Hello to all of you!
Some time ago I saw the ATTiny40 at an incredibly low price, about 0.18$ each unit. Now, I got a soic-20 breakout board to test them and play with them a bit. But I have found that this has been worse than I imagined. The interface to program it is the TPI, which I have did some research and I haven't seen a lot about it. Up to this time, the most valuable resource I found is an Arduino sketch that simulates this TPI interface. But, as you can see entering to that web site, is that the sketch was done to program a ATTiny10. It is adapted to program an ATTiny40, but I have found even less information about this chip and the TPI.
Now, I'm trying to debug what is the problem it does not work. I've dived on the datasheet of ATTiny40, trying to fix any error between the "documented protocol" and the protocol that is implemented on the arduino sketch. A big part of it is working, I am able to do a dump of all the ATTiny40 memory (I guess that is called the NVM). The programming sequence documented commands to do a "chip erase" following of the actual memory writing. But I can't even get working this chip erase instruction. I haven't modified the arduino sketch from the previous webpage, because the protocol seems to be exactly the same.
I have two chips and, when doing the memory dump of the first, I see this:
registers, SRAM
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0010: 00 00 00 00 00 00 00 00 00 00 00 10 00 00 00 E0
0020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0030: 00 00 00 00 00 00 03 00 00 8C 00 03 00 00 00 00
0040: E9 58 5A 9C 1C A8 EE 53 1F 3B 16 15 A7 99 0B 96
0050: A3 7B B6 01 57 46 CF 16 EE 56 CF 9D 5E CC D6 FD
NVM lock
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F00: FF FF
configuration
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F40: FF FF
calibration
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F80: 8C 02
device ID
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3FC0: 1E 92 0E FF
program
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
4000: FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
4010: FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
4020: FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
[...]A positive thing is that the NVM lock seems to be all 1's, so according to the docummentation, it is writable. Doing a chip erase does not erase the memory, and the program memory is always at 1's level (all that 0xFF shown there)
But other chip I have shows a completly different behavior: The LSB bits on the NVM lock are 0's, so it shows a "programmed" state. Those bits should also be eraseable with a chip erase command, but the chip erase command does not changes anything. Also, the program memory is filled with 0's instead of 1's
registers, SRAM
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0010: 00 00 00 00 00 00 00 00 00 00 00 10 00 00 00 CF
0020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0030: 00 00 00 00 00 00 03 00 00 77 00 03 00 00 00 00
0040: 4E D6 0E F1 DF EF 1F 67 BA DF 85 BA C7 F9 FD FB
0050: 53 CC 6C BF AB BB 5F C9 74 FD 3B 6C 3C F4 F2 F9
NVM lock
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F00: FC FC
configuration
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F40: FF FF
calibration
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3F80: 77 02
device ID
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
3FC0: 1E 92 0E FF
program
+0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +A +B +C +D +E +F
4000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
4020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[...]So, now my question is: Does anyone has had this issue with the ATtiny40 (or other ATtiny similar family as Attiny5/9/10/20). I don't know what is "normal" to be found on a new factory chip, if it should be filled with 0's or 1's. Am I missing something?
