Results 1 to 20 of 20

Thread: XTS3000 from hell

  1. #1
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default XTS3000 from hell

    Hi Guys,

    I have taken on an XTS3000 from a HAM friend that is doing my head in. Just wondering if anyone has any ideas - my feeling is that the controller has an intermittent FBGA solder joint thats temp related but I'm not sure.

    Symptoms:-
    Radio constantly reboots some days and will not power up. Other days, it will reboot for 10 mins then come good and be solid as a rock. Some days, it works perfectly.

    Initially I thought that the codeplug was corrupt, then the radio powered on fine and was working OK. Just to add to the confusion, there are no FAIL or ERROR messages displayed ever.

    If anyone can chime in that would be great. I think the f&*kup fairy has flown into the radio and it needs to be extracted

    Cheers,
    Matt


  2. #2
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,478
    Thanks
    519
    Thanked 1,091 Times in 570 Posts
    Country: Christmas Island

    Default

    Take the board out and use a heat gun to re-cook the BGA chips. Be careful to not overheat them and fry the IC's.

    This is a LAST RESORT thing to do, it's easy to fry the PCB or one of the IC's if you're not careful.

    Also, wrapping all around the board with aluminum foil and exposing just one chip at a time to reheat will prevent adjacent components from melting or getting overheated.

  3. #3
    Join Date
    Feb 04, 2012
    Posts
    1,608
    Thanks
    65
    Thanked 300 Times in 150 Posts

    Default

    If you have a steady hand you an appply SLIGHT pressure on the chip be carefull you dont move the chip. I have seen BGAs where one point got less solder than it needed.

    I built a jig a few years ago for doing this work. All it is is a clamp to hold the board and a spring loaded suction cup. The cup is a silicone based sucker from an old tool kit. The spring was from a standard bic style pen. The idea being the suction cup holds the chip so to dosnt shift and the spring places a gentle downward pressure.

  4. #4
    Join Date
    Dec 21, 2011
    Posts
    4,051
    Thanks
    2,964
    Thanked 5,774 Times in 1,707 Posts
    Country: Canada

    Default

    Replace the 50-pin compression connector between the vocoder board and the controller. Clean the gold contacts on both boards, with a pencil eraser. (make sure you blow out the rubber debris)

    Part number is 2805214Z03. $13.39 USD on MOL.

    Do this BEFORE reflowing all the chips.

    Most "hobbyist" 3000s have been taken apart far too many times, and sometimes, people don't put them back together properly or don't take proper care in making sure the compression connector is correctly seated. The result is one or more bent pins. (or skin/oil contamination which affects reliability of the connector).

    You may also wish to check/replace the 20-pin compression connector which connects the RF board to the controller. The p/n is 2805216Z0, and price is $16.00 USD.

    Always suspect mechanical failure, first, unless it's an obvious or known electronic failure.

  5. #5
    Join Date
    Feb 04, 2012
    Posts
    1,608
    Thanks
    65
    Thanked 300 Times in 150 Posts

    Default

    Good point Mars. If it had been a flex circuit connector I would have suspected a cracked trace.

  6. #6
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    Update - this wasn't a hobbyist radio - in fact I doubt it had even been apart more than a few times at the service depot. In the end, I pulled it fully apart, cleaned everything, installed a new RF flex cable and battery connector/gasket combo.

    Put it all back together and its working well - the clue that made me suspect it was the battery connector was the service manual detailing how all the power lines go through the RF deck first. From there it was just a process of elimination to find the suspect parts. As I removed the RF deck I noticed the rear of the battery connector that mates to the RF deck had a pin that was slightly recessive, when I tried to bend it forward it snapped clean off which indicates metal fatigue/incorrect installation at some point. So that part got tossed in the bin and I ordered some fresh bits from my friendly /\/\otorola dealer and that seems to have done the trick!

    Cheers,
    Matt

  7. #7
    Join Date
    Dec 21, 2011
    Posts
    4,051
    Thanks
    2,964
    Thanked 5,774 Times in 1,707 Posts
    Country: Canada

    Default

    Thanks for the update, Matt. This just reinforces the fact one should always suspect mechanical failure, before electronic failure, unless known otherwise.

  8. #8
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    Hehe yep! The actual process of elimination was more like... codeplug, then firmware, then controller board, then RF deck.. then finally to the battery connector - which murphy dictates would be the cause of the problem and also make sure its buried underneath everything else so its the last thing to test

    In any case I'm just glad its fixed!

