Wardialer Rev2
I finished the revision of the wardialer. The biggest difference is that I moved the speed control from software driven to hardware driven. This was done by using an external 40106 chip as a clock. There are several gates left for other exciting things, but only two were used for the clock. The clock then drives the PICAXE.
Unfortunately, the PICAXE interrupt is not a triggered interrupt. As you can see in the Wikipedia, IRQs are a way to prevent wasted time polling a pin. Well, since this is not the case, there is quite a bit of wasted time.
Typical code looks like this:
SetInt %000010000, %000010000 ' Pin 3, High Main: goto Main: Interrupt: Pause 100 If Pin3 = 1 then Interrupt <do your stuff > SetInt %000010000, %000010000 Return
Basically, you poll for a high pin. Once you detect it, you repoll for the low pin. Once that happens you can do what you want and rearm the interrupt for the high state.
This cuts your processing slice in half. I suppose you could fix this by clocking through a flip flop and resetting the flip flop to make it a monostable circuit. Here is a page that does this by setting the value of the input to 0. Not so bad…it would extend the processing cpu slice…but…you know…we seem to be ok so far, so lets deal with that later if needed.
If you look at the code, you will notice that I take advantage of the space time by only reading the pot (READADC) if I know that Im idle for that 1/2 cycle.
This revision has 5 modes. Here are some samples. These were taking by plugging the circuit directly to my soundcard. There is a TINY bit of postprocessing. I added a lowpass because the spaces were a little cracky. When the DTMF cuts off…it cuts off right away. I never heard it going into a guitar amp or through some effects. Ill add a lpf to the next revision.
Mode 1
Mode 1 cycles through all 16 states upwards.
Boomp3.com
Mode 2
Same as mode 1 but downwards
mode 3
This is the one that sounds most like somebody dialing digits. There is a zero shifted to the register between random digits.
Boomp3.com
Another sample
again…one more
Mode 4
This is a random digit pattern, but without the spaces. Its kinda monotonous really.
Mode 5
This adds a few more spaces into the equation. Kinda interesting.
The actual circuit is not too much different.
Note that a Low pass filter at around 7 to 10khz is a REALLY good idea. DTMF is really two sine waves at fixed frequencies, and all telephone audio is bandpassed…so adding anything more complicated would be a waste of time.
I did try and make a pcb for this circuit. Surprisingly it worked the FIRST time.
I have the pdf of the circuit too…but…you know, it needs some work still. If you interested, email me at robstave aht yah hoo dot com (stupid robots) to get the pdf. Or, if your interested in any kind of collaboration/copyleft let me know. I still might consider making a kit that I can sell out of the thing, but it still has a way to go before Im completely happy with it.
Back to PICAXE projects