Author Topic: Dip Switches  (Read 4478 times)

0 Members and 1 Guest are viewing this topic.

Online bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2236
  • Country: us
Dip Switches
« on: April 12, 2019, 03:07:13 pm »
I'm confused about something. I have a really old garage door opener (Craftsman 139.653000) and want to change the code which requires changing dip switches. The manual states (which makes sense) both the receiver and remote need to have the same dip settings.

The remote seems to have three positions (+, -, and 0) and I don't understand why. Also, the receiver has push-button dip switches and I don't know if up/down mean on or off. Can someone tell me the meaning of the three-state dip switch and also whether up is a 'on' or 'off' on the other?

On a side note, something doesn't seem correct because I changed the dip switches in the receiver only, and the remote still activates the door opener. I returned the dip switches to their original state, changed the dip switches in the transmitter, and the opener still works. So it's like the dip switches don't affect the 512 security possible combination options; although the manual states otherwise.

Here are pics of the dip switches:





 

Offline tsman

  • Frequent Contributor
  • **
  • Posts: 599
  • Country: gb
Re: Dip Switches
« Reply #1 on: April 12, 2019, 03:28:49 pm »
The receiver sounds faulty if the dip switches don't affect it at all as they're definitely there to set the house code.

For the transmitter: + = ON, - = OFF and 0 = unconnected.

For the receiver: Pushed in on the left = ON and pushed it on the right = OFF. It is a rocker switch so you can only push in one side.

The IC in the transmitter has a trinary code hence why it has 3 different states on the switch. It'll probably be something like the old Motorola MC14502 encoder IC. The decoder IC in the receiver they're using is binary only which is why it only has 512 different codes and only has ON/OFF on the switch.

The MC14502 encoder datasheet says:

Quote
Each transmitted trinary digit is encoded into pulses. A logic zero (low) is encoded as two consecutive short pulses, a logic one (high) as two consecutive long pulses, and an open (high-impedance) as a long pulse followed by a short pulse.
 

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1718
  • Country: gb
Re: Dip Switches
« Reply #2 on: April 12, 2019, 03:32:02 pm »
That is indeed confusing - about changing the DIP switch settings and the device still working, often these things use the PT2272/PT2262 or equivalent (HT12D/E series) devices that have ten tri-state address settings.  That might explain the use of a tristate DIP switch - each of the 8 are either, VCC, GND or High-Z (not connected to either).

It seems the receiver is using a two-state DIP switch, although it is hard to tell from the photo it looks like pushing down on one end of the switch, raises the other, so its either connected, or not.  You can check this by looking on the underside to see if it has two rows or 8 pins, or three.

If you have access to the encoder chip in the transmitter, and a logic analyzer, you can see the output which is explained in the datasheet for the encoder chip (whichever one its using). But to me it certainly looks like the transmitter address is 10ZZ10ZZ1, whilst the receiver has 101101101. Odd.  It could be the receiver doesn't use a standard decoder, but a microcontroller, that can accept more than one address (a learning receiver).
 

Online bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2236
  • Country: us
Re: Dip Switches
« Reply #3 on: April 12, 2019, 03:41:14 pm »
Sorry, I should have stated that pushing down on one raises the other side (on the receiver dip switches).

Using the 'off' side of the dip switches as reference, I can't tell if when the button is down, whether that's 'off' or it needs to be up to be 'off'.

Initially I tried to figure out this by comparing the 'up' and 'down' state of each switch to the transmitter, however, when I saw the transmitter was tri-state, I got really confused because the switch states didn't match.

My focus was on dip switches 1-3. So I moved all three on the receiver only, and the transmitter still controlled the door opener. I returned the switches, and moved all three on the transmitter, and was still able to control the garage door opener.

I'll investigate it more this weekend, but I'm confused. A contractor has essentially stolen my spare remote, so I'm trying to change the code to avoid him accessing my garage.

 

Offline madires

  • Super Contributor
  • ***
  • Posts: 8274
  • Country: de
  • A qualified hobbyist ;)
Re: Dip Switches
« Reply #4 on: April 12, 2019, 04:26:21 pm »
If you like to improve security you can buy kits with key fobs and a matching receiver module supporting rolling code. The receiver module has a relay for interfacing the garage door opener. And don't forget to disable the old receiver circuitry. I've done this mod last year for two garages and it was fairly easy. IIRC, I paid 40 bucks for each kit with two fobs and a two channel receiver.
 

Online bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2236
  • Country: us
Re: Dip Switches
« Reply #5 on: April 12, 2019, 05:24:14 pm »
Hmmm my garage door opener is weird, and, although I haven't spent much time trying to understand it, nothing makes sense. I have a module plugged into an outlet that feeds apparently the receiver located on the wall, and then a set of wires that go up to the opener.

