Victor SP Motor Controller

[IanBerry]

I've been working with my school's robotics team and they asked me to build a quicky PWM out of a 555 timer to run a Victor SP motor controller. I set it up to oscillate at 15KHz at 5 volts per the data sheet requirements, but due to the capacitance of breadboards when I put it on perf board it ended up being 17KHz. I thought this wouldn't be too much of an issue. Well, when I plugged it all in low and behold it didn't work. I got looking at the data sheet again and there seemed to be some very conflicting numbers. It says the frequency should be 15KHz but it also says the period should not exceed 2.4 ms which if I'm not mistaken is about 416 Hz. If someone could take a look and explain this to me? I'll attach a screenshot and a link to the datasheet.

http://www.vexrobotics.com/217-9090.html

[vltr]

Motor controllers like the victor most likely will use a slightly different version of PWM than what you're used to for control.  Here's a decent explanation of what's going on with the input signal: http://www.endurance-rc.com/ppmtut.php

[james_s]

Are these intended to use RC servo signals? If so, you can buy servo testers for a few dollars that have a knob to set the PWM value.

[FreddyVictor]

as james_s suggests, looks like a rc type
the 15KHz is the output, not the input frequency
just supply a pwm input with freq between 10Hz and 345Hz (100ms - 2.9ms)

looks like the page you linked has gone, presumably discontinued (appears top of page here)

[IanBerry]

Oh, okay, that makes sense. But what about the specification that says a max 2.4 ms high input pulse time? Does that mean it uses the PPM like in the link vltr sent?

[FreddyVictor]

No, PPM is somewhat different to normal PWM
PPM is a way of sending multiple - not just 1 - control signals in a single wire - saves having separate wires/connections.

your attached img says 0.6 -> 2.4ms - it just means that the 'on time' can be can be just that period of time - see 'Generic PWM pulse' on the linked page
and you need to repeat this signal at the rate (freq) in my prev post

the normal rc spec is 1.0 -> 2.0ms range, so they're just saying that their range is greater than this but you obviously need to keep within this range for correct operation

[IanBerry]

So what about the maximum period of 100 ms. If the limit for the on pulse is 2 ms than the pwm would only be operating at 2% duty cycle? And if I can only go down to a period of about 3 ms and the maximum on time is about 2 ms. Wouldn't my maximum possible duty cycle be 66%?

[FreddyVictor]

no, the 100ms period is the frequency - ie: how many times you need to send your pulse (in this case, the minimum)

it's slightly confusing since frequency can be expressed in ms as well as Hz

so:
100ms = 0.1s .'. 1/0.1 = 10Hz
so minimum of 10 times per second you need to send the pulse of between 0.6ms -> 2.4ms (min & max respectively)

If you have an Arduino handy, then you can use the Servo library
It generates (IIRC) a pulse 50times/sec (50Hz which is 1/50 = 0.02ms) with the required pulse length

[IanBerry]

Sadly I do not have an Arduino. I only have analog things on hand.
So does what you're saying mean that to control the motor control I need the increase/decrease the frequency of the 2 ms pulses?
Or does it mean I need to increase/decrease the amount of 2 ms pulses within a longer period?

[FreddyVictor]

Sadly I do not have an Arduino. I only have analog things on hand.

this really would make life alot easier if you did !

So does what you're saying mean that to control the motor control I need the increase/decrease the frequency of the 2 ms pulses?

no...
the frequency of the pulses - how often you send then is typically fixed  - choose between a range of 10 -> 345Hz as already mentioned

Or does it mean I need to increase/decrease the amount of 2 ms pulses within a longer period?

sorry to be err... negative again, but no...

to change the speed of the motor, you need to change the length of the 'on' high pulse
typically 1.0ms is slow/stopped and 2.0ms is fast - this will depend on the controller. they gave a min/max range of 0.6 -> 2.4ms

Example:
| marks a change - all are at 50Hz
For stopped: signal_low | signal_high_for_1.0ms+0.0ms=1.0ms | signal_low_and wait_19ms_before_repeating (so 1.0ms+19ms=20ms = 0.02 and 1/0.02 = 50Hz)
For half-speed: signal_low | signal_high_for_1.0ms+0.5ms=1.5ms | signal_low_and wait_18.5ms_before_repeating (apply same logic as above)
For full-speed: signal_low | signal_high_for_1.0ms+1.0ms=2.0ms | signal_low_and wait_18ms_before_repeating (apply same logic as above)

[IanBerry]

Oh okay, that makes sense. Thank you! I'll put something together to do that and see if it works.

[Ian.M]

Should be good enough to get you going.  Substitute any small signal Schottky diode for D1.  LTSPICE sim attached.