Revive a shorted v4.0 HM

[Steve_M]

I managed to short out my old v4.0 HM a while ago but didn't realize it wouldn't run anymore after doing so. In doing so, it also took out an old raspberry pi. The short happened when I had it out of the case and accidentally bumped the metal braid from a probe which ended up touching the exposed pins on the modded cat5 jack I had on it.

I've now got my first gen 256MB model B hooked up to it, but there's no communication between RasPi and the Atmega. I've swapped the atmega and it's not any better.

Symptoms:

LCD display only shows blocks on top row
Green and Yellow LEDs remain on
Unable to flash the atmega chip

The voltages listed in the image below have all been verified.

Flashing the atmega shows:

Stopping LinkMeter OK

LinkMeter platform is BCM2708
AVR fuses ERROR

Starting LinkMeter OK

What's next on the check list?

 

[RalphTrimble]

Since you swapped the ATMega I guess we can count that out, and you've measured +5v and+3.3v so I guess the PS is working... If you look at the schematic there are only a few components between the rPi header and the ATMega, the shift register is connected to a few of the same ATMega pins as the rPi header, and a resistor and cap or two.... So maybe look at the shift register? That is if you are sure the ATMega(s) you are using is ready to be programmed? I recall something about resetting the fuses on the ATMega, maybe you need to look into that?

 

[Steve_M]

I swapped the Atmega into my test/franken/TC botched v4.2.4 unit and was able to run avrupdate on it

# avrupdate

Stopping LinkMeter OK

LinkMeter platform is BCM2708
AVR fuses 62d907 blank ATmega chip detected, flashing default
hmdude: compiled on Oct 3 2014 at 09:58:46
Using port: /dev/spidev0.0
Loading ihex file: "/lib/firmware/hm.hex" (23712 bytes)
Device signature: ATmega328P
Low: 0x62 High: 0xd9 Ext: 0x07 Lock: 0x3f
Chip erased
100% |##################################################| 23712 (9.2s)
Verifying...
100% |##################################################| 23712 (8.2s)
Writing low fuse: 0xff
Writing high fuse: 0xd7
Writing extended fuse: 0x05

Starting LinkMeter OK

# avrupdate
Stopping LinkMeter OK

LinkMeter platform is BCM2708
AVR fuses ffd705 OK

Starting LinkMeter OK

I then swap the chip back into the HM 4.0 board and I still get:

# avrupdate
Stopping LinkMeter OK

LinkMeter platform is BCM2708
AVR fuses ERROR

Starting LinkMeter OK

 

[Steve_M]

I can actually boot the HM v4.0 unit with flashed chip from the v4.2.4 unit. The Web UI comes up and I can even get readings from all 4 probe jacks.

With regards to the LCD and Shift Register, here are the measured voltages I just took:

LCD Pins:

1) 0.0
2) 5
3) 0.6
4) 0.07
5) 0.00
6) 0.06
7) 4.8v
8) 4.8v
9) 4.8v
10) 4.8v
11) 0.04
12) 0.02
13) 0.01
14) 0.01

Shift Register Pins:

Top Row:
16) 0.08
15) 0.07
14) 0.06
13) 0.04
12) 0.03
11) 0.01
10) 0.01
9) 0.01

Bottom Row:
1) 3.5
2) 0.08
3) 2.3
4) 0.01
5) 0.01
6) 3.3
7) 3.46
8) 0.0

According to Bryan from another thread, I should be seeing:

LCD
1 - GND
2 - 5V
3 - ~1V (adjust the pot until this reads about 1V)
4 - Should be steady but bounce around once a second
5 - GND
6 - behaves like 4
7,8,9,10 - no connection
11,12,13,14 - behaves like 4
15,16 - don't care

If those are bad then check at the shift register starting with the top row of 8 pins (looking at the solder side of the board as in your photo there) from the left.
16 - 0V (LED)
15,14,13,12,11,10 - behaves like LCD4
9 - GND

Bottom row starting at left
1 - 3.3V
2 - 0V (LED)
3,4,5,6 - behaves like LCD4
7 - 3.3V
8 - no connection