<https://lh3.googleusercontent.com/-1CZT9VDaixc/WE2lvJMxr9I/AAAAAAAAWdM/wWu9VhkzQzUriDJTViM-OCImyc2ej_EPACLcB/s1600/bb_pru_pwm.jpg>
Quite some time passed, but now I ran across the same issue and caught it 
on scope. It is quite easy to get one by wiring sine or triangle siggen to 
pru pwmgen and watch pwm or filtered signal on scope.
Spikes like this occur quite often.

My cape for BeagleBone, named Furaday Cape, and its users would benefit 
from high quality PWM signal. I am just not up to writing PRU drivers on my 
own.

2015 m. rugsėjis 15 d., antradienis 17:04:27 UTC+3, Charles Steinkuehler 
rašė:
>
> On 9/15/2015 8:46 AM, Marius Alksnys wrote: 
> > Transistors are switched directly. 
>
> You can probably do OK using the PRU, but the hardware PWM will 
> definitely give you more resolution.  If the main problem is the 
> spikes, it's possible there's a goof in the code that could improve 
> things a lot if fixed.  If you can check that the proper commands are 
> going to the PWM driver via halscope and see if there's actually an 
> extra long (or short) PWM pulse generated, it would verify this is 
> what's happening. 
>
> I'd try to reproduce it here, but I don't have the hardware to do a 
> closed loop handy, and I suspect the bug (if present) is 'tickled' by 
> the servo hunting that happens when at rest (which is pretty hard to 
> replicate without real-world hardware). 
>
> -- 
> Charles Steinkuehler 
> [email protected] <javascript:> 
>

-- 
website: http://www.machinekit.io blog: http://blog.machinekit.io github: 
https://github.com/machinekit
--- 
You received this message because you are subscribed to the Google Groups 
"Machinekit" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
Visit this group at https://groups.google.com/group/machinekit.
For more options, visit https://groups.google.com/d/optout.

Reply via email to