  9. #9
    Join Date
    Feb 04, 2012
    Posts
    1,608
    Thanks
    65
    Thanked 300 Times in 150 Posts

    Default

    Thanks for letting us know how it worked out. I happy it was a clear case of a bad connection/componant. I hate it when its intermittant like a cracked trace on a flex circiut. The the only solution is to play musical parts untill the problem either transfers or becomes permanent and can be located.

  10. #10
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    And this is why I say it has the devil inside it - this is what you get on power up!

    There must be a bicolour LED in the rotary knob as well - and for whatever reason its getting reverse polarity!

    XTS_red.jpg

  11. #11
    PRC148 No Longer Registered

    Default

    The encoder will turn red like that when the radio is bootstrapped for firmware loading.


    Of course, it shouldn't do that normally.

  12. #12
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    Yep - seen it before when it doing firmware upgrades - whats got me curious is that the voltages required to put the radio into bootstrap mode only come from the SRIB itself - which makes me wonder where the hell the controller is getting the bootstrap signals from..!

  13. #13
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,478
    Thanks
    519
    Thanked 1,091 Times in 570 Posts
    Country: Christmas Island

    Default

    Some of the very early prototype XTS3500's also had the Red LED used for normal illumination instead of the green.

    Just curious does it light up red after bootup or is it normal green? If it's still red then the LED's are either just reversed or both of them are red. If it goes normal after the radio finishes self-test, then something more weird is happening.

    If it only does this during self-test briefly on powerup it might be falsely detecting a VPP on the universal connector - the circuitry that detects bootstrap mode is indicating VPP is present so it lights the red LED, but since VPP isn't actually there, it doesn't really go into bootstrap mode it just thinks it's going to but doesn't and winds up booting up normally.

  14. #14
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    After putting the radio back together it was working perfectly for about an hour or so, then the red TX light came on with the red rotary. I power cycled the radio, and this is what its been doing since:-

    http://www.youtube.com/watch?v=Jl0v38iShdc

    I think theres been two problems - the faulty battery connector which was causing the constant reboots due to low voltage through the bad contact, and this other strange fault that seems to be in the controller board.

    ARRGHH!

  15. #15
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,478
    Thanks
    519
    Thanked 1,091 Times in 570 Posts
    Country: Christmas Island

    Default

    Wow, that is wierd - the controller is going ape-sh*t! (That's a technical term!) I assume the controller is also not waking up when that happens - the program is running amok, and it's writing random data to random ports in hardware. One of those ports is the LED latch, so it's causing the blinkenlights. If it happened after an hour it sounds like it might be thermally related, try putting it in the freezer for a while (or use freeze spray on the controller) and see if it starts working right again. If it's sufficiently screwed up it may have written data to the EEPROM so the codeplug might be corrupted - if it comes back it might come back up with a FAIL 01/82 or somesuch checksum error on the codeplug. Sounds like the controller finally went bonkers, there may be no fixing that one...

  16. #16
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    I'm inclined to agree - it went nuts after and hour or two on my desk at work doing nothing, and its been like this since - even after powering down and letting it cool down. I'll give the freeze spray a try but I think the controller is pretty much shagged..

  17. #17
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    Cold shock treatment didn't help either... hmph.

  18. #18
    Join Date
    Dec 21, 2011
    Posts
    4,051
    Thanks
    2,964
    Thanked 5,774 Times in 1,707 Posts
    Country: Canada

  19. #19
    Join Date
    Apr 09, 2012
    Location
    Australia
    Posts
    745
    Thanks
    151
    Thanked 348 Times in 149 Posts
    Country: Australia

    Default

    Hey mars, All done - changed out with known good ones from a working radio, and still the same result

  20. #20
    standardmissile No Longer Registered

    Default

    Looks like that was QC'ed by Murphy. Also it could have been a last friday/first monday assembly line radio. I had a similar problem with a Thales/Racal handheld, lucky for me the little ba***rd was still under warranty and it was sent back to logistics and done as an exchange.

    Also have you checked the battery/contacts? Shoddy contacts on the power supply could have caused a voltage fluctuation/inrush that damaged the circuitry.

    When we had that cluster of a blackout here in the Southwest not to long ago, the inrush of current when SDGE half-assed restarting the grid blew right through the power circuit on my TV at home and fried my D-Comb as well as the circuit that controls SVideo input. I spent 3 weeks grabbing at my 1/4" hair trying to figure the issue out, while my buds were nagging at me to play Call of Duty with them as soon as we got off duty. I finally nailed down the problem to blown caps, fried resistors, and one completely dead PT board, because I REALLY love my Sony 51" rear projection HDTV.