If I unplug the 120V AC from the garage door opener and push the transmitter, the module plugged into the outlet clicks (obviously a relay). I'm assuming this feeds 24V AC to the receiver on the wall, but don't know why this module has a relay.

Maybe the receiver activates the relay inside the module, and sends a signal to the garage door opener?

In any case, I'm uncertain if I can incorporate such a system into this ancient opener. Also, my car is equipped with buttons that learn the code and eliminate the need for keeping an opener in the car. So a rolling code probably will not work.
 

Offline georges80

  • Frequent Contributor
  • **
  • Posts: 930
  • Country: us
Re: Dip Switches
« Reply #6 on: April 12, 2019, 06:16:50 pm »
Homelink (even our older 2003 vehicle has that) deals with rolling codes.

I'd just get a modern garage opener for 1) better rolling code security and 2) better safety features such as optical beam. 20+ years ago when expecting our first child I replaced the antiquated Sears garage opener, it's a pretty simple task.

Since then I've added a MyQ Wifi connected opener that pairs with the 20+ year 'new' garage opener to send compatible rolling codes to open/close the garage and also tell you when it was last opened/closed (or still open when you drive off forgetting to close it). Also handy when the kids visit and want to open the garage to get into the house.

I see no reason to deal with an ancient garage opener when for $100 - $150 you can get something modern and safer. Unless you are really tight on $.

On your 'problem', did you cycle power on the garage opener/remote after changing the dip switch settings? Might be the new setting is only read on a power cycle.

cheers,
george.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Dip Switches
« Reply #7 on: April 12, 2019, 06:21:25 pm »
Push down on the Off side turns it off, push down on the On side turns it on, it's just a covered rocker switch, all rocker switches use this convention.
 

Online bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2236
  • Country: us
Re: Dip Switches
« Reply #8 on: April 12, 2019, 06:35:46 pm »
Quote
On your 'problem', did you cycle power on the garage opener/remote after changing the dip switch settings? Might be the new setting is only read on a power cycle.

I had power unplugged on the receiver during changing the dip switches, so reapplying power should have caused the receiver to not recognize the transmitter with the 'old' code should a power cycle be necessary. It's a good suggestion nevertheless.


 

Online bostonmanTopic starter

  • Super Contributor
  • ***
  • Posts: 2236
  • Country: us
Re: Dip Switches
« Reply #9 on: April 15, 2019, 01:44:06 am »
I discovered the problem.

Turns out the receiver must have broken at some point or someone designed it differently. The receiver, although has had power applied to it, has basically been a "dummy box". The electronics was powered, but it didn't have any connections to enable the garage door opener.

The module that I assumed was giving this box power turns out to be the receiver. It doesn't have dip switches, but has a switch that (which I spent time trying to figure out) needs to be pushed after the remote (with the new code) button is pushed.

When the remote is pushed (after the box learns the new code), the box receives the signal, This in turn closes a relay which shorts two wires, these two wires meet at the "dummy box" where all the switches to open/close the door meet. Another wire goes up to the garage door opener.

From what I could tell, the garage door opener is triggered when two contacts are shorted. So it's why all the wires meet at a single point.
 

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1718
  • Country: gb
Re: Dip Switches
« Reply #10 on: April 15, 2019, 03:15:44 pm »
Ahh yes, the "learning receivers" are everywhere on ebay/amazon/alibaba etc.. because they're so much easier to set up, and can "learn" from remotes with different encoders.  I'll second the advice on the rolling code systems.  It's easy to say that the basic systems are insecure - but that assumes someone knows what they're doing in the first place (unlikely), but a rolling code system is essentially unbreakable.
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 8274
  • Country: de
  • A qualified hobbyist ;)
Re: Dip Switches
« Reply #11 on: April 15, 2019, 03:46:27 pm »
The old school codes are basically an address plus function (push button ID). They are easy to record and replay. The rolling codes are a little bit more complex and add a varying element to make replay attacks harder. They aren't unbreakable but require some effort by the attacker. It's a compromise between security and simplicity (one-way communication). More secure systems like challenge-response would require two-way communication.
 

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1718
  • Country: gb
Re: Dip Switches
« Reply #12 on: April 15, 2019, 08:40:02 pm »
The old school codes are basically an address plus function (push button ID). They are easy to record and replay. The rolling codes are a little bit more complex and add a varying element to make replay attacks harder. They aren't unbreakable but require some effort by the attacker. It's a compromise between security and simplicity (one-way communication). More secure systems like challenge-response would require two-way communication.

Yeah I wasn't going to go into the latest side channel attacks as they cover most existing systems used in cars.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf