Announcement

Collapse
No announcement yet.

MTSVO-SC open/close is not reliable

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • MTSVO-SC open/close is not reliable

    Hi,

    This is the third unit that I've had so far, and I'm trying to find some other MTSVO-SC/K300 owners that might be able to shed some light on this subject.

    Unit 1 had a bad inverter. RMA.
    Unit 2 worked fine for a few months. Then the open/close started to go crazy. It would open, but not turn the screen up, then get stuck open trying to close back up. It was very tricky to get it to close back up. RMA.

    Unit 3 (present) also works fine, but a few days ago the open/close started to go crazy again. This is common between two units, so I'm thinking that a simple part is just not reliable or is running into trouble that can be fixed easily. Nearly the same thing is happening - the screen comes out, starts to come up, and stops at random positions, almost never in the correct one (at least 90% fail rate). Sometimes, it will not come up at all, and at others it can get 1/3 or even 1/2 way up to the normal position.


    Note that I have a dash panel that blocks the full swing by a very small amount, and this hasn't been a problem. The motor would just stop when it reached the last opened position. Mostly.


    Do any owners of this model see the above symptoms? I have conjectured two possibilities:

    1) Calibration drift. A sensor output was forgotten/corrupted in memory, and a full open/shut has to be done. The random nature of the problem seems to discredit this idea.

    2) Tilt position sensor corruption. Grease has found its way into a tilt sensor, causing its output to become erratic.

    I do not want to RMA again, because I have a sneaky suspicion that the next unit will fail the same way at about the same time.
    Has anyone disassembled one of these? What sensors are there, and where and what type are they? Any memory doesn't die with power off. Is there a service menu/NVRAM clear? Power off means the BATT line is actually disconnected and now power can reach the unit, not just ACC/key off.

    I haven't tried putting the unit on a bench and allow full deflection yet.
    Any help on this subject would be appreciated.

  • #2
    I have fixed this problem. I'd like to share what I've discovered about the auto open/close mechanism.

    There is an ALPS RDC506 low profile SMT potentiometer that provides an analog output against a 5V reference signal. This device is designed to detect a 320 degree swing of whatever arm/panel/whatever is attached to it. This has a gear on it that connects to a gear that drives the screen to angle up or down.

    I found that it was very easy to take this sensor out by supplying a battery (3V) to the slide in/out motor to expose the sensor completely (in the normal open position, it can't be removed, but it is visible. Upon removing the gear, I found exactly what I was looking for. A silicone based gear lubricant was used during assembly at the factory, the angle sensor is not tolerant of such contaminants. Some grease eventually found its way under the gear (it looks like a thin film was placed there on purpose also), and in a small hole on the sensor. Because silicone grease is a dielectric, it caused the random open circuit conditions that made the control board get horribly confused. After cleaning out the sensor (surface only, using a tissue and meter probe to clean the small windows on the gear side) as well as I could without having any contact cleaner, the mechanism works like a charm.

    I don't think this is a design problem, but rather a poor choice was made in the manufacturing process. This angle sensor can _not_ be contaminated by grease/water/whatever, or it will fail to provide accurate position data. Otherwise, it's rated at 1 million cycles, and is so simple that it has to be ultra reliable, barring any manufacturing beyond specifications.


    If anyone else has this problem in the future, I hope that this will help you out, and save some time waiting on RMA. A good contact cleaner rated for potentiometers should help, although mine seems just fine so far without it.

    Comment


    • #3
      do you have any pics, so people who don't have this screen know what you are talking about.
      it might be useful to troubleshoot other lcd's with faulty mechanisms

      Comment


      • #4


        '01
        Hang on while I defragment my car...

        the Acura ITX project
        the Acura ITX skin for RoadRunner
        project progress: 99%

        Comment


        • #5
          Sorry, I don't own a digital camera (or a conventional one), so I was not able to take pictures at the time. I can provide instructions that are as detailed as I can muster, so I hope this helps you.


          Tools:
          #1 philips screwdriver, good quality (not worn out)
          1/4" standard (flat) head screwdriver. Bring a few of the smallest ones you can find because it's a tight fit.
          3-5VDC power source (I used 2 AA batteries in series)
          electronics grade contact cleaner (optional) OR compressed air canister (optional)
          lint-free paper towel


          Prepare to unmount the unit and be able to disconnect the power while the screen is opening or closing.

          Open the screen, so that it is all the way out (for normal operation), at any angle from hoizontal to maximum. Disconnect the power as soon as the screen is in an appropriate position. If you can get the screen horizontal, it will be easier to handle on the work surface.

          Remove the unit, using whatever means necessary to properly disengage it from your particular mounting hardware. Obviously, make sure you don't damage the exposed screen. Place it on an appropriate work bench, and use static discharge protection if available. Otherwise, use care while handling any bare traces or wires inside. Touching the case will provide sufficient protection.

          On the front bezel (SRC/open buttons), there are two philips screws that can be removed with a #1 philips head screw driver. Remove them, and make sure you don't loose the screws or the little washers. Use your fingers to gently pull back the edges of the bezel while pulling away from the unit to release the clips. It helps to release one side first. Once the bezel is off, you can just hold it out of the way so you don't have to disconnect the wires.

