set for 45 volts and a pair of DM542’s in the bottom, and klipper will never know they are there. Remove the tmc2209’s from printer.cfg and use the nano’s en,step/dir right from the headers on the nano.
One question about that. In LCNC’s typical break-outs, we run a 5 volt buss to the + end of the drivers input opto’s, and assume the active state is low. Since that is likely a direct connection to the stm407 on the nano, is the low state capable of sinking more current than the high state can source? The drivers on the typical cnc setup are ttl and can sink lots more current than they can source. In that event, can I break the ground and connect that to a 5 volt line like we do in LCNC?
That’s a big enough wall of test for one message.
Take care and stay well, Sinoes
Obviously I’ve not played with any of these FPGA’s now in everything. This tech is lots newer than I am. ;o)>
Then I cabbage a pair of 60mm core nema-17’s off a couple 50/1 harmonic drives I made a couple years ago. Spend 3 + hours changing them, and find they are at least 15mm too long, and the base hits the crossbar supporting the Z bearings on the sp-5. The only shorter than that is some dual shafted, like the ender 5 uses for its puny Y, they might clear by a mm… Murphy is determined to screw this old man over at every turn. Later, after I change them again.
