Digital PI-1

[Peter Morf]

I made an assembler program for a PIC16F874 to control the PI-1's timing. It has two hex rotary switches to set timing. A LED indicates if the setting overrides the set frequency (when timings are too long to fit in the set period), if so... the second sample delay is set back to 120us automatically. I even started on a program for an 4x20 LCD whith a menu system, but that would require 2 PICs, one for timings and one for driving the LCD, so I'm starting with the hex switches.
I am also working on the PCB that includes the PIC.
Is this of interest??
I havent tested it yet... but when I'm finnished building it, I thought I should post it on this forum.
Regards
Peter Morf
Sweden


; SPECS:
;
; Frequency: 800-2300 Hz in 15 steps of 100Hz
; Pulse width: 15-75us in 15 steps of 4us
; Sample pulse width: 15-75us in 15 steps of 4us
; Sample delay: 4-79us in 15 steps of 5us (from the end of the TX pulse)
; Sec. sample delay: 0-256us in 15 steps of 16us (from the end of the 1:st sample)
;
; Auto correction and indication when frequency overrides
; __15-75__________________________________________
; Pulse*: |___|
; _________________ 15-75 _________________________
; 1:st Sample*: < 4-79us >|_____|
; _______________________________________ 15-75 ___
; 2:nd Sample*: |_____|
; *
;************************************************* **********
; *
; _______________ *
; |____ \_____/ | *
; -|MCLR RB7||RA0/AN0 RB6||RA1/AN1 RB5||RA2/AN2 RB4||RA3/AN3 RB3|

[Georg Aster]

Re: Digital PI-1

For driving LCD you can use PIC Tutorial Series by Nigel Goodwin http://www.winpicprog.co.uk/



PIC Tutorial Series

[FJ]

Re: Digital PI-1

Hi,
No, it doesn't require 2 pics. I've done it with one. I'm using a 16F877 at 20Mhz. The key is to use interrupts and not time delay loops. I use the PIC to generate the timing for pulsing the coil and sample gates, PWM to generate the tone, Pushbuttons to enter a menu to change parameters, and an LCD to show the parameters on line 1 and a bar graph to show relative signal strength on line 2. This is using one PIC.

My boat towable version uses a second PIC but for GPS communications. The metal detector PIC sends "HIT" data via the USART to the GPS PIC's USART and the GPS PIC outputs the hit strength, time, date, and GPS coordinates to an RS-232 port to log to a computer.
Good luck,
FJ

[Georg Aster]

Re: Digital PI-1

I've done it with 3PICs 2xPIC16F676 and 1PIC16F628 This eliminate all the parts from IC8a to IC9.

[sevzirfo]

Re: Digital PI-1




I made an assembler program for a AT90s2313-10PI to control the PI-1's timing.

[Rodis]

Re: Digital PI-1

There are different ways of using a microcontroller in a PI that have different levels of difficulty.

1. The easiest would be to just have the microcontroller take over the timing functions. The micro does not have much work to do and the program is small and should be easy to get running. This would eliminate the timing chips from the design.

2. The next level of difficulty would be to have the microcontroller sample the signal at the output of the integrator (IC8a). If you use 2 ADC channels you can sample from both sides of C19 and get both motion and non-motion signals. The microcontroller does not have to be very fast because sampling at about 100 Hz would be enough. Then the microcontroller could add the threshold and chop the signal to make the audio output. The microcontroller would have to do a lot more work than the first case and there is more software to write, but it still does not have to do much math. This would eliminate all the parts from IC8b to IC9.

3. The next level is to sample before the integrator eliminating IC8a. This is much more work because now the sampling has to be done at the pulse rate or higher. And now there is a lot of math to do to filter the samples. An advantage of doing this is that you can take extra samples at different delay times to try to identify the target.

I am working on this third case but case 2 is a lot less work for only a small difference in the amount of analog parts. And I am not certain that I will be able to get the same performance possible from case 2. I hope that I can, but I will not know for sure until after I do all the work.
-Robert

I will also work on case 3. Case 1 has been done in some of the other projects (GoldPic & Stuart), and case 2 is how Tesoro does the SandShark. If case 3 performs poorly, then back off to 2.
- Carl

[Peter Morf]

Re: Digital PI-1

Hi!
I think I'll start with case 1 and make it run with an LCD in the same PIC (as LF wrote).
It would be fun to increse the difficult level =) but are the PIC's ADC fast and accurate enough for this purpouse (10-bit)?