Genie Garage Door problem

Discussion in 'General Electronics Chat' started by Tonyr1084, Sep 18, 2016.

  1. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    The door works from the remote but does not work from the wall switch. The safety beams are working properly. The wall switch has a red LED that is lit. The wiring is all good, but when you push the button to open or close the door nothing happens. Nothing at all.

    The wall switch has three functions: 1) to open and close the door. 2) to activate the garage door opener light. 3) to lock the garage door opener (prevent it from running).

    Upon opening the switch there's quite a bit more circuitry in there than I thought would be. The board is about 1" by 2". I even tried shorting the wires from the switch but nothing happens.

    I get the feeling it's going to be needing a new wall switch but I hate to spend money without fully diagnosing the problem(s). Any advice? And if anyone says "Google it!" FORGET IT! Already tried. Found a site that would answer all my questions if I promise to pay a fee. Couldn't find ANY free troubleshooting guide.
     
  2. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    I fear it may be broken.

    Unplugged, damaged cable, fuse?
     
  3. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    If the cable were unplugged or damaged then the red light would not be lit. Same if there were a fuse involved.

    Two wires go from the remote to the wall switch. The switch has a red light on it. (LED). Cables are OK. Red light indicates as much.
     
  4. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    If those are the only two wires going to the switch then either the LED is in series with the switch or in parallel with it. If the LED remains lit then in the first case the switch is shorted but in the second case the switch is open regardless of its position.

    Impeccable logic unless there is some other circuitry in the switch box.
     
  5. #12

    Expert

    Nov 30, 2010
    16,247
    6,744
    Do you mean something like a circuit board that is 1 inch by 2 inches?
     
  6. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    Yes. As for exams, always read the question first.
    Could we have a photo of the board?
     
  7. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    It's not as think as you simple.

    This is a "Series 3" wall switch. It has three buttons. First opens and closes the door (as stated before). Second is a lock to prevent unauthorized openings from the wall switch (as stated before - sort of) and a third for controlling the lights (as stated before).

    The circuit line has 5.17 VDC unloaded (with the switch removed). Sitting here now I'm thinking I should have tested the voltage with it connected. I CAN do that if I need to.

    I have a series 2 control switch, but when I put it on the ONLY things that happens is when I hit the door button the two garage lights come on. They go off if I push it again. The lights also come on if I push the light button - and go off with the next push. IF I push the lock button then nothing happens. No light, no other actions. When NOT locked the lights can be switched on and off. Nothing more. No matter what, a series 2 won't actuate the door. It will take a series 3 wall switch to operate normally.

    So I have the S3 on the bench. According to the installation instructions there is NO polarity. Yeah, I know, 5 Vdc. I can see D3 & D4, so I assume there's also a D1 & D2. Maybe they're rectifying polarity and sending some sort of pulse back up the line - same as they do with the safety beams. Those also operate on two wires each with no polarity. They did that so the average home owner can install them without having to follow polarity.

    Anyway, I'm wondering what I should test on it. IF there's any sense in testing it. I don't know what I'm looking for, nor do I know where to test for what sort of signal other than to power it with 5V and scope it.

    U1 is a TI5551 chip. Assuming it's a Texas Instruments 555 timer. Though exactly how it's configured I don't know. There's (as you can see) a white membrane covering the board preventing me from tracing out the circuitry. But since the series 2 DOES actuate the lights then maybe I can conclude that since the series 3 does NOT actuate the lights either it's stuck in lock mode or the switch unit has failed.
     
  8. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    Told you it's much more than just a simple switch.

    UN-impeccable Logic. NO?

    And U1 pin 1 is in the upper left corner. Sorry, missed that detail.
     
    Last edited: Sep 18, 2016
  9. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    Scoping this thing. Powered from 5VDC, either way the LED lights up. No obvious polarity.

    When hitting the "Door" button, scope set to sweep 50µS per division, I get a square wave of 1.4 divisions at 7mV. Duty cycle is high for 1.2 divisions (I'm not good with scopes - don't blast me please). There's also quite the peak on the leading edge of the positive going wave form. Same drop off on the negative going wave form.

    The "Lock" button produces a square wave of 5.4 divisions (same voltage) and the "Light" button gives me a square wave of 3 divisions.

    The 7 millivolt thing doesn't seem right.

    When I reverse the polarity of the power I get roughly the same readings. Instead of 7mV I get 6mV. It may be time to spend a few bucks and just get a new switch. I'd hate to spend that money for nothing.
     
  10. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    OK, THROW OUT THOSE RESULTS! I HAVE A BNC TEE ADDAPTER THAT APPEARS TO BE FAULTY.

    ˚J˚
    ¬

    Sweep rates are the same but the voltage is more like 320mV square wave.

    Don't buy electronics from Amazon, they'll lead you to the wrong conclusion. 320mV is a HUGE difference from 7mV. Tested the output from my supply and got hardly any reading. When I removed the TEE I got the expected 5 volts.

    Learn't that lesson well!
     
    Last edited: Sep 18, 2016
  11. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    Do you have a resistor in series with that 5V supply?
    Is the 320mV across the supply?

    TL5551 is a cmos version of the '555 and if that is the only chip then the signalling is likely to be the frequency, pulse width or mark space ratio of a square wave. The switch box circuit may be something like this with the switches connecting different timing resistors to the '555.:
    upload_2016-9-19_9-34-14.png
     
  12. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    If you've looked at the last photo I uploaded you see two screw terminals, one marked "BW" and the other marked "W". Those are the only inputs / outputs there are. The voltage across the two terminals is 5 V. The pulse train when you push the buttons is 320mV. So somehow the head unit is sending a 5 volt signal down the line and listening for any changes in the load. The frequency of the load is being read and the head unit is deciphering the code. The shortest code comes when you press the door button. The longer code (lower frequency) is associated with the light button. The longest code (lowest frequency) is the lock button.

