Telemetry sudden stopped on XJT module

[jhsa]

if you power the radio with both horizontal trims inwards, does it enter bootloader mode? If so, connect it to the PC. Is it recognized and do both drives show up on the computer screen?

João

[voodoo614]

I have to press horizontal trims outward, but yes I can get into bootloader. And the computer recognize the TX. I am able to read EEPROM and flash firmware.

I always wondered about the trim thing. I remember reading that you have to press inward, and I never got it to work. But outward always worked.

[jhsa]

It should be inwards for bootloader mode and outwards for maintenance mode.

There is something really no right with that radio.. Did you do any mods? Did you buy it from someone else or new??

João

EDIT: Being able to see the two drives means that the processor is running.
Please download the radio's eeprom onto eepskye, go to the general settings (first tab above model1) and make sure the contrast setting is at least at 27 or so..

João

[voodoo614]

I will check the drivers and the bootloader again when I get home this evening.

As far as the contrast, it is currently 25. Why should I have it to at least 27?

I bought the TX from Hobbyking. No mods.

[jhsa]

I wanted to say around 27

[voodoo614]

Just curious for contrast why it should be around 27? Is it to see the LCD or my initial problem with telemetry? I wasn't sure you read that I got the radio to power up again by bridging R9. My initial problem with telemetry still persist.

[jhsa]

Aahhhhh, I missed that post.. Sorry..

Sent from my thor using Tapatalk

[voodoo614]

FYI, just rechecked the inward trims and my TX doesn't going into bootloader mode. For me to upgrade firmware, I have to plug in USB and turn on the TX.

[MikeB]

What version does the bootloader report when you plug in the USB and power on?

I'm probably going to need to build a special test version of ersky9x to investigate your loss of telemetry further. I can't think of any reason why it doesn't work.
Just in case there is some setting that is causing the problem, try:
1. Backup your EEPROM (use eepskye Burn|Read EEPROM memory to file).
2. Using eepskye, create a new EEPROM.
3. Write the EEPROM to the radio.
4. Start the radio normally, the sticks won't be calibrated, but just skip all the alarms.
5. Change the protocol of the external module to XJT (PXX).
6. Power on the receiver and see if you receive RSSI.
7. Replace the backup of the EEPROM.

Mike.

[voodoo614]

RSSI is 0.

[MikeB]

What version is the bootloader?

Mike.

[voodoo614]

Forgot. Bootloader is 1.7

[MikeB]

The bootloader is at version 2.6 for the 'PRO. Even 1.7 should run at power on with both horizontal trims held inwards.
Please check in the "DiagSwtch" menu that you trim switches are all working.
I've just posted a test version for the 'PRO that includes a serial diagnostic option. Power on holding the RIGHT horizontal trim to the RIGHT. The radio should start with a screen showing "DIAG" and a '?' in the centre of the display.
Connect the Tx signal of COM2 (on the FUTABA trainer connector) to the SPort pin of the module connector.
Once per second, the character in the centre of the display should change, stepping through 'A' to 'Z'. This shows serial data is being received on the SPort connection at 57600 baud correctly.

Mike.

[voodoo614]

Thank you for taking the time to help me diagnose this problem.

The trims are working in diagswtch. I updated bootloader to 2.1, which is what r220 firmware provides. Now inward trims works. My TX is weird. I had that weird voice problem that we never solve. Where if I don't leave the splash screen on, the voice would disappear.

Anyway, I loaded the test version. All I got was DIAG in the upper left corner. No "?". When I connect COM2 TX to S. Port, I just get an "_". No count up of the alphabet.

I actually get an left arrow symbol first then the it stays at the underscore symbol.

[jhsa]

About the trims problem, you possibly had the wrong firmware flashed to your radio. That is why this new firmware solved the problem Not your radio being weird

João

Sent from my thor using Tapatalk

[MikeB]

I'll try to add some more information on the display, and maybe some extra test as well.

Mike.

[voodoo614]

Is there anything in the hardware that I can check?

[MikeB]

You could check that the bottom pin of the module connector is connected to pin 2 of J6 on the main board, and then on to R82 on the main board (100 ohm resistor).

Also, when testing the special version I posted, try with COM1 both not inverted and inverted.

Mike.

[voodoo614]

same result with inverted or non-inverted.

I get continuity both for R82 and Pin 2 of J6. resistor does read 100ohm.

