- March 12, 2018 at 7:52 pm #54969
Just stumbled across this
Imagine the controller being centrally located near the computer…cat5 cable carries the signals to the drivers that are practically directly connected to the steppers…only caveat is now you have to wire power directly to the drivers on the motors…
But you no longer get current/voltage losses thru long runs of cable, can run higher voltages to the drivers because the drivers and the controller are no longer on the same board
Similar to the ustepper, but a much more inexpensive and simple…remember the controller only sends the signals to the driver, and this would just be a driver connected to the stepper. The signal is still generated by the microcontroller..but is now fed remotely to the motor drivers on the back of the stepper.
The ustepper takes it MUCH further by using a magnet and hall effect sensors and directly mounted microcontrollers to ensure steps are exact, but this is MUCH to overkill for our purposes..but the simple poor mans remote stepper driver might be something a genius at R&D could look into.
Because just like with the stepper breakout board I have shared on OSHPark…the only thing needed to make the stepper driver control the motor is the Step, Direction, and Enable signals, and clean DC voltage
Just saw this and though that it might be interesting idea for large builds of the MPCNC or the LowriderMarch 14, 2018 at 9:04 am #55135
A better way to wire them up would be nice.March 14, 2018 at 2:25 pm #55157
David RabbitParticipantBut you no longer get current/voltage losses thru long runs of cable, can run higher voltages to the drivers because the drivers and the controller are no longer on the same board
Is this the main payoff? Because you could also get that simply by using thicker gauge wire with the existing configuration. 🙂March 18, 2018 at 8:27 pm #55614March 19, 2018 at 8:24 pm #55704March 20, 2018 at 2:59 am #55710
Dui, ni shuo de duiParticipant
Im wondering if this is actually a good or a bad idea.
You’ll end up having signal wires (very low power) running all over your CNC. That’s a big risk of catching nasty noise.
Plus if you also have to run power wires separately, I’m not sure to see the point.
Except of course if there is some kind of encoder to make it a poor man’s servo, that’s the only case I see that could make sense.March 20, 2018 at 3:39 am #55711
Jeffeb3ParticipantYou’ll end up having signal wires (very low power) running all over your CNC. That’s a big risk of catching nasty noise.
It’s a tradeoff. Lower power means more susceptible to noise, but digital means higher tolerance for noise, since the signal will have to be more than 2.5V off to matter. Twisted Ethernet cable would be fine in almost every case (not sure about the plasma cutter you have, Dui).March 29, 2018 at 11:15 pm #56613
My build uses three 23 gauge cat6 cables – complete with 8p8c plugs – for the axes. In my brief test they seem to work. Should be plenty wires (for endstops of whatever) and plenty thick. If I stow or tear it down I can use them for what they are.
March 30, 2018 at 10:29 am #56646
- This reply was modified 3 weeks, 6 days ago by buurin.
David RabbitParticipantAsk, and it just shows up
Just found this!
I like this idea. What kind of communication protocol is being used to talk to the remote driver boards? Doesn’t look like I2C, SPI, or serial. Or is there none, and it’s sending just straight TTL, the same as if the driver were on a local board?
You must be logged in to reply to this topic.