    That's it - just two wires ringing 5V down the line. The head unit doesn't care which wire gets wired where. Same is true of the safety beams. They also have only two wires going to them. And they are wired in parallel. One unit is the transmitter and the other is the receiver but they're both wired to the same two terminals on the head unit. Just exactly how Genie is doing it I don't know. All I know is that it works.
     
  13. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    Is that 320mV with the switch connected to the main unit or connected to a 5V power supply?
    If it is with the power supply, can you measure it while the switch is connected to the main unit?
     
  14. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    That'd be very difficult. The garage is detached, and I don't want to drag all my equipment outside. What do you think that will yield?
     
  15. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    If it works as I suspect I would expect a bigger signal.
    Also, on your PSU all three buttons send a signal, it would be good to know whether they do when connected to the main unit.
    Also, also the main unit may 'communicate' back to the switch in some way.
     
  16. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    I may give it a try later on today. However, other than a 555, four diodes and several transistors (designated Q1 through Q5) some resistors and a few capacitors I don't see the switch being any kind of "Smart" device where communication would be necessary. As far as I can tell, the lock button should disable the system upon the first push and enable it on the next. It would appear that the "Lock" command is stored in the head unit so that no other inputs other than the proper coded input will change that.

    The whole reason why I'm having trouble with this is because a few days ago I re-sided the garage. I removed two Halogen 300 watt lamps and installed a single motion detector light. The two Halogen's were individually wired. Since I was going from two to one I had to flip the breaker off in order to remove the old wiring safely. When I restored power the GDO (Garage Door Opener) didn't work. I was going to troubleshoot it but when I came back it was working. Don't know what changed. But yesterday I interrupted power again so I could install the motion detector light. When I restored power - again, the GDO wouldn't work. Thinking I had a bad breaker I began testing for voltage. At first I didn't get indications of power at the head unit. So I pulled the board out and checked out the connections and such. Everything appeared to be in order.

    Two years ago I had a new Genie go bad after just 4 months (motion detector circuit failure). Genie sent me a replacement and allowed me to keep the defective head unit. I fixed the motion detector circuit - it was just a bad plug. Yesterday I swapped control boards to see if the older head unit was at fault. Nope! Nothing changed. I tried an older "Series 2" wall switch (similar electronics but different codes) and the GDO still wouldn't operate. However, I DID have control of the lights and the lock. Series 2 won't operate the GDO, so I probably need to just go buy a new series 3 wall switch. Can't be that expensive. But I like to mess with things and gain an understanding of how they work.

    From what I've learned, the GDO supplies a low voltage (5v) to the switch and to the safety beams. Those devices communicate back along the same lines by variances in the load. That's why you can't just bypass a safety beam by jumping the two wires together. Full voltage or dead short - the GDO won't work. Use a variable resistance and again, the GDO won't work. It has to be a specific signal. Load fluctuations. Going to the garage with scope and power supply in hand isn't going to be easy. I'll probably just go buy a new series 3 and HOPE for the best. I HATE throwing money at a problem without fully diagnosing the cause. But sometimes you have to stab at it and just hope to hit an artery.
     
  17. MrChips

    Moderator

    Oct 2, 2009
    12,414
    3,353
    Now I'm curious.

    Genie_front.jpg Genie_back.jpg

    W-connection is neutral (GND).
    BW is signal sitting at +12V.

    Trace for Light Button:
    Geine_light.jpg

    Trace for Door Button:
    Genie_door.jpg


    I should have checked the duration.
     
  18. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    Mr Chips: My instruction manual on the hookup states that the head unit does not care which wire goes where. "Polarity does not matter" they say regarding both the wall switch and the safety beams. I have the Genie 1000. Both control boards send out 5 volts. Interesting you get 12 volts. I wonder if I have a bad wire - weak connection, not broken, not shorted. But why would the series 2 control the lights?

    My other GDO, a Genie 2000 is the series 2, from where I got the switch. The series 3 system - you can not short the two wires together and have the door do anything but the series 2 system, if you touch the two wires together they trigger the door. Maybe when i replace the switch I should replace the wires too. Hey! Couldn't hurt.
     
  19. AlbertHall

    Well-Known Member

    Jun 4, 2014
    1,878
    370
    There are quite a few reports of these GDOs getting 'confused' after a power outage. The cure seems to be to unplug it for a minute then plug it back in. It's worth a try.
     
  20. Tonyr1084

    Thread Starter Active Member

    Sep 24, 2015
    529
    86
    Tried that several times. Disconnecting to test power cord and other things, still no response. However, what you say makes sense because the first time I interrupted power the GDO stopped working. Later, perhaps an hour, it worked again. Yesterday letting it sit for four hours made no difference. But alas' I'll try one more time. Who knows. Maybe I don't have to do anything but put it back together.

    [edit]

    Tested the line with VOM. Without the switch on the line there's 5.17 volts. With the switch ON the line there's 4.4 volts. At the head unit there's 4.44 volts. I'm thinking something has gone semi-shorted on the switch. Wires seem to be good enough. No real reason to expect them to have failed. The 40 mV difference from the head unit to the switch - it IS a small wire. Line loss can easily account for the drop in voltage. And maybe 4.4 volts is not enough for the switch to operate properly. Keep in mind I get the same readings with two different head units. Genie SilentMax 1000.

    I'm going shopping.
     
    Last edited: Sep 19, 2016
Loading...