Working on my 1st 3D printer and building a Delta.

Working on my 1st 3D printer and building a Delta. One item I am running into is finding a breakout board in the 3D community that can handle minimum 48V and 1.68A or 60V and 1.25A. As the Nema17 motors that i am looking at have a sweet spot of either 50V (Going to use 48V as power supplies are easy to get for that) or 60V. As such I am willing to use standard CNC drivers but the issue is either I use a separate rail to power the drives and set up the pins independently for the controls. This Delta is going to be driven by Ball screws (1204).

Equation that I am using for finding the Voltage is the square root of inductance * 32.

any reason you are wanting to use a 48volt ps instead of the standard 12 or 24 volt?

and look at

you can buy them from ebay as well

these motors will make your delta sing unless you are doing something crazy with it. save your money on the motors and powersupply and spend it on your controller for a nice 32bit controller

That equation for PSU voltage doesn’t really apply to 3d printers. It’s intended for large motors in CNC mills with high gear reductions, where you need extremely high motor RPM to achieve high mechanical power and reasonable tool speed. In 3D printers, we don’t care much about mechanical power and use much lower transmission gearing ratios. So the motor RPMs required are lower, and the PSU voltage required is lower. A 24v PSU and low-inductance stepper motor will be absolutely fine for any motion speeds your extruder can keep up with.

That said, why on earth are you using screws? A 4mm lead ballscrew will severely under-perform a belt-and-pulley drivetrain in a 3D printer, at higher cost.

Ryan, a 4mm pitch ballscrew will have less issues and longer life span then the belt drive. The reason for the 48V is working closer to optium voltage for a given motor insures less current used and also much less chance of loss steps. which in turns increases the rate at which you can run the machine. That is the reason for the want to have the voltage near the 50V that the drivers are close to wanting. In accuracy of drive train from most to least it goes linear drive, ballscrews, and then beltdrives. There are other options as well but rack and pinion has way too many requirements to run efectively to be worth the invest (though on a CNC Gantry router that completely changes depending on main material to be proceesed).

Voltage does not only have to do with mechnical advatnage when dealing with steppers.

@Michael_Marino what you’re describing is a very expensive and slow 3D printer. That’s fine if that’s what you want to build. Just understand that 3D printers (particularly deltas) and CNC mills have substantially different mechanics and design optimization challenges.

For starters, there are no tool loads to speak of in printers, so the near-entirety of torque requirements on the motors are inertial. Big screws add considerable rotating inertia, yet the large mechanical advantage and stiffness buys you little benefit in terms of moving the light-weight delta mechanism after the ballscrew nut. This is why >99.9% of deltas use belts: they’re plenty stiff and strong and long-lasting for the light-duty service conditions. In comparison, screws require lubrication and considerably slow down the printer.

3D printer motion controllers also cause pretty violent drivetrain impulse-loading at toolpath corners to traverse them with non-zero speed and prevent extruder residual pressure blobbing. This is perfectly acceptable in a relatively flexible light-duty printer, but can can cause damage in high-stiffness mill-style drivetrains. It’s a different trajectory control philosophy that implicitly relies on a certain class of drivetrain. (The main exception is MachineKit.)

Then there are top speed issues. A delta may require up to 3x higher carriage speed than end-effector speed, due to arm geometry. At 4mm/rev, if you want to reliably hit a pretty normal 150mm/s travel speed, you will need 6750 RPM at the motor! Good luck with that, even at 50v. To say nothing of motion controller step pulse frequency limitations.

When people put screws in deltas, which is rightfully rare to begin with, they almost always use faster screws than 4mm/rev.

350mm/sec sounds reasonable to me and that is not max speed for that screw size. I own and run a CNC that is capable of that while moving a gantry and maintaining accurate positioning. Lubricating a screw nut is a monthly item and not a lot of work (zerk fitting and grease gun). Have you ever done real world work with CNC?? When milling many types of materials outside of metals (I think you are stuck in the metal side of CNC which is only one part), I have machines that are doing 500mm/sec accelerations on slow machines and some higher then 1.1m/sec accelerations. It depends on what you are doing with it. yes these are digital drives that are doing this and yes they are meant for serious continuous work load. many not cutting anything harder then cast acrylic or sky board and cutting at a range of speeds.

With 3d printing the control that the screw gives is accuracy in step and positioning in .08 of a mm over a 300mm for C7 screws. that is without positional sensors on the steppers. that is the limitation of a the screw. Belt drive can not offer that, physical fact. using optimum or near optimum voltage reduces the current used and insures that the steps are registered by the motor. This in turn increases the acceleration that is poosible with a given drive that in turn allows the CAM to much more smoothly control the motion of the combined steppers (depending on the software used to drive it and the how far ahead the software has ability to look).

I appreciate the offer of information. Like I said, I have years experience in CNC and while there is some important points that are different when dealing with 3D printers, inertia is constant for given values. Which brings me back to the voltage issue as having the ability to get near or at proper voltage will directly increase the velocity and acceleration. Which when dealing with corners is extremely important (Lithophanes require rather high accelerations on the Z to make any reasonable production cycle possible).

Thanks for the input folks. Learned a good bit.

Most 3D printer deltas run 2-3 m/s^2 acceleration AND 20mm/s “jerk” at corners. (Not real physics jerk, it’s a poorly-chosen name.)That is an instantaneous commanded velocity change that requires the motor to “pull-in” to the new speed in less than about 10 milliseconds. Those jerks often require 10-30 m/s^2 accelerations for brief periods to avoid the motor losing sync. Normal CNC motion controllers simply don’t do that. It’s a unique quirk to 3D printer controllers derived from GRBL.

My estimate from your voltage and screw (doing math with some typical motor specs) is that your column carriages will be utterly maxed out with minimal torque at 200mm/s, and have to run at less than half of that to reliably handle jerks and avoid skipping steps. Which means your top practical end-effector speed might be ~120mm/s at bed center and 40mm/s at bed edge.