openrcforums

Updated version of the test code at viewtopic.php?f=10&t=465&p=5861#p5861.
This reports which key(s) is/are seend pressed.

@kilrah - copied the relevant code, not so easy to get the actual menu screen, the menu driver is not running at this point in the code.

Mike.

MikeB wrote:Updated version of the test code at viewtopic.php?f=10&t=465&p=5861#p5861.
This reports which key(s) is/are seend pressed.

Mike.

OK, now it says its the Key 1 (left). The multimeter shows the switch (frog) is working correct.

OK the LEFT key is on pin 16, the last one on the right at the bottom. Check the voltage all along the mauve track, the circular pad and the pin of the CPU. With the LEFT button not pressed, you should get around 5 volts, and with the LEFT button pressed you should get around 0 volts. The pin of the CPU is the most important point to measure.

Mike.

Tricky problem.
After diggin' a little bit with the multimeter: Pin 16 is 0v, when TX is off, it's short with GND, also pin 16 is short with the upper pins on the left switch. Between the switch points there is no short. The 200 Ohm is ok the capacitor seems to be OK, no short.
Turning the TX on the pin 16 is 0.01mv. ??

That's a good idea... Make a "diagnostics" version.
I'll get cracking!

I think it is obvious that there is a short to ground..the question is where..
.

Here you go guys:
1) For checking keys:
http://code.google.com/p/er9x/downloads ... eydiag.hex

2) For Checking the analog inputs:
http://code.google.com/p/er9x/downloads ... nadiag.hex

jhsa wrote:I think it is obvious that there is a short to ground..the question is where..
.

Thanks to mike and erazz for their diagnostic fw's. I think these will be of good use also for others in the future with similar problems.

After checking the whole thing my resume is the following (If something is wrong, please feel free to correct me):

In normal state Pin 16 should emit 5V and if tied to GND via the left switch, an event occurs.

From the left switch one pin is GND. The other I have traced successfully to the 200 Ohm resistor, there's no short till this point, the capacitor has no short, so I assume he's alive. On the other side of the resistor towards the pad (violet in Mikes pic) there is a short to ground. Optical inspections tell me no problem on the pcb. Only the violet section is short to GND. So only thing I can imagine is - atmega has an "oink" on this pin, is this possible?? I've measured the other I/o switch pins, they show near to 5V with power on. The Pin 16 says 0,02mv.

So brick or not???? (My fault, I've not checked the TX before flashing for correct operation, was brand new and directly flashed - no reason at this point to switch on with the limited original FW)

Stefan

Busy at the moment, but I'll try to do another diagnostic to test the pin.

Mike.

The pin is programmed to have a pullup resistor on it, internal to the atmega. This is the only pullup, so if it is not working, then you won't get +5V. As a test you could try pulling the signal up with a resistor external to the atmega, around 10K would be a suitable value to try.

Mike.

MikeB wrote:The pin is programmed to have a pullup resistor on it, internal to the atmega. This is the only pullup, so if it is not working, then you won't get +5V. As a test you could try pulling the signal up with a resistor external to the atmega, around 10K would be a suitable value to try.

Mike.

Ok, so pressing a 10k resistor to the violet marked pad near pin 16, other end to multimeter's red pin (black to GND) should give me 5v?? Tested it but no effect, gives same 2mv as without. ???

Stefan

No.

One end of the resistor on pin16.
The other end of the resistor to +5volt.
Red probe (+) of multimeter on pin16.
Black probe (-) of multimeter on GND (-).

This should (hopefully) give you a meter reading of around +5volt

Reinhard

10k Resistor now between +5V and Pin16, again no 5V on the AtMega pin 16. Hope dies last. So now it's a brick (spare part donor).
Again - A big thanks to all for their generous help.

Stefan

last possibility:
can you unsolder pin 16 and bend it off the board (without breaking it) so it is nomore in contact with the board ???
Then do the same check with the pin and/or with the printed wire that pin 16 was attached.
This finally will tell you where the brick is.

Reinhard

yes, it will tell you if the short is in the atmega or on tne board

You guys never give up!!

Hey, it's not so easy to solder and unsolder that small parts when you have glasses like ashtrays (4,5 dioptres).....

But:

After unsoldering pin 16 I now have 4,9 V on the pin and no short on the pcb. ????????????????????????????????

Stefan

you know what? I think that if the problem is that pin only, your tx should turn on now normally.. did it?? or did it still hang??

So, while desoldering the tiny pin 16 I noticed a small resistance when trying to put the x-acto under the pin to bend it up. Seemed to be a micro small piece of solder. I thought this could possibly cause the shortage. After cleaning all with isopropyl I soldered the pin down again and voila, now the function on the left button is working.

Brick is alive again.

Thanks to a great community - teamworks seems to solve every problem.

Stefan

Gr8 news


Sent from my iPhone using Tapatalk

Awesome!

Now it would be interesting to hear from ekill if he has also made same progress using the testing firmware or solved his problem.

Stefan

hurrah... Those are good news. I bet u'll have a drink 2nite. I will.

Stuff like this always puts a smile on my face. It will also give you a little bit more satisfaction every time you fire up your radio and fly your plane.

jhsa wrote:hurrah... Those are good news. I bet u'll have a drink 2nite. I will.

No drink but took a walk with Mr. Bear:



Stefan

Ran just the keydiag so far. Stops at this screen. I have 4.37v at pin 16.

Earl

@ekill, Could you try my test code at viewtopic.php?f=10&t=465&p=5861#p5861, with a known good eepe file loaded. I'm not sure what Erazz's code does, mine reports where it gets to in small steps. It may well be you get to the screen shown, then when you press a key, it gets stuck before the screen gets updated with something else. My test keeps updating the screen.

Mike.

Mike - All keys work except left. The only epee file loaded is a "new model" file


Earl

Deja Vu??

I didn't think it would read any different...just hoping for a different clue. If I get around 5 v at pin 16 and the left switch is physically ok and the cap is removed, doesn't that mean that the short is between the left switch and ground? Or can the short be elsewhere?


Earl

Hi Earl,

The difference between my "diag" version and Mike's version is that Mike's version tries to load regularly and report where it gets stuck whereas my version dumps most of the startup and just shows the diagnostics screen.

Unfourtunately, I forgot to drop the reading of the EEPROM and that's why you get a "bad eeprom" message.
Can you try to load the diag version and then load a model meory?