A little DTMF help

[Travin77]

If you hold the numbers depressed too long, the program puts the same number in the next array space. So, I am going to have to revisit the decode section. It is interesting that the dtmf chip is disabled at the time of data entry of the decoded number into the array. Also, only one verbal response plays.

Travin

No need to post a reply to myself. I debounced the software with a Pause 500 just before "until c > 7", and it worked like a champ. I tried several different pauses during tone entry and never had another repeat. So, thinking I had it solved, I placed the corrected code into the original code and whalla, the same problem. So, I guess the if then statements on the password section are bunk. It still won't play the error message if the entry is wrong, it won't go to the menu message if the entry is right, and it still starts over after pressing a random number of buttons on the phone after all eight digits of the password have been input.

[BigWumpus]

You wrote:

reset = 0
PAUSEUS 30 ' Hold GT for > 18mS
reset = 1


Remember, you wrote a 30uS pause, but you need a 18mS pause !



PS: How to write your Programm in 40 lines ?

[Travin77]

Thanks for the catch. I guess I just wasn't careful. I know my program is incredibally long for something so easy, I just figured since the pic couldn't do something as simple as play the damn error message or menu message, why should I trust it to jump to a subroutine. I figure my next step is to start removing chips from the equation. I am going to hook up the ol LCD screen and use the lcdout command instead of the audio. That way I can try to se if what is going on visually.

Travin

[BigWumpus]

You wrote:

DT0 VAR PORTB.4 'pin 2 (D1)
DT1 VAR PORTB.5 'pin 1 (D2)

out of the manual of the chip, I think you have mixed up Pin1 and Pin2.

[Travin77]

Those pins are correct. I used someone elses code off of this forum and never updated the note. There is now no problem with decoding or the tone entry into the array as far as I can tell. I tried about twenty different combonations leaving the keys pressed for different amounts of time and every time it played back without any problems. The only change I made was the pause 500 before "until c > 7", which solved the entry problem Now the issue seems to be the if password[0] <> "1" then error. I am not sure why but it can't manange to do something that simple.

[BigWumpus]

maybe:

@ device pic16F876A, hs_osc, wdt_on, lvp_off, protect_off

include "MODEDEFS.BAS"
define OSC 20
adcon1=7
trisc = %00000000

'_________pic to mc145436a assignments______________________
dtmf_ready VAR PORTA.0 'pin 12 (DV)
select_dtmf VAR PORTA.1 'pin 3 (Enable)
reset VAR PORTA.2 'pin 5 (GT)
DT0 VAR PORTB.4 'pin 2 (D1) ??
DT1 VAR PORTB.5 'pin 1 (D2) ??
DT2 VAR PORTB.6 'pin 14 (D4)
DT3 VAR PORTB.3 'pin 13 (D8)
ce var portA.3 'pin from isd chip

'____________________variables____________________ ______
dtmf VAR byte ' Stores most recent DTMF digit
dtmf1 var byte
password var byte [8]
codeword var byte [8]
codeword1 var byte [8]
c var byte
e var byte

'___________________Initial Conditions__________________
start:
pause 100:high ce:high reset:Low select_dtmf:e = 0
repeat
until dtmf_ready
portc = %00000000:ce = 0: pause 20:ce = 1: pause 2000:c = 0
repeat
Repeat
until dtmf_ready 'waits for number to be depressed
select_dtmf = 1 ' Enable the data output from the MT8870
dtmf = 0:dtmf.0=DT0:dtmf.1=DT1:dtmf.2=DT2:dtmf.3=DT3
LookUp dtmf,["_1234567890*#"],dtmfassword[c]=dtmf
Repeat : until dtmf_ready=0 ' Loop here until DTMF signal stops
reset = 0: PAUSEUS 30:Reset=1 ' Hold GT for > 18mS
select_dtmf = 0 ' Disable the MT8870 data output
gosub Output_DTMF:c = c + 1
until c > 7
c=0:repeat:dtmf=password[c]:gosub output_DTMF: Pause 1000:c=c+1:until c>7
goto start
end

Output_DTMF:
Lookup dtmf-"0",[%10101001,%10010111,%10011001,%10011011,%10011101, %10011111,%10100001,%10100011,%10100101,%10100111],PortC
pause 1:ce = 0: pause 1:ce = 1
return

[Travin77]

Your code compiles with errors. I don't quite understand it enough to change either. it doesn't like you lookup table, either of them. Also, it apprears you may have an extra repeat so I removed it. Thanks for the code, I am going to fiddle with it for a while.

Travin

I see know what you were doing. My compiler says it is missing an end block on the repeat.

[BigWumpus]

It compiles fine with the MP (instead of MPASM). (228 words in a 16F876A .. snif)
You must put some lines together.... they are wrapped.

Maybe it is interesting to check the OEM-signal of the ISD-chip to wait for the end of a message, before sending the next...

The "pause 20" around the ce=x can be shortened to "pauseus 10".

Why are you using 20Mhz ? It is VERY fast for this slow devices !
at 4MHz you can throw away some pauses....

[Travin77]

I did that. I had a routine that was interupt based on the EOM pin during which the pic would do nothing until the EOM bit was pulsed low. I found that I really didn't need to do that. I didn;t gain anything from it. My problem isn't the decode, its the if then statements. The pic just won't do them. It won't go on in the program, it won't play the error message, it just wants to loop back to the beginning. Do you have another way of doing the password section? As to why I am using 20 mhz, I don't know, thats just what I was using. I will try a 4 mhz osc. Now I guess my question for that is it still as OSC_hs or just external?

Travin

[BigWumpus]

Your code looked good:

if password[0]<>"1" Then goto error
if password[1]<>"2" Then goto error
...
if password[7]<>"8" Then goto error
menu:

error:



You can extend the Output_DTMF-routine to wait for the EOM-bit in the beginning.
And you can add some code in order to use this routine to play the other menu-messages, too.