Connect with us

Subaru dash clock problems - pictures...

Discussion in 'Troubleshooting and Repair' started by Graphite, Jan 24, 2012.

Thread Status:
Not open for further replies.
Scroll to continue with content
  1. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Hi - I found your forum looking for some advice on the possibility of repairing my Subaru Legacy dash clock & fuel info display.

    The problem is that sometimes the whole display turns off as though it is getting no power. Over the last week or 2 the display has spent gradually more time off than on. When it comes back on everything is displayed correctly, when off absolutely nothing is lit at all. Makes me think power side related rather than anything more complex.

    The bottom of the clock gets hot when running - I have just connected it lose in the dashboard to try to find out if there are any consistencies to when it is off or on.

    Looking for any dry solder joints I took the unit apart to see if anything obvious - looks like it may be heat related from the centre resistor in the bottom of the clock? There is a brownish haze that has spread equidistant from the centre resistor on the rear of the PCB and also the white band at the bottom of the front of the PCB is only brownish in the region of the centre resistor. :confused:

    Pictures below...

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    Any ideas?

    My first thought is to swap out that resistor - would a failing resistor be generating excess heat? and why would it work sometimes and other times not work???

    Any advice appreciated

    Thanks

    Mick :D
     
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,500
    2,840
    Jan 21, 2010
    None of those resistors look to be a problem.

    I would suggest that the high voltage to drive the display is failing. I would look closely at the inverter.
     
  3. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    The 2 resistors on the right are labled '18Ω ·J' and the one on the left '10Ω ·J' They are about 12mm long.

    From a 2004 vehicle.
     
  4. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Thanks - Which bit is the inverter?
     
  5. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    There are two 18R and one 10R in series. The 18R's will thus get considerably hotter than the 10R.
    One may wonder why they didn't use three 15R's, or two 15R's and one 16R to distribute the heat more evenly.
    Though resistors can fail intermittently the problem is more likely cracks in the PCB tracks or "dry" solder joints.
    There is evidence of failing solder joints on the 18R resistors in the third picture.
    The resistors are there to provide a suitable heater current for the VFD. It is a rather inefficient albeit cheap method.
    If you measure end to end on the display corner "supports" you should get around 1.5-2.0 Volts.
    If this drops to zero when the display goes dark then the next step is to check the voltages between the resistors.
    There should be 0V between neighbouring pins of the resistors for example.
     
  6. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    OK - thanks dry joints was my first suspicion - is it worth just re-melting those joints?

    I'll check the things you suggest too.
     
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,500
    2,840
    Jan 21, 2010
    I can't see exactly what Resqueline is suggesting, but yes, if that is the problem then it may be fixed by a single touch of a soldering iron (but better to resolder it properly)

    I believe those displays require a high voltage to operate. Somewhere there will be a small inductor and either an IC or a couple of transistors and other assorted components which step up the voltage.

    They may be under the display.

    If the problem is a bad solder joint then this is all moot.
     
  8. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    I'm not so sure there is an inverter. Although these displays usually runs on 27V or more, they can manage with 12V (or even less).
    If there had been an inverter then it would have been the natural source for the filament voltage too. With resistors for this I'm sure they've skipped the inverter.
    See the attached blow-up of the third picture, where I have circled the suspect joints. Vf is the filament (heater) terminals of the display, where you'll measure 2V or so.
     

    Attached Files:

  9. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Thanks - I have 'blobed' some solder on the areas you pointed out - tested - working fine at the momement. So at least I haven't 'broken it' LOL :D

    [​IMG]

    Not the tidiest job - but I'll see how it goes. I have re-assembled the unit in the dash - I'm fed up of driving round in my car with a gaping hole there :rolleyes:

    Thanks for being there for me guys :cool:
     
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,500
    2,840
    Jan 21, 2010
    I'll file that one away...

    OK, now I see them.

    The fix looks good to me too, but I'll leave it up to people with better eyes :D
     
  11. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    LOL - better camera maybe??? :D

    before...

    [​IMG]

    after...