[MikeB]

I've just posted a new, special test version. As before, power on holding the RIGHT horizontal trim to the RIGHT. The radio should start with a screen showing "DIAG" and a couple of '?' in the centre of the display. There will also be a 4-digit (hex) number on the bottom line with another character to the right (starts as '!').
Once per second, the character in the centre of the display should change, stepping through 'A' to 'Z'. This shows serial data is being sent on the SPort connection, and looped back correctly.
Now, connect the Rx signal of COM2 (on the FUTABA trainer connector, top right pin) to the SPort pin of the module connector (bottom pin). The second character in the middle of the display should now change with the first one.
This confirms the SPort pin is sending data OK, so is connected OK.

One possibility is the main processor clock is wrong. Please time how long it takes for the character to change ten times. It should be 10 seconds.

Do you have a serial port available on your computer (a real one with RS232 buffers, not a FTDI type device)?

Mike.

[voodoo614]

I posted a video below.

I tried it inverted and non-inverted. Same result. No count up of the alphabet.
Connecting COM2 RX doesn't change anything.

A real COM port on a computer... What is that?

I actually do have an old laptop that has a real serial COM port. I can dust it off.

[MikeB]

I'll need to try something else as a test, this is an odd problem, the SBUS signal doesn't seem to be sending or receiving.
You could try checking for some correct connectivity on the board.

One side of R82 connects to the pin in the module bay (you have already checked this.
The other side of R82 connects to:
R77 (10K)
D9
U7 pin 4.
The other side of R77 connects to:
R80 (10K)
Q1 (transistor base)

Mike.

[MikeB]

I may have had an error in the test version.
I have posted a new .zip file with two test versions in it, with names including Diag1 and Diag2.
Please try the Diag1 one first, this should display the words High and Low alternately, showing that the SPort signal is looping back correctly.
Then try the Diag2 one, which is like the original test version but with the possible error fixed.

Mike.

[voodoo614]

I tested both version

Diag1- I got exactly what you described. High and low alternating.

Diag2- Same as the video I posted yesterday. I tried both inverted and non-inverted.

I checked all the trace continuity, and they all checked out like your said.

[MikeB]

OK, I'll try to sort a version of "Diag2" that uses a slower baudrate and see if that works. I'm testing at 57600 baud at present (since that is the SPort baudrate).
If you run "Diag1" again, you should see a 4-digit number on the line under the second "High"/"Low" line. Is this number just 0000 or 0001 or does it show larger values?
For information, the upper line with "High"/"Low" on it is the output level, the lower line with "High"/"Low" on it is the input level and the number is how long it takes for the input to respond to the output.

Mike.

[MikeB]

"Diag3" test version posted. This starts at 2400 baud, goes through the alphabet, then increases the baudrate, displaying the current baudrate on the third line. This repeats for 6 different baudrates.
See if any of these baudrates work looping characters back.

Given that "Diag1" works, indicating everthing is connected, it is very odd that "Diag2" isn't working.

Mike.

[voodoo614]

I tested diag1 again. You can see the video below.

Diag3 - the only thing that changes is the count at bottom left. The baud never changes. No alphabet count. Connecting COM2 RX does nothing I waited at least 5 mins.

[MikeB]

Diag1 is wrong, both lines should read the same, i.e. both High or both Low. it seems the invert Com1 signal is not working correctly.
I'll see if I can come up with something with which to test it.
You could try loading Diag1 again, go into the Telemetry menu and make sure the invert Com1 is not enabled. Do this by actually setting invert Com1 on then off again. Then run the Diag1 test again.

Please also confirm the RTC is still running with the correct time.

I'm wondering if you have a fault with the I2C connection on the board. This is used to control the RTC (an hence the Com1 inversion) and also the volume control. It might explain why COM1 inversion is not reliable and also why you might lose voice output as the volume could be left a "off", so not being set correctly.

If possible please check resistors R40 and R42 (should both be 2K2) and resistors R70 and R71 (should both be 10K). Check both the value and that they look to be properly soldered.

Mike.

[voodoo614]

RTC is active and running. Inverted and non-inverted were tried multiple times.

As for all the resistors, they all gave me 1.8K instead of what you mentioned. I have verified that my multimeter is accurate with a known value resistor. Is there any other component I should check for values?

[MikeB]

Each 2K2 is in parallel with one of the 10K ones, so a reading of 1.8K is correct.

I'll hopefully post something that will test the inverter control later today.

Mike.