ComputerVoltageSources

Just to not mix multiple themes in messages ...

Consider the need to detect narrow triggers. I started to look into
this with my PSIM but none of the I/O pin interrupts are on external
connectors. Having an edge or level interrupt would be very useful to
detect triggers. In several of my programs I resort to monitoring
changes in CV and, of course, gate signals to determine if there
should have been a trigger. It would be real nice to have a way to
capture the trigger directly.

If you can't do this via interrupts, then I'd consider at least one
edge capture & reset circuit for an external trigger input. I do
think you can do it with interrupts. Let's hope that some of the 4
additional pins on the AtomPro28 have external level or edge
interrupts.

Dave

There are no Atom Pro 28M or 40M pinouts from Basic Micro yet, though they
have placeholders for the data sheets and the port assignments are shown in
the 28M schematics. A Renasas 3664 only has 4 IRQs and the 24M already
brings out 3 of them, and the fourth is used for the serial port TX signal.


To reduce input count requirement, what about making the mode selectable?
Program a port IO expander to set the modes, which toggles analog switches
to toggle the modes. If we want to make this compatible across programs,
now is the time to do it before the hardware is designed.

John Loffink
The Microtonal Synthesis Web Site
http://www.microtonal-synthesis.com
The Wavemakers Synthesizer Web Site
http://www.wavemakers-synth.com

> -----Original Message-----
> From: ComputerVoltageSources@yahoogroups.com
> [mailto:ComputerVoltageSources@yahoogroups.com] On Behalf Of djbrow54
>
> Just to not mix multiple themes in messages ...
>
> Consider the need to detect narrow triggers. I started to look into
> this with my PSIM but none of the I/O pin interrupts are on external
> connectors. Having an edge or level interrupt would be very useful to
> detect triggers. In several of my programs I resort to monitoring
> changes in CV and, of course, gate signals to determine if there
> should have been a trigger. It would be real nice to have a way to
> capture the trigger directly.
>
> If you can't do this via interrupts, then I'd consider at least one
> edge capture & reset circuit for an external trigger input. I do
> think you can do it with interrupts. Let's hope that some of the 4
> additional pins on the AtomPro28 have external level or edge
> interrupts.
>
> Dave
>

It seems like we could just put a simple RC pulse strecher on the triger input.

That should allow enough time for detection with a software scan.

There is always goin to be some latency. 5-10 milliseconds is not a lot.

Or do we really need to generate an interupt from the trigger input jack itself?

--- In ComputerVoltageSources@yahoogroups.com, "John Loffink" <jloffink@...> wrote:
>
> There are no Atom Pro 28M or 40M pinouts from Basic Micro yet, though they
> have placeholders for the data sheets and the port assignments are shown in
> the 28M schematics. A Renasas 3664 only has 4 IRQs and the 24M already
> brings out 3 of them, and the fourth is used for the serial port TX signal.
>
>
> To reduce input count requirement, what about making the mode selectable?
> Program a port IO expander to set the modes, which toggles analog switches
> to toggle the modes. If we want to make this compatible across programs,
> now is the time to do it before the hardware is designed.
>
> John Loffink
> The Microtonal Synthesis Web Site
> http://www.microtonal-synthesis.com
> The Wavemakers Synthesizer Web Site
> http://www.wavemakers-synth.com
>
>
> > -----Original Message-----
> > From: ComputerVoltageSources@yahoogroups.com
> > [mailto:ComputerVoltageSources@yahoogroups.com] On Behalf Of djbrow54
> >
> > Just to not mix multiple themes in messages ...
> >
> > Consider the need to detect narrow triggers. I started to look into
> > this with my PSIM but none of the I/O pin interrupts are on external
> > connectors. Having an edge or level interrupt would be very useful to
> > detect triggers. In several of my programs I resort to monitoring
> > changes in CV and, of course, gate signals to determine if there
> > should have been a trigger. It would be real nice to have a way to
> > capture the trigger directly.
> >
> > If you can't do this via interrupts, then I'd consider at least one
> > edge capture & reset circuit for an external trigger input. I do
> > think you can do it with interrupts. Let's hope that some of the 4
> > additional pins on the AtomPro28 have external level or edge
> > interrupts.
> >
> > Dave
> >
>

I probably started this and I'm having second thoughts. The issue
with stretching the pulse is that you want to detect a trigger only
once. So, having detected it with a stretched pulse, you now need to
ignore it until it goes away. My initial thought was that you could
assign the AtomPro pins appropriately such that you could use one of
the hardware level interrupts. I'm assuming these are edge and not
level. However, if they are level, then you have the same issue. In
thinking about this, I'm starting to come to the conclusion that you
can detect triggers via timer interrupts.

I general run my timer interrupt at 1 mS. Most of my triggers are at
least this wide (Is this a true statement in general?). Assuming
triggers are at least 2 mS, it should be pretty simple to sample the
digital input and look for a state change in a timer interrupt
routine. This state change could then set a global variable that a
trigger had occurred. The software could poll the variable and decide
what to do with it whenever. If you need immediate action, you can
always do this inside of the timer interrupt routine.

