Instant Interrupts - Revisited

[mister_e]

Uh... my calculation were wrong... (1/12000000)*65536*2='round 10mSec. ARRGGH i don't have my stuff here to compare with mine... Darrel is on-line anyway

[mister_e]

NO.. let's look at that...

T1CON = $11 ; Prescaler = 2, TMR1ON

i remind that TiCON can be use in 16 bit or 8 bit.. i'm sure you use it in 8 bit mode now... or i missing something obvious about the 16/8 bit.. maybe just in read/write something... mmm something look weird in this line...clock source, mode or else.

[Darrel Taylor]

Well yeah, I'm online. But that's not going to help much since I don't have a Clue about using the USB stuff. Just never got around to trying it before.

I'm still hoping you guys can figure it out. So that when I do get around to trying it, I won't have to work as hard.
<br>
DT

[mister_e]

if i can finish the house-bench-lab-cleaning stuff soon, i'll have access to my stuff. I have this one working.. but i don't remind wich timer i used and how i used it... the T1CON line seems weird... clock source is still my most possible guess.

[Demon]

Ok, I disabled that code to disable/enable the USBSERVICE interrupt before the gosubs and now the interrupt is getting executed a LOT of times.

I can't say at what interval, but the counter is scrolling on the LCD pretty fast. The hundreds are going by just a bit faster than every second, but I still lose connection.

Darrel, how can I adjust the speed of the interrupt?

Robert

[Demon]

I'm going to solder a wire to 1 pin, blink that pin in the interrupt and connect my scope on the wire.

I don't know much about scopes, but I might be able to figure out the frequency.

Robert

[Demon]

Uh... my calculation were wrong... (1/12000000)*65536*2='round 10mSec. ARRGGH i don't have my stuff here to compare with mine... Darrel is on-line anyway

Steve, I'm trying to figure this out using math (gotta start sometime) and I'd like to know where you got the 12000000 from.

I read this link you posted in another thread:
http://ww1.microchip.com/downloads/e...Doc/41214a.pdf

And I got this formula from Tip #7:
Tosc x 4 x 65536 x prescaler

I am using a 48MHz crystal so I get this:
(1/48000000) x 4 x 65536 x 2 = 0.0109226 (repeating)

D'uh, ok, I see it now, 48 / 4 = 12, that's where you got your value. So, as you said, that interval rounds up to just over 10ms, that 'should' work.

I'm sure it has something to do with the interrupt occurring right in the middle of USB processing and interfering with the connection. I don't see what else it can be; the interrupt seems fine, the USB works.

Robert