| Author |
 Topic  |
|
|
tekky
Australia
26 Posts |
 Posted - 05/29/2019 : 07:25:18
|
Can anyone help me please.
I repair circuit boards for a living ... and in the past I read the buffer and data off an ATmega16-16AU while it was still in-situ on the board.
I then re-programmed a corrupted ATmega16-16AU on a second board successfully and all worked fine.
Sorry for shouting ... but I CAN NOT REMEMBER IF I POWERED UP THE BOARD when I did this.
Last week I had another board which mostly worked ... but seemed to be a bit odd in its timing ... so I went to reprogram it.
I couldn't remember if I needed to power up the board as well ... but the programmer was prompting me "Power on Target Device is Required" ... so I powered it up.
Instantly smoke and flames shot out of the ATmega16 and it was cooked!
Sadly, my programmer was also toast ... so spent the bucks and had another delivered to Australia.
After updating drivers etc on my new GQ-4x4 ... I still can not make it read even a known good one. It's still prompting me to power it up ... but I'm not game to burn up my new one!!!
Most times when I attempt to "ID" the chip ... it comes up unknown ... but several times it has read the ID code correctly ... but has not read the code from the micro.
I had written very good notes as to how to do this whole procedure ... EXCEPT whether to power it up or not!!!
And I had the purpose built cable filed with the notes ... so no mistake there.
I've attached notes as to how I have connected it up.
Thank you for any help.
Download Attachment: 
32051 |
|
| Reply #1
anniel
2572 Posts |
Posted - 05/29/2019 : 08:22:32
|
quote:
Originally posted by tekky
Can anyone help me please.
I repair circuit boards for a living ... and in the past I read the buffer and data off an ATmega16-16AU while it was still in-situ on the board.
I then re-programmed a corrupted ATmega16-16AU on a second board successfully and all worked fine.
Sorry for shouting ... but I CAN NOT REMEMBER IF I POWERED UP THE BOARD when I did this.
Last week I had another board which mostly worked ... but seemed to be a bit odd in its timing ... so I went to reprogram it.
I couldn't remember if I needed to power up the board as well ... but the programmer was prompting me "Power on Target Device is Required" ... so I powered it up.
Instantly smoke and flames shot out of the ATmega16 and it was cooked!
Sadly, my programmer was also toast ... so spent the bucks and had another delivered to Australia.
After updating drivers etc on my new GQ-4x4 ... I still can not make it read even a known good one. It's still prompting me to power it up ... but I'm not game to burn up my new one!!!
Most times when I attempt to "ID" the chip ... it comes up unknown ... but several times it has read the ID code correctly ... but has not read the code from the micro.
I had written very good notes as to how to do this whole procedure ... EXCEPT whether to power it up or not!!!
And I had the purpose built cable filed with the notes ... so no mistake there.
I've attached notes as to how I have connected it up.
Thank you for any help.
Download Attachment:
32051
Yes, the target needs power. What did toast in the programmer? |
 |
|
| Reply #2
tekky
Australia
26 Posts |
Posted - 05/29/2019 : 15:40:01
|
Thanks for the reply anniel.
The 2 obvious parts inside the programmer that failed were U6, a TQFP44 IC with its numbers ground off ... and T20, a 3904??
Is there any chance I can identify them and repair my original GQ-4X
Regarding this board needing power ... the whole time I've been blaming myself for powering it up when I shouldn't ... but now I'm left afraid that my new one will go up in smoke if I try again.
Is it possible that I did it in the wrong order?
I had selected ATmega16 ISP on the programmer and had connected it up to the target board ... and because I was getting no result and the programmer was prompting me ... I switched on the power to the target board.
The failure was instant and spectacular!
Is it possible that I should have selected ATmega16 ISP on the programmer ...
Powered up the target board ... AND THEN connected the cable from the programmer to the target board.
The reason I say this is because ... after scouring through all my notes from the first time I reprogrammed one of these boards ... I came across a comment I had written ...
"Note ... when plug programmer lead into (target board) ... the "Switch" led goes out and "Transmit" led comes on".
