HomeLink HA Interface #4 - Slowdown to use a breadboard

So life happened again.  I wish I could just get down with development and not have to stop for something, but that is never the case in my life.  Maybe when I am retired it will get better; hopefully more time available for my hobbies.  Right now I am in the middle of a kitchen update and work activities that are ramping up again.

I decided to go ahead and get a couple of Raspberry Pi Zero breadboards to see if I can solder the required resistors and the level shifter, along with a couple of breadboard electrical connectors.  I have it in mind to run power to the RPi breadboard and then connect the 850LM to the breadboard as well. That way, the power goes to one location and is distributed from there.  It would be nice if I could find a simple board that provides 5v power without being huge.

That being said, I fritzed a layout using the Adafruit quarter proto-board.  I used pretty much the same connections, except that I purposefully put most of the power traffic cop on the proto-board.  So there are now five entities that will need to go into a 3D print, counting the LiftMaster 850LM.  The fritzing layout (not showing the 850LM) follows:

This was pretty much as simple as I could get it.  I am using a slightly different Buck-Converter than the Adafruit one shown in the Fritzing diagram, but the principle is the same (9v -> 5v power for the RPi Zero W).  I am assuming at this point that when the relay switches and drops the voltage to zero that the 9v power will not drop below the point where the RPi will "brown out."  The resistors are 220 ohm which will give a draw of 40 ma from the 9v power supply during the relay switch. Note that the 850LM channel 1 is tied to GPIO4, channel 2 is tied to GPIO5, and channel 3 is tied to GPIO6 on the RPi Zero W.  The fritzing RPi is a Zero but that is okay for illustration purposes since the GPIO pinouts are the same.  The use of a RPiZW is for wifi access in my IOT vlan.

HomeLink HA Interface #3 - Started the breadboarding process using an RPiZW, Buck converter, and 850LM

So I was able to start breadboarding the whole HomeLink HA Interface last night with mixed success.  I first of all did the tests in HA #2, followed by connecting everything up to my breadboard test setup.  The overall setup that I am going for looks like this:

The converter board will actually be a breadboard connected to the RPiZW gpio pins.  I am planning to initially mount the Down Converter (Buck) to the same breadboard.  However, I need to get everything set up first before moving the parts to the RPiZW breadboard.  So, the initial physical setup looks like this:

The RPiZW is rather small compared to the LiftMaster 850LM, so piggy backing everything on one breadboard for the prototype is probably ok.  I also need to figure out a model to print from my 3D printer to contain everything.

HomeLink HA Interface #2 - Experiments with the LiftMaster 850LM

So I decided to go ahead and test out the LiftMaster 850LM to see what I was up against. This device will become central to how I am going to interface my home automation system to my car.  I have always been a little upset that there were no interface devices to link my car to other than a simple switch or a garage door.  That is shallow thinking on the part of the automotive industry and the people who make garage door devices.  Anyway I digress.

The 850LM is going to be used to detect button clicks from my car and convert those clicks to MQTT messages that will be sent to my home-assistant controller via an MQTT broker. I expect to interface this on a Raspberry Pi Zero W (RPi0W) using its gpio pins.  However, there are a couple of issues that need to be resolved. The gpio pins on the RPi take 3.3v and the lowest voltage DC that the 850LM can handle is 9v.  Fortunately, I noticed previously that the Adafruit bi-directional level shifter can take a high side voltage up to 10vDC.  That means that I can use the level shifter between the 850LM and the RPi0W. The main reason that I want to use the RPi0W for this interface is that I can run the software under a Docker container, and allow for another container to House a Bluetooth proximity sensor / MQTT setup. Note that the RPi0W needs a 5v DC power source, but generates its own 3.3v power for the gpio.

The 850LM has a power port as well as a screw connection marked plus and minus. It has a total of three relay connections; the first two having NO / GND screw terminals and the third having NO / GND / NC screw terminals.  So my first question was how do the power port and power screw terminals relate.  The second question was when the relays switch is it momentary (governed by how long you press the button on the car), is it momentary (fixed close then open time), or is it more of an on / off switch (changes as you click the car button). Fortunately, I should be able to use a remote control to do the testing.

Test 1: I powered up the 850LM using a 5v to 9v usb cable that I had lying around by plugging it into the power port.  I had to use an adapter because the center pin on the power port was thick (white size). Once powered up, I had no problem setting up the remote control by following the 850LM instructions (click yellow button on 850LM, led next to button flashes, click button on remote, led goes out).  Once I did this, a click on the remote caused the relay to close and then open on its own.  So I answered the second question.  It makes sense because most garage door controllers have wall switches that momentarily close a connection which starts the garage door motor running.

Test 2: I then checked the power terminal pins and discovered that there was on voltage in the millivolt range.  This means that I cannot power the 850LM and draw power from these terminals when the 9v supply is plugged into the power connection.

Test 3: while clicking on the remote, I checked the voltage levels appearing between the NO / GND terminals.  This did not measure any change from zero volts.  This means that I will need to use a pull-up resistor for transitioning on the level shifter.  It also means that I will need to deal with negative logic; high showing up on the gpio means False / 0 / not clicked, and low means True / 1 / clicked.  I might end up using interrupts instead of running everything through a loop in order to not interfere with other container programs.

Test 4: I connected my 9v power source to the power screw terminals. I checked the on / off with my remote and everything worked out just fine.  So I will be connecting the 9v source (via a barrel to screw terminal adapter) to these terminals, along with the input side of a Buck converter (to give 5v from the 9v source) and some connection wires to power up the high side of the level shifter.  What I may do instead is put everything on a breadboard connected to the RPi0W gpio and minimize the connections needed (use the breadboard as the connection points for all of it).

I wonder if I can mount the Buck converter on the breadboard as well, that would save some space for me.  I am also concerned that the Buck converter might interfere with the functioning of the RPi, but I will have to test that out as everything is built.