Dynomotion Forum

I have a KFLOP/KSTEP setup and would like to output one of the unused KFLOP PWMS (any one of 1-7) on the KSTEP RELAY1 output.
Alternatively, I could add an extra opto board, wire its input to a KFLOP JP6 PWM output and run more connections to my CNC router, but since the relay1 output is already wired all the way to my gantry, it would be really nice if I could simply mux a PWM output to IO1 on the KFLOP.
Is this possible?

Thanks,
RJ Wilcox

Hi RJ,

There isn't any such mux option. I assume you are using PWM0 for the KStep Analog output?

You might hardwire a connection from KFLOP JP6 to JP7.

Would a software PWM be fast enough?

Yes, I'm using PWM0 for analog spindle frequency control.

Both suggestions give me some better possibilities.

If I hardwire a PWM output on JP6 to IO1 on JP7, can I assume that as long as IO1 is never programmed as an output, there will never be an output conflict between the two signals? Or will I need to cut the IO1 trace on JP7?

The PWM software suggestion is intriguing but I don't think it will give me enough resolution at a 1KHz PWM frequency. With a 90us scheduling clock, there would only be 11 possible pulse widths in the 1ms period.

Thanks for the quick response,
RJ

Hi RJ,

If I hardwire a PWM output on JP6 to IO1 on JP7, can I assume that as long as IO1 is never programmed as an output, there will never be an output conflict between the two signals?

Yes

The PWM software suggestion is intriguing but I don't think it will give me enough resolution at a 1KHz PWM frequency. With a 90us scheduling clock, there would only be 11 possible pulse widths in the 1ms period.

Do you need 1KHz PWM frequency? You can all use a delta-sigma approach to get better resolution and performance. You might see this article.

I think the hardwired solution fits best. The low frequency PWM is interesting, however. I'll give it some thought.

Thanks again for an excellent product and support,
RJ