    [​IMG]

    Marks out of 10? ;) - Probably shouldn't ask, but tips on how to do it better... 'just in case' there is another time :D
     
  12. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,500
    2,840
    Jan 21, 2010
    One of the connections (about a third of the way in from the left) appears not to have wetted the lead fully. Note that the top of the joint is concave. You can also see that the solder hasn't flowed over the pad smoothly. This is generally caused by the soldering iron being too small (or not hot enough) so that the joint cannot be heated up sufficiently.

    There is a little too much solder on the connections about 2/3 of the way from the left. The left side also shows some evidence of poor wetting on the lead.
     
  13. nobrob

    nobrob

    22
    0
    Sep 7, 2011
    My advice: practice.

    Get some solder, your iron, and some scrap metal (a scrap circuit board from a broken device, for example). Put some solder on, put some more on, and keep going. Get to know how long it takes your iron to melt your solder and for it to wick around/over whatever it is you're heating up.

    The theory is that both pad/wire/whatever and solder need to be hot enough. The solder won't just stick nicely to a cold surface.
     
  14. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Thanks guys... yes correctly observed soldering iron is on the small side with small pointed tip - and wire wasn't long enough to get the part into good light either :(

    But - good news - been out in the car a few times today and the clock has functioned perfectly so far :D

    - Thanks again!
     
  15. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Darn it! Clock has stopped working again after a few weeks...

    If I plug it in and apply a light twisting pressure from one end to the other it will work again for a short while...

    Any ideas - or further tests to try?

    Cheers :D
     
  16. mrmodify

    mrmodify

    157
    32
    Feb 13, 2010
    If you can twist board and make it turn on and off you most likely still have a bad solder joint.

    I use a very powerful magnifying glass or a jewelers loop to find the bad joint. They stick out like a sore thumb when magnified.

    I will heat the joint with a iron and suck the solder off the pad, clean with a wire brush (lightly) and then solder. Heat joint first then dab solder at tip of iron and pad till solder starts to flow good then add solder. You will have to be fairly quick or you can fry the parts or/and the board.

    When I can't find it by the method above I have also taken a plastic welder (Harbor freight) which is basically a high heat heat gun and passed it over circuit board a couple of times which will remelted the solder.

    Goggle HP DV2000 or DV9000 laptop video repair they will show placing the mother board in the oven set at a certain temp for certain amount of time to remelt the solder. you will have to remove or watch out for some plastic parts:D
     
    Last edited: Mar 16, 2012
  17. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    I did a bit of video of how easy it is to get to work with a gentle persuader on the PCB - and then it wasn't so easy as just a gentle bump - but a light bend end to end did it...

    [​IMG]
     
  18. niftynev

    niftynev

    22
    0
    Mar 20, 2010
    Graphite - I'd replace that big black 47uF 160V capacitor as a real suspect mate. Often, a HV inverter for a fluorescent display will have a capacitor across its output. If this cap dries out, the inverter can hiccup or just stop working, & the display will blank out. Bending the board may be moving its leads around as well.
     
  19. Graphite

    Graphite

    33
    1
    Jan 24, 2012
    Do i need the exact capacitor to replace? - or can I get a 'similar' size? which values are important?

    Sorry for the noob questions :D

    Have found some on ebay - but for pack of 20 :rolleyes:

    http://www.ebay.co.uk/itm/47UF-160V...CAPACITOR-RUBYCON-YK-X20-fbe4b6-/120841714471

    I have asked the seller how much for one LOL :D
     
  20. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,500
    2,840
    Jan 21, 2010
    If bending the board changes stuff, then I doubt that the capacitor is the problem.

    The problem is still most likely to be a poor solder joint, or a crack in the board.

    I note that there appears to be surface mount stuff on the top side. It may be useful to go over it with a loupe. There may be a broken solder joint or even a broken component. Surface mount components are far more sensitive to the stresses caused by flexing the board because many don't have flexible wire leads to absorb the forces.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Thread Status:
Not open for further replies.
Electronics Point Logo
Continue to site
Quote of the day

-