          Place the unit on the table, up side down, so that the screen is facing up. If the screen is deflected past horizontal, you can use a book or some other means of holding the unit so that it is flat on the table without interference from the screen. I don't recommend using the edge of the bench or table, because the weight of the screen could make it easier to drop the unit.


          SENSOR1 is on a very small circuit card, placed between some of the tilt gears. It's easy to find, because it has yellow and a black 5V/gnd wires attacehd to it. In order to remove it, you have to extend the monitor drawer action beyond its normal maximum out position.

          Using a 3-5V power source, attach power to the drawer in/out motor. With the screen facing away from you, this is the one on the LEFT. It's a tricky reach, because the terminals are below the monitor ribbon cable, but it's not difficult to make contact for the small amount of time needed to extend the drawer. The polarity should be + to +, but I'm working off of memory. make the drawer extend about 1/2 inch more, so that the SENSOR1 board has room to be removed.


          The sensor board is attached to a steel support plate, and is secured to the drawer frame with a small brass post. Use the standard head screwdriver to remove the post in the same manner you would a regular screw. The screw below the post shouldn't fall down, because the frame is threaded to provide an anchor point.

          At this point, try to find an index on the gears, so that you don't loose the position. Don't worry if it freewheels out of place if you bump it with your finger.

          Remove the sensor board by gently pulling up and tilting it towards the screen (away from the back of the unit). It should come out pretty easily, but could require some jiggling. Be careful not to break the signal/ground wires by overstressing them by pulling or bending.

          Once the sensor is out, remove the gear (it pulls straight up), and you will find a criminal amount of grease under it. Use a shop towel, or anything lint-free to pull up the grease. Once it is all off the surface of the sensor, clean the gear. Notice that there are two small windows on the sensor where you can see the wiper arm. If you are having tilt related problems, these windows will have grease in them, interfering with the wiper arm's ability to make good, consistent contact. You can use an electronics grade contact cleaner, or simply repeatedly blow into the windows with your lungs or with the air can. If using compressed air, use care not to allow the liquid to contact the sensor, and use moderate to low pressure. The grease will exit at the sensor hub.

          When reasonably sure that the sensor is clean, reinstall it. This is pretty straightforward, so I won't bother with the details of reversing the disassembly order.

          Don't replace the front (lower) bezel yet, in the event that the position has accidentially shifted. If you find that the gear is off its original position, you can loosen the fastening post, and it will pull up enough to clear the gear for manual rotation, without the need to re-extend the drawer.

          If the gear is 180 degrees off position, the screen will tilt to about 50% full deflection, but manually pressing UP will allow it to reach near full deflection. The easiest way to align the gear is to guess a starting point, and move one tooth at a time until proper operation is achieved.

          If anyone does this and has pictures, please send them to me, or just throw them up here.

          Comment


          • #6
            Thanks, that's a pretty detailed walk-through. The one thing that I'm not clear on is the part about extending the drawer that additional 1/2 inch using two AA batteries.

            How do you cause the drawer to retract that 1/2 inch when you're reversing the process later, and will it automatically retract the exact distance that it extended, or is it a judgement sort of thing where you just stop it at a point where it looks about right?

            '01
            Hang on while I defragment my car...

            the Acura ITX project
            the Acura ITX skin for RoadRunner
            project progress: 99%

            Comment


            • #7
              Using an external power supply on the retract/extend motor bypasses the controller logic. When you turn the unit back on (plugged in), the computer senses the position is at full extended, and will begin the retract procedures. When you push the open button again, it will just stop in the normal full open position.

              There are two optoisolator type sensors that detect the position of the drawer. One is in the rear, and one is in the front. When the infrared light is blocked by a small fin on the frame, the controller senses the position of the drawer. The blocked sensor is the position, as illustrated below.

              The sensor looks like this on its edge: |_|

              BACK low, FRONT high: fully retracted, stop retract motor. User intervention is required to change.

              BACK high, FRONT low: fully extended, stop retract motor, and start tilt motor until ANGLE is >= memory angle, or start retract motor when closing and ANGLE == minimum.

              BACK high, FRONT high: undefined state (ignored), drawer is in transit. If power was disconnected in this state, the user has to push the drawer back manually so BACK is low.

              This design is safe, because an emitter failure in either or both sensors will cause the motor to stop.

              One final note: The ANGLE (SENSOR1 sub-***'y above) input must be at minimum, so the tilt motor could try to run past horizontal. If this happens, disconnect power, and rotate the sensor using the aforementioned service procedure to realign it. Be aware of the 180 degree off possibility as well. The motor is not strong enough to cause damage in the event of a sensor misalignment, so don't panic.

              Comment


              • #8
                Does anyone know where to get a replacement LCD panel for this display? I managed to damage the contacts on the ribbon cable attached to the LCD panel while working on mine. EVerything works fine on the display now, except there is a verticle stripe where the LCD remains dark.

                Comment

                Working...
                X