Motors for different configurations and uses

Hardware
1

Here is where I’d like to discuss motors for different machine configurations, uses and price points. I am developing the Maslow further for the open source market and would like to have contribution of the community and build solutions that people with existing systems can upgrade to if they’d like.

2

The motor requirements are:

  1. must not be able to be back-driven by the load of the sled (must not creep
    when the machine is turned off)

    currently we use a worm gear gearbox, I suspect that a cyclodial gearbox would
    be better (less wear, stronger configurable to high gear ratios)

  2. must have a high enough resolution on the chain/line length measurement
    (currently we are around 0.01mm which is ~1/50 of our desired machine
    resolution)

    This is probably overkill, but you need at least 4x machine resolution for
    your actual permission, and some additional resolution or you will end up
    hunting back and forth around the target position

  3. it needs to have enough torque to support the sled along the top of the
    machine with enough power to move things quickly.

    Currently the static forces along the top of the machine are ~30 pounds of
    tension on each chain and the motors are rated at ~60 pounds of force and are
    marginal

Beyond that, there are the usual trade-offs of size, voltage, current and final
drive speeds to consider.

David Lang

1 Like
3

Cost is another factor…

4

Definitely, looking at costing out three motor solutions. Original+ to take advantage of the higher capacity of the new controller, mid range to allow for higher feed rates and an upper tier that can take the cutting forces of a spindle going through steel plate.

5

remember that gravity is one of your forces available, and you can only adjust
it by adding weight (which increases friction as well), you need to add a LOT of
weight to get significantly higher cutting forces.

the bottom corner problem is that the cutting force available is the tension on
the chain going to the far motor, so with the stock maslow, you only have 3.2
pounds of cutting force (minus friction losses)

David Lang

6

https://odriverobotics.com/#features

7

Yep, I’ve been watching them for a good while (and have one of their boards)

they are also a motor/encoder setup, but using hobby brushless DC motors and a
separate encoder (the encoder info I posted above was started from the odrive
discussions)

There would still be a need to run those motors through a gearbox to eliminate
backdriving. I really want to do some testing on cyclodial gearboxes and see how
easy/cheap we can make a strong enough one (can we 3d print one that’s strong
enough?, can the critical part be machined easily?)

thanks to quadcopters, small, high speed (low voltage and high current
unfortunantly) brushless DC motors are very cheap, if we can gear them down to
reasonable speeds and then control them precisely, they can be a very good
option.

David Lang

8

I wouldn’t try to 3d print that unless you get it sent out for SLM for the greater strength. As far as CNC I’ve got a 4 axis CNC with the precision if needed. The odrive can run something similar to the brushless motors used for under 1 KW spindles. That leads to the possibility of sizing a motor sufficient for baseline spindle and movement drives. Quantities of scale get easier if all four motors are the same with just different hardware attached.

9

At $120 each for only 2 axes an odrive takes a giant chunk out of the $500/600 price point

1 Like
10

odrive is drastic overkill for a maslow, we don’t need to run multiple KW
motors to move it. (we currently allow 2A per motor, not the 100A that the
odrive supports)

it shows what could be done, and given the much lower requirements for our
motors, something similar could be built for a lot less money.

11

Odrive is open source so we can use the design to help. Thinking of using the high powered Odrive design for another option. Going to be offering several different options to build the machine you’d like.

12

I’m working on a 2 motor device based on the Maslow drivers, software, etc. I’d like to find a similar motor but with twice the power. Any suggestions?

Also, are planetary gear boxes better for forward/reverse compared to the Maslow’s gearbox?

13

planetary gear boxes suffer from being backdriven, so they don’t work well. We
had someone who did a maslow with steppers, and he tried planetary gears, he
ended up replacing them with worm gears