I think I can code this up in a pretty minimal overhead mechanism. I
should have thought of it earlier as I already have interrupt routines
for input averaging the analog inputs.

This is probably a simpler solution that would work and maintain code
compatibility and minimal hardware.

Dave

--- In ComputerVoltageSources@yahoogroups.com, "Grant Richter"
<grichter@...> wrote:

>
> It seems like we could just put a simple RC pulse strecher on the

triger input.

>
> That should allow enough time for detection with a software scan.
>
> There is always goin to be some latency. 5-10 milliseconds is not a

lot.

>
> Or do we really need to generate an interupt from the trigger input

jack itself?

I'd say its a bad idea to rely on some minimum or maximum trigger
length... I've seen sub-millisecond triggers... and some much longer.

If you get into triggers more than a few milliseconds long, they are likely to
affect the sound of the modules they are controlling.

If its a problem for a particular user, they could do the RC trick
(which does NOT work if the trigger is already generated by an unbuffered
RC) ... or add an active one-shot for their particular issue.

Can't make everyone happy :^P (except me)

H^) harry

djbrow54 <davebr@...> wrote: I probably started this and I'm having second thoughts. The issue
with stretching the pulse is that you want to detect a trigger only
once. So, having detected it with a stretched pulse, you now need to
ignore it until it goes away. My initial thought was that you could
assign the AtomPro pins appropriately such that you could use one of
the hardware level interrupts. I'm assuming these are edge and not
level. However, if they are level, then you have the same issue. In
thinking about this, I'm starting to come to the conclusion that you
can detect triggers via timer interrupts.

I general run my timer interrupt at 1 mS. Most of my triggers are at
least this wide (Is this a true statement in general?). Assuming
triggers are at least 2 mS, it should be pretty simple to sample the
digital input and look for a state change in a timer interrupt
routine. This state change could then set a global variable that a
trigger had occurred. The software could poll the variable and decide
what to do with it whenever. If you need immediate action, you can
always do this inside of the timer interrupt routine.

I think I can code this up in a pretty minimal overhead mechanism. I
should have thought of it earlier as I already have interrupt routines
for input averaging the analog inputs.

This is probably a simpler solution that would work and maintain code
compatibility and minimal hardware.

Dave

--- In ComputerVoltageSources@yahoogroups.com, "Grant Richter"
<grichter@...> wrote:

>
> It seems like we could just put a simple RC pulse strecher on the

triger input.

>
> That should allow enough time for detection with a software scan.
>
> There is always goin to be some latency. 5-10 milliseconds is not a

lot.

>
> Or do we really need to generate an interupt from the trigger input

jack itself?






SPONSORED LINKS
Music instrument stores Electronic Instruments Module

---------------------------------
YAHOO! GROUPS LINKS


Visit your group "ComputerVoltageSources" on the web.

To unsubscribe from this group, send an email to:
ComputerVoltageSources-unsubscribe@yahoogroups.com

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.


---------------------------------





[Non-text portions of this message have been removed]

yeh...
what Harry said!
I've been thinking about this too and i endorse Harry's position.
In the wild, there's just no 'right' way position every kind of
pulse... and satisfy every kind of application.

Passive RC,
Active one shot (ala 555),
or software,
are all 'best' for different circumstances and making a choice now
(kinda) limits the flexibility.
of course...
(you can always dremel off the offending components) and solder right
on to the exposed traces)
(but i don't wanna gross anybody out)
;'>
-doc


--- In ComputerVoltageSources@yahoogroups.com, Harry Bissell Jr
<harrybissell@...> wrote:

>
> I'd say its a bad idea to rely on some minimum or maximum trigger
> length... I've seen sub-millisecond triggers... and some much longer.
>
> If you get into triggers more than a few milliseconds long, they

are likely to

> affect the sound of the modules they are controlling.
>
> If its a problem for a particular user, they could do the RC trick
> (which does NOT work if the trigger is already generated by an

unbuffered

> RC) ... or add an active one-shot for their particular issue.
>
> Can't make everyone happy :^P (except me)
>
> H^) harry
>
> djbrow54 <davebr@...> wrote: I probably started this and

