PIC16F690 EEPROM not retaining data

[xnihilo]

Hi,

I've got a very strange behaviour.

I write data to the internal eeprom with statement such as:



write healed_,healed
READ healed_,tempvar
IF (tempvar != healed) then
SOUND speaker,[100,200]
LCDOUT $fe,1,"writeRoR"
PAUSE 2000
ENDIF


and get no message, which means the write could be verified.
Then, when I turn off the circuit and turn on again, I display the values in the EEPROM and they are those I set at the time I flashed the PIC:



DATA @10,0,0 'first hit
DATA @12,0,0
DATA @14,0,0
DATA @16,0,0
DATA @18,0,0
DATA @20,0,0


It happens as if the EEPROM does not keep the data I WRITE.

Would anyone have an idea why??

[mister_e]

Healed, Healed_ AND TempVar are Bytes?

[mackrackit]

DATA @10,0,0 'first hit
DATA @12,0,0
DATA @14,0,0
DATA @16,0,0
DATA @18,0,0
DATA @20,0,0

Is the above in the code you are running.

I display the values in the EEPROM and they are those I set at the time I flashed the PIC:

If so, every time the PIC starts those values are written. If this is not the case then...

[mister_e]

Nope.. DATA is performed at programming time only. :o

[xnihilo]

Healed, Healed_ AND TempVar are Bytes?

healed is a BYTE var
healed_ is a CON pointing to the EEPROM location where I backup value of healed variabme
tempvar is a BYTE var

[xnihilo]

DATA @10,0,0 'first hit
DATA @12,0,0
DATA @14,0,0
DATA @16,0,0
DATA @18,0,0
DATA @20,0,0

Is the above in the code you are running.


If so, every time the PIC starts those values are written. If this is not the case then...

Really?
I thought that DATA @... was only used at the time the PIC is programmed only.
Then How should I set initial values at programming time???

[xnihilo]

Nope.. DATA is performed at programming time only. :o

Ah, that's what I thought, otherwise DATA would make no sense.
There is another instruction to alter EEPROM value: it is READ and WRITE

[Archangel]

Hello xnihilo,
what does your code which makes healed_ into an address location look like ?

[xnihilo]

Thank you for your help but this time I could solve the problem:

I was using a routine that displays at the begining of the program the contents of the EEPROM at location 10 to 21 using an index var (hitnr).
Then later in the program I was writing to EEPROM using the index var BUT i forgot to reset the index var to its initial value after using it at the begining of my program so I was thinking I was writing to location 10 while the index var was already at 12. So when restarting the pic, I was displaying values of locations 10 and 11, that were still at 0 so I thought EEPROM write failed somehow. I was at the wrong memory offset...
Stupid error, I should get some sleep :)

Have a nice day.

[mackrackit]

Stupid error, I should get some sleep

ME TOO. I was close.