The only way for those leds to be on ... is if the target board IS powered up.
|
|
|
| Reply #3
anniel
2572 Posts |
Posted - 05/29/2019 : 16:21:16
|
quote:
Originally posted by tekky
Thanks for the reply anniel.
The 2 obvious parts inside the programmer that failed were U6, a TQFP44 IC with its numbers ground off ... and T20, a 3904??
Is there any chance I can identify them and repair my original GQ-4X
Regarding this board needing power ... the whole time I've been blaming myself for powering it up when I shouldn't ... but now I'm left afraid that my new one will go up in smoke if I try again.
Is it possible that I did it in the wrong order?
I had selected ATmega16 ISP on the programmer and had connected it up to the target board ... and because I was getting no result and the programmer was prompting me ... I switched on the power to the target board.
The failure was instant and spectacular!
Is it possible that I should have selected ATmega16 ISP on the programmer ...
Powered up the target board ... AND THEN connected the cable from the programmer to the target board.
The reason I say this is because ... after scouring through all my notes from the first time I reprogrammed one of these boards ... I came across a comment I had written ...
"Note ... when plug programmer lead into (target board) ... the "Switch" led goes out and "Transmit" led comes on".
The only way for those leds to be on ... is if the target board IS powered up.
What was the power source of the board? Any chance the polarity was reverse? |
|
|
| Reply #4
tekky
Australia
26 Posts |
Posted - 05/29/2019 : 17:36:33
|
No, there was no chance of that. The board has a 24v ac supply and had been working ok (other than a small timing error).
I had just checked the Vcc to the ATmega16 and it was exact at 5.0v
The more I think about it ... I must have had the original unit powered up when I retrieved the program from the micro (several years ago).
Maybe this failure occurred because I had the GQ-4X4 plugged in when I powered up the target board ...
Maybe I was supposed to have it already powered up ... then when the GQ-4X4 is plugged in ... the Reset pin immediately pulls the micro into Reset ... and then can be read or written to safely ..
Or maybe, it was just an unlucky coincidence and I am wasting time worrying about blowing up my new programmer!!! :) |
|
|
| Reply #5
anniel
2572 Posts |
Posted - 05/30/2019 : 03:02:14
|
quote:
Originally posted by tekky
No, there was no chance of that. The board has a 24v ac supply and had been working ok (other than a small timing error).
I had just checked the Vcc to the ATmega16 and it was exact at 5.0v
The more I think about it ... I must have had the original unit powered up when I retrieved the program from the micro (several years ago).
Maybe this failure occurred because I had the GQ-4X4 plugged in when I powered up the target board ...
Maybe I was supposed to have it already powered up ... then when the GQ-4X4 is plugged in ... the Reset pin immediately pulls the micro into Reset ... and then can be read or written to safely ..
Or maybe, it was just an unlucky coincidence and I am wasting time worrying about blowing up my new programmer!!! :)
Any chances your 10 pins header was reversed? That would put VCC on your RESET and GND on your MISO etc. That should do nice fireworks. |
|
|
| Reply #6
anniel
2572 Posts |
Posted - 05/30/2019 : 03:07:52
|
quote:
Originally posted by tekky
Thanks for the reply anniel.
The 2 obvious parts inside the programmer that failed were U6, a TQFP44 IC with its numbers ground off ... and T20, a 3904??
That transistor is wired up to what? |
|
|
| Reply #7
tekky
Australia
26 Posts |
Posted - 05/30/2019 : 04:28:33
|
quote:
Any chances your 10 pins header was reversed? That would put VCC on your RESET and GND on your MISO etc. That should do nice fireworks.
Haha, it would indeed ... but no ... because it failed so spectacularly ... I went back over what I had done super carefully and the plug was definitely in the correct way. In this case 3 wires on the bottom and 2 wires on the top row.
In fact, Vcc is not even connected on this plug ... if I had powered up the ATmega16, it was through the target boards power supply and not the GQ-4X lead.
Should I have applied Vcc via the programming plug ... that would try and fire up the micro ... but I suppose the Reset line would help keep the micro under control and not allow it to go into its normal operating mode.
PLEASE NOTE ... that is definitely NOT what I did when I successfully read the micro and re-programmed a couple of these boards in the past!!! |
|
|
| Reply #8
tekky
Australia
26 Posts |
Posted - 05/30/2019 : 04:32:14
|
quote:
That transistor is wired up to what?
I don't know ... once I saw the blown IC with its markings ground off ... I didn't look any further. I imagine it may be a micro ... so even if I did know its number ... I wouldn't have the program for it anyway ... unless I could buy it via MCUMall.
Do they sell spare components for their boards? |
|
|
| Reply #9
anniel
2572 Posts |
Posted - 05/30/2019 : 06:12:49
|
quote:
Originally posted by tekky
quote:
Any chances your 10 pins header was reversed? That would put VCC on your RESET and GND on your MISO etc. That should do nice fireworks.
Haha, it would indeed ... but no ... because it failed so spectacularly ... I went back over what I had done super carefully and the plug was definitely in the correct way. In this case 3 wires on the bottom and 2 wires on the top row.
In fact, Vcc is not even connected on this plug ... if I had powered up the ATmega16, it was through the target boards power supply and not the GQ-4X lead.
Should I have applied Vcc via the programming plug ... that would try and fire up the micro ... but I suppose the Reset line would help keep the micro under control and not allow it to go into its normal operating mode.
PLEASE NOTE ... that is definitely NOT what I did when I successfully read the micro and re-programmed a couple of these boards in the past!!!
No wires overheated in the harness? |
|
|
| Reply #10
anniel
2572 Posts |
Posted - 05/30/2019 : 06:12:58
|
quote:
Originally posted by tekky
quote:
That transistor is wired up to what?
I don't know ... once I saw the blown IC with its markings ground off ... I didn't look any further. I imagine it may be a micro ... so even if I did know its number ... I wouldn't have the program for it anyway ... unless I could buy it via MCUMall.
Do they sell spare components for their boards?
Knowing the function of this transistor may help understand what went wrong.
Appart from the ZIF socket I don't think they sell parts but you could try by contacting them directly. |
|
|
| Reply #11
tekky
Australia
26 Posts |
Posted - 05/30/2019 : 07:09:48
|
No, no wires overheated ... just the flash and smoke ... and coz I'm experienced elect tech (haha) instantly flicked the switch off (still had my finger on it).
I'll have a closer look at the burnt transistor in the programmer in the morning when I go back out to work.
I didn't really expect to be able to buy spare ICs for the programmer ... and at the new price of USD $99 it is probably not worth chasing ... though by the time we convert to AUD, then add freight ... it still cost me $225 or so to get it here.
I use it all the time ... sometimes daily ... then sometimes only weekly ... but it scared me not having access to it for those few days. That's why I don't wanna risk blowing up the new one. Gotta admit I was amazed that it only took about 4 days to get to Australia from Canada ... pretty impressive. |
|
|
| Reply #12
anniel
2572 Posts |
Posted - 05/30/2019 : 07:22:41
|
quote:
Originally posted by tekky
No, no wires overheated ... just the flash and smoke ... and coz I'm experienced elect tech (haha) instantly flicked the switch off (still had my finger on it).
I'll have a closer look at the burnt transistor in the programmer in the morning when I go back out to work.
I didn't really expect to be able to buy spare ICs for the programmer ... and at the new price of USD $99 it is probably not worth chasing ... though by the time we convert to AUD, then add freight ... it still cost me $225 or so to get it here.
I use it all the time ... sometimes daily ... then sometimes only weekly ... but it scared me not having access to it for those few days. That's why I don't wanna risk blowing up the new one. Gotta admit I was amazed that it only took about 4 days to get to Australia from Canada ... pretty impressive.
Once the original factory smoke is escaping it's too late for the switch.  Maybe you can get a can of generic smoke refill on FleaBay to get it back working. |
|
|
| |
Topic |
|
ATmega16 flamed when went to program