I'm having second thoughts. The issue

> with stretching the pulse is that you want to detect a trigger only
> once. So, having detected it with a stretched pulse, you now need to
> ignore it until it goes away. My initial thought was that you could
> assign the AtomPro pins appropriately such that you could use one of
> the hardware level interrupts. I'm assuming these are edge and not
> level. However, if they are level, then you have the same issue. In
> thinking about this, I'm starting to come to the conclusion that you
> can detect triggers via timer interrupts.
>
> I general run my timer interrupt at 1 mS. Most of my triggers are at
> least this wide (Is this a true statement in general?). Assuming
> triggers are at least 2 mS, it should be pretty simple to sample the
> digital input and look for a state change in a timer interrupt
> routine. This state change could then set a global variable that a
> trigger had occurred. The software could poll the variable and

decide

> what to do with it whenever. If you need immediate action, you can
> always do this inside of the timer interrupt routine.
>
> I think I can code this up in a pretty minimal overhead mechanism. I
> should have thought of it earlier as I already have interrupt

routines

> for input averaging the analog inputs.
>
> This is probably a simpler solution that would work and maintain code
> compatibility and minimal hardware.
>
> Dave
>
> --- In ComputerVoltageSources@yahoogroups.com, "Grant Richter"
> <grichter@> wrote:
> >
> > It seems like we could just put a simple RC pulse strecher on the
> triger input.
> >
> > That should allow enough time for detection with a software scan.
> >
> > There is always goin to be some latency. 5-10 milliseconds is not a
> lot.
> >
> > Or do we really need to generate an interupt from the trigger input
> jack itself?
>
>
>
>
>
>
> SPONSORED LINKS
> Music

instrument stores Electronic
Instruments
Module

>
> ---------------------------------
> YAHOO! GROUPS LINKS
>
>
> Visit your group "ComputerVoltageSources" on the web.
>
> To unsubscribe from this group, send an email to:
> ComputerVoltageSources-unsubscribe@yahoogroups.com
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of

Service.

>
>
> ---------------------------------
>
>
>
>
>
> [Non-text portions of this message have been removed]
>

Trigger success!

It turns out I read the AtomPro24 datasheet incorrectly the first
time. IRQ1 is on P8 which is the AUX input/output. You have to program
the PMR1 register (took a while to figure that out) to select this
input as an interrupt. The interrupt detects negative edges. I can now
generate an interrupt on a trigger edge. Great!

It seems to work reasonably well with a positive trigger. It just
detects the trailing edge. However, you get one extraneous interrupt
when you enable IRQ1 since the level is low.

Inverting the signal for a negative trigger works well and eliminates
the extraneous interrupt when you enable IRQ1.

Just make sure that P8 is used as a digital input. The only way I see
to eliminate the extra interrupt for a positive trigger would be to
invert it on the PCB or via an external module as I have done to test
this. I'll keep playing with it as I have time to see if I can figure
out a way in software to reset this first interrupt. Off to read the
hardware manual.

Dave


--- In ComputerVoltageSources@yahoogroups.com, "drmabuce" <drmabuce@.
..> wrote:

>
> yeh...
> what Harry said!
> I've been thinking about this too and i endorse Harry's position.
> In the wild, there's just no 'right' way position every kind of
> pulse... and satisfy every kind of application.
>
> Passive RC,
> Active one shot (ala 555),
> or software,
> are all 'best' for different circumstances and making a choice now
> (kinda) limits the flexibility.
> of course...
> (you can always dremel off the offending components) & solder right
> on to the exposed traces)
> (but i don't wanna gross anybody out)
> ;'>
> -doc

Found a couple of more registers to control irq's. I can now select
the inputs for rising or falling edges. I can also clear extraneous
inputs for an active high signal. For some reason they only occur if
the input is normally low.

Anyway, to detect triggers via interrupts, just make sure that
external access is available for p6 - p8. I would select p8 to make
it compatible with the aux i/o on the PSIM.

Dave


--- In ComputerVoltageSources@yahoogroups.com, "djbrow54" <davebr@...>
wrote:

>
> Trigger success!
>
> It turns out I read the AtomPro24 datasheet incorrectly the first
> time. IRQ1 is on P8 which is the AUX input/output. You have to

program

> the PMR1 register (took a while to figure that out) to select this
> input as an interrupt. The interrupt detects negative edges. I can

now

> generate an interrupt on a trigger edge. Great!
>
> It seems to work reasonably well with a positive trigger. It just
> detects the trailing edge. However, you get one extraneous interrupt
> when you enable IRQ1 since the level is low.
>
> Inverting the signal for a negative trigger works well and

eliminates

> the extraneous interrupt when you enable IRQ1.
>
> Just make sure that P8 is used as a digital input. The only way I

see

> to eliminate the extra interrupt for a positive trigger would be to
> invert it on the PCB or via an external module as I have done to

test

> this. I'll keep playing with it as I have time to see if I can

figure

> out a way in software to reset this first interrupt. Off to read the
> hardware manual.
>
> Dave
>
>
> --- In ComputerVoltageSources@yahoogroups.com, "drmabuce" <drmabuce@
> ..> wrote:
> >
> > yeh...
> > what Harry said!
> > I've been thinking about this too and i endorse Harry's position.
> > In the wild, there's just no 'right' way position every kind of
> > pulse... and satisfy every kind of application.
> >
> > Passive RC,
> > Active one shot (ala 555),
> > or software,
> > are all 'best' for different circumstances and making a choice now
> > (kinda) limits the flexibility.
> > of course...
> > (you can always dremel off the offending components) & solder

right

> > on to the exposed traces)
> > (but i don't wanna gross anybody out)
> > ;'>
> > -doc
>