Post Reply 
[wp34s] Permanently damaged machines when flashing
07-24-2015, 10:49 PM (This post was last modified: 10-16-2015 10:53 AM by emece67.)
Post: #1
[wp34s] Permanently damaged machines when flashing
Today I've permanently damaged another (this is the second one) wp34s during flashing.

The symptoms are: during flashing (MySamba progress bar is moving) the machine hangs in some way, MySamba reports that it does not receive the ACK from the machine and, from now on, the machine is absolutely unresponsive.

I've tried even to flash it thru the JTAG port to no avail, the machine is, definitely, dead.

This is the second time it happens to me.

During the post-mortem analysis I've found that:
  • Both machines have their cells at 2.9 V
  • My USB to RS232 converter (a 3V3 one) sends levels up to 3.4 V thru the TXD line, this is 0.5 V above the battery voltage
  • My DIY cable does not power the machine, during flashing the machine is powered from its batteries

and the SAM7L128 processor datasheet (that's the processor inside the hp20/30B) claims that the input voltage on any input cannot be greater than 0.3 V above the supply voltage.

Thus I suspect the combination of a slightly greater than expected output voltage on the RS232 port and the not so fresh batteries can be the culprit.

I cannot be sure this was the root cause, but I advise to use fresh batteries when flashing wp34s machines.

The HP cable, does power the machine besides the batteries?

Regards.

C├ęsar - Information must flow.
Find all posts by this user
Quote this message in a reply
07-25-2015, 06:47 AM (This post was last modified: 07-25-2015 07:05 PM by Perre52.)
Post: #2
RE: [wp34s] Permanently damaged machines when flashing
The maximum supply voltage Vd of the Atmel processor is 3.6V.
However the input data pins can have a Vd + 0.3 V. That's 3.9V without any damage to the processor.
I flashed mine yesterday with a BV101-4 USB to serial interface cable.
The max voltage on the Tx pin of this interface is 3.85V.

I did fry a calculator but not by programming the calculator. I used a power supply as a voltage source for my calculator during programming. And the voltage on this supply was 9.0V ! I noticed the fault but it was to late.

Try measuring the current when the calculator is on. It should be around 5 mA.
If you measure this value chances are great that your calculator isn't dead but just "sleeping."
This is also the reason that I use a power supply instead of the batteries. I can measure the current and know if my calculator is on (5mA) or off (0mA) during programming.

The HP programming cable uses a ADM3202 IC from Analog Devices.
This device draws its current from the calculator batteries.
You can find more about this cable here: http://hhuc.us/2008/Interfacing%20to%20t...b%20V3.pdf
Find all posts by this user
Quote this message in a reply
07-25-2015, 07:29 AM
Post: #3
RE: [wp34s] Permanently damaged machines when flashing
It is also possible that the input protection diodes went into "Latch Up" when the TX voltage went .5V above
the power supply voltage. To clear the "Latch Up" state remove the batteries and short the 3V and GND terminals of the battery for a few seconds to drain any power supply capacitors. Then reinstall the batteries to see if the calculator wakes up again. If you had left the calculator in the "Latch Up" state for any length of time it is likely that the batteries suffered a significant drain, and it might be wise to start with a fresh set of batteries after the battery-removal / Discharge procedure. Good Luck, Barry
Find all posts by this user
Quote this message in a reply
07-26-2015, 01:29 AM
Post: #4
RE: [wp34s] Permanently damaged machines when flashing
So, I wasn't going crazy then...

I encountered mechanical failure in one of my units ages ago, but I beat my head against the wall for too long believing it to be a software or computer issue. I donated the defunct unit to a more tech-savvy member, and he confirmed it was dead, but did not speculate why.

I have not flashed mine for ages, as I have been away from wp34s usage for over a year, and still have 3.2/rev 3469. If I proceed to upgrading the units--I have three--I will heed your warnings.

Les

PS. What is interesting is that I flash my units the older way, with a cable that was once provided by Gene. I don't have the USB upgrade. Does that make a difference?
Find all posts by this user
Quote this message in a reply
07-26-2015, 05:06 AM
Post: #5
RE: [wp34s] Permanently damaged machines when flashing
(07-25-2015 07:29 AM)BarryMead Wrote:  It is also possible that the input protection diodes went into "Latch Up" when the TX voltage went .5V above
the power supply voltage.
This is something I didn't know. Learned some new things today.
That's problably the reason why Atmel gives the max voltage at Vd + .3V.
If You go higher You could have "Latch up"
I assume Vd is the voltage of the battery or the power supply. So if we have a 3.0V power supply the max input voltage would be 3.3V. If we have 3.6V power supply the max input voltage would be 3.9V.
Am I correct or did I miss something ?
Michel
Find all posts by this user
Quote this message in a reply
10-15-2015, 04:54 PM
Post: #6
RE: [wp34s] Permanently damaged machines when flashing
This is why I put 1k resistors in series with the TX / RX when I flashed mine (Link). Can't be too careful when it comes to multiply powered devices intercommunicating.
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)