2011/10/10 andy pugh <[email protected]>:
> 2011/10/10 Viesturs Lācis <[email protected]>:
>
>> What can potentially be wrong?
>
> It might be a race condition, that program.run fails if mode is not auto.
>
> I have been playing around, and if you net halui.mode.is-auto to
> halui.program.run (and make sure that the pin is on a momentary
> button, or wired through a oneshot) then it seems to run reliably
> simply by setting mode.auto
> I have found an unexpected side-effect that the program auto-runs if
> you press "reload" which might not be as-desired..
>
> You might be able to use two one-shot components, one triggering
> mode.auto from the button, and the second triggering from the falling
> edge of the first and with the output setting program.run.

Thank You for the idea! I feel very high probability that You might be right.
Current scheme is:
1) input pin is driven by pneumatics controller - it goes high, when
part is fixed in place;
2) signal from input pin is filtered through debounce module;
3) filtered signal from debounce is routed to oneshot component to
check for rising edge;
4) oneshot output goes to and2 module - second and2 input is from
pyvcp button, enabling this whole process;
5) and2 output goes to halui.mode.auto and halui.program.run

After writing the previous email I tested some more and got better
results - now it seemed that 1 out of 10 cases did not start execution
of file. But I did not really change anything in settings, only the
width of oneshot output pulse was increased from initial 0,25 seconds
to 5 seconds.

I will try to check, if Your suggestion helps, when I will get my
hands on the machine next time.

Viesturs

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to