Multi-Button CSM with Pic16LF722A

[richard]

since the isr TB1 is an accurate timebase you can
make a new var

 kio_inhibit var byte

modify the I/o key routine

IF KP = 0 THEN '1/0 BUTTON PRESSED
        ;use  p_level to see what state machine is in  0=off ,non zero = on
        if kio_inhibit==0 then
           kio_inhibit=152    ;152*32mS = about 5 seconds
        if  p_level then     
            ccp1con=0 ;must be on kill it
            led=0
            p_level= 0
        else         ;must be off  power to max
            p_level= 5
        endif
       endif

and add this to the isr

if kio_inhibit then
    kio_inhibit =  kio_inhibit -1
    endif

now the I/O key is inhibited for 5 sec after being acted on

I like ART's quote and its so true

PBP is very well documented, however you’ll notice all examples of command use are also examples of exactly how not to write a cyclic program.

you need to adjust your way of thinking to loops , states and flags . it opens up a whole new vista

[RWright]

Wow, you are awesome. I'll have to try it out to see what it does. Thanks!

[RWright]

Hello again,

Was that timestamp of 3:12 as in 3:12AM?! I did the modification as above. It works like you explained. It doesn't cycle now, but the 5-second delay is way too long to be able to
power it up again. If I change the kio_inhibit value to a shorter value, will it just end up cycling again? Is there a means that if the 1/0 button is released it resets the counter to zero so that the button can be used immediately again? The long delay is fine for when the button is being held for an extended period of time. Thanks!

[RWright]

Hello again!

Just curious as to why there isn't any reference to the PR2 register in the program, for setting up the PWM? If I wanted to increase the pwm frequency to 25Khz, I take it I would have to go to a higher oscillator frequency and use different prescaler values?! Thanks.

[richard]

Just curious as to why there isn't any reference to the PR2 register in the program, for setting up the PWM?

flicker is virtually invisible to the human eye for switching frequencies above 50 Hz, switching losses increase with frequency , other than that the frequency is irrelevant .
fine control of the pwm frequency is possible using pr2 at the expense of resolution . since the frequency is largely irrelevant why not have maximum resolution

If I wanted to increase the pwm frequency to 25Khz, I take it I would have to go to a higher oscillator frequency and use different prescaler values?

depends on required resolution, @4mHz 25kHz is possible with 1:1 prescale and pr2=39 , however maximum duty occurs @ pwm_duty_reg = 160 , ie pw range is 0 to 160 (instead of 0 to 1023 steps as before)

[Aussie Barry]

Hi All,

Has anyone had any experience with a 2 channel CSM slide switch?
I would like to implement a slide switch rather than an up / down button arrangement to select the PWM output for a LED string.
In my case, the granularity is quite coarse - only eight steps of PWM selection.

Any input would be greatly appreciated.

Cheers
Barry

[RWright]

Hello,

Regarding the flicker, you know this and I know this, however, we are trying to match an existing product that is set at 25khz. It appears at lower frequencies
it is causing their power supply to make audible noise at the PWM frequency. And sure enough, at 1khz, at certain pwm levels, we can hear it in the power supply. I don't have to keep the oscillator at 4khz, but I assume making any changes to the main Oscillator will also affect the touch circuit, as well. I did see in the pwm section of the datasheet that the resolution drops depending on the oscillator frequency after I posted this. Thanks again!

[RWright]

Sorry for the repeated post.

[richard]

for barry

Has anyone had any experience with a 2 channel CSM slide switch?
I would like to implement a slide switch rather than an up / down button arrangement to select the PWM output for a LED string.
In my case, the granularity is quite coarse - only eight steps of PWM selection.

have now, seems to work nicely

slider as per http://ww1.microchip.com/downloads/e...tes/01250a.pdf
made on ds board the bottom side is unconnected, slider pad are 10mm*40mm the top is grounded
covered with sticky tape (durex for a sniker from the euro's)
see pic

[richard]

a slider design like this on single sided board is worth trying out,i think it would be more sensitive and the board could be painted to encourage the user to slide finger along the horizontal line. next time i'm doing some chemical etching I will make one [if I remember]