I took apart an old magnetic door alarm, where when the contact is broken it counts down from 10 seconds for the user to input the code (which ranges from 1 to 8, there is no 0 or 9 key). This alarm is so old there is no information on the internet. I took it apart, and so far all of the internal components have been working, which is great for a hobbyist such as myself!

I am currently trying to reverse engineer it with just a 9v battery and a multi-meter, but I am having some trouble.

On the front it appears to say 894HB MEYER, and DG4 - 102 - D.

Here are two pictures: enter image description here enter image description here

Four of those pins used to be connected to an IC labelled LS 7220 LSI 8122 PM. This was used to check the correct pass-code was entered in the correct order. The other pins were not connected.

In the first picture, at the bottom, you see the eight pins. According to the original plastic alarm the pins are connected to buttons in this random order: 5 1 6 2 3 7 4 8

I connected the white wire to the negative terminal of a 9v battery. I did a continuity test where I connected one side of my multi-meter to the other 9v battery terminal, and then touched each one of the eight pins. Strangely they all had continuity, even when I did not push any buttons.

Could someone tell me how it works so I can use it for my DIY projects?


4 Answers 4


EDIT This is the logic diagram of how this board should be used.


simulate this circuit – Schematic created using CircuitLab

This has common cathode diode-OR switches arranged in 1 row and 8 columns with external pull down resistor to read when a button is closed the input select appears at the Vout. .

This is a similar 4x4 keyboard but not exactly the same which uses pullup instead with row and column scanning to access 16 buttons in the same number of lines so a button close is a low logic level. (negative logic)

enter image description here

  • \$\begingroup\$ What is the role of the diodes? \$\endgroup\$ May 31, 2021 at 20:33
  • 2
    \$\begingroup\$ They isolate the column voltages from switches in other columns and mux’d.low on each row \$\endgroup\$ May 31, 2021 at 22:15
  • \$\begingroup\$ Just so you know - a pin only has continuity when I press a button. The reason every single pin had continuity before was because I connected the wrong wire. \$\endgroup\$
    – questioner
    Jun 1, 2021 at 8:47
  • \$\begingroup\$ @Tony, interesting... The same (maybe decoupling, isolating, blocking, separating, detaching) diodes are used with the same purpose in fusible PROM. It is interesting to see the general idea behind them... the logic function they perform (maybe wired OR?). For example, row signals can be considered as SELECT signals for the common-cathode diode groups (like in 7-segment dynamic indications). \$\endgroup\$ Jun 1, 2021 at 12:45
  • \$\begingroup\$ -1 This button board is not a matrix at all. That's a good diagram of a matrix, but on this one each button corresponds to one lead. \$\endgroup\$ Jun 3, 2021 at 14:21

Here is the datasheet for the 7220:


It looks like the grey wire is common to all of the switches and the whitе wire is common to all of the diodes. The diodes could be part of a diode-OR/NOR circuit that detects the press of any button, independently of whatever the chip does.

8122 might be a date code for the chip. Photos?


I solved it myself in the end with some help from IRC.

The grey wire should be connected to one side of the battery, button one is connected to pin two which the multi-meter should bridge to the other side of the battery.

Hey presto - continuity only when pressed.

  • 1
    \$\begingroup\$ +1 for the self-confidence! \$\endgroup\$ May 31, 2021 at 20:40
  • 1
    \$\begingroup\$ You don’t want to short out the supply. Ther must be a current limiting resistor to sense voltage rise. \$\endgroup\$ May 31, 2021 at 22:14
  • \$\begingroup\$ @TonyStewartEE75 What ohms resistor should I use? \$\endgroup\$
    – questioner
    Jun 1, 2021 at 8:45
  • \$\begingroup\$ See my schematic for the value. \$\endgroup\$ Jun 1, 2021 at 11:08

This isn't a regular keyboard, and the diodes aren't there to allow multiple buttons to be pressed at once without shorting out the matrix.

On this board the gray wire is V+ and each of the 8 leads on the left correspond to each button. Each of those is also connected to one diode which all have their cathodes connected to the white wire. The purpose of the white wire is to allow the logic to know when any button is pressed.

The logic takes 4 wires which must be triggered in the correct order. These 4 wires are connected to 4 of the 8 leads on the left side in a custom order. This sets the password. If the password is 1,2,3,4, and someone enters 1,2,7,3,4, the 7 will trigger the white wire without triggering one of the 4 wires, which tells the logic that an extra button was pressed, and the logic can then reject the password. Without the white wire and the diodes, 1,2,7,3,4 would be a valid password.

The white wire is probably indirectly connected to the RESET on the LS7220 IC, but with some simple logic to only trigger RESET if one of the 4 sequential inputs is not pressed. This saves the user having to wire up all the other buttons to the a RESET line.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.