I'm wondering if an Arduino with the appropriate frequency rf transmitter can open garage doors that use rolling codes (new code every button press). I am talking about my own garage door, which has remotes, I have access to the learn button, etc.
My concern is if the programming is some high level stuff that figuring out would require essentially hacking a protocol aimed at security.
At the brainstorming stage, the ghetto approach would be to simply use a real paired remote, and use the Arduino to "press" the button, either mechanically or open it up and incorporate a relay/transistor/something to trip whatever a button press trips, or permanently connect the button and pulse it at the battery contacts.
But, let's see what my options are. Can an Arduino with rf transmitter be learned by a garage door opener?
Eventual end goal is to be able to open/close the garage door and see if it is opened or closed via phone over wifi. I still have a lot of learning to do for the wifi part, but that's another topic.
It sound like a challenge response type system and to integrate with that you would probably need the cryptographic algorithm and duplicate the key (seed) exchange mechanism used during the pairing.
Another approach would be to drop the existing radio / security parts of the door mechanism, replacing thes with your own.
Example Arduino Garage Door Opener Is Security Minded | Hackaday
Anyway, I'm abandoning the remote avenue. I took apart the control panel inside the garage, that I somehow totally completely forgot about, that is hardwired to the opener unit, and there's a simple microswitch button that opens/closes the garage door by completing a junction between two traces/solder points. Will be stupid simple to run a parallel switch for the Arduino to work with.
Actually, I'm pretty dumb- if I put a transistor across those button leads (collector and emitter, base to the Arduino) then what info do I need to use the right transistor? And is a transistor even the best solution here?
For nice, clean separation without having to understand too much of the existing mechanism, I'd opt for a relay module. You are then simply emulating a microswitch and completely isolated from the rest of the door open/close mechanism.
An opto coupler may work as well. Check the voltage on the button pins, and the current between both. Take care when measuring the current, this will activate the doors.
The button is passing 16VDC @ 30mA. A relay seems overkill.
I understand the wall controller switch. It's using a simple 2 conductor bell wire to run to the main unit. But it also has a 'Light" microswitch as well as a "Lock" microswitch, which I'm guessing pulse a duration below some threshold before the "open" command. I don't have an oscope with me to actually see in detail.
INTP:
The button is passing 16VDC @ 30mA. A relay seems overkill.
From an engineering point of view, I agree with you. From a practical point of view, and bearing in mind that 5volt relay modules can be had for $0.99 I'd tend towards a simple solution where the existing environment does not have to be looked at in any detail.
I have a relay, but I also have an SS8050 transistor that came with my starter kit. Even if that isn't correctly spec'd, and cost is your point, transistors are much much cheaper than relay units.
I really need you to clarify how you see a bigger relay as a simpler solution, and what you mean with this "existing environment" understanding detail etc
And if I want to give the Arduino the ability to mimic the "Light" and "Lock" commands, a relay will not do as it will possibly not pulse fast enough, being a mechanical doohickey.
OK. The relay module in my post has 4 relays for a price of $0.99. If it can be done with a microswitch, it can be done with a relay. I'm certainly not saying using a relay is cheaper than using a transistor. It is almost certainly not. All I am saying is that the extra cost of the relay module is so small that it hardly matters.
Anyway, it is your project. You sought opinions and have got them. I wish you luck with it. It sounds interesting and will no doubt bring a useful result.
Your responses are intriguing me and I am seeking clarification not as a challenge, but as a desire to learn because what you are saying isn't jiving with my understanding, and I'm the kind of person who always questions my own understanding before doubting others.
How is a relay a 'simpler solution' than a transistor? Is there more to using the transistor than I am aware of?
A relay offers perfect isolation of both circuits, and certainly acts like a mechanical contact. You don't have to worry much about AC/DC, voltage and current, if you want to use it instead of or in parallel to another mechanical contact. But it requires some activation power, often more than a digital pin can supply, and has a limited lifetime.
Reed relays are different, of course, should be used with care.
A typical optocoupler supports only DC of the right polarity, limited current and voltage. But it doesn't have moving parts, resulting in low driving power and a long lifetime. SSR are similar to optocouplers, also for AC and high power loads.
Your idea with a transistor is critical, because it would require direct contact between both circuits, with possible shortcuts or mains exposure.
Something would have to go terribly terribly wrong for mains exposure, like having the garage on fire and then sprayed with a fire hose.
Opener motor unit is plugged into a mains outlet. On the motor unit are some spring clip terminals for the DC voltage to and from the two safety sensors (a sender and a receiver unit on either side of the garage opening to detect obstacles) and to/from the wall mounted button unit. All of these DC cables are scrawny 22-24AWG bell wire.
So an optocoupler is just an LED and something like a light dependent resistor, just super shrunken down.
Nice. I imagine a slower max switching frequency than a transistor. Maybe doesn't matter.
This controller has 3 momentary microswitch type buttons. The main one that opens/closes the garage is a simple short between the two wires running to the controller. No components involved.
A 'Lock' button has a 22uf cap in series between the two wires, and a 'Light' button has a 1uf cap in series between the 2 wires. As this is all DC, I guess the main unit recognizes the slight dip in voltage as a signal from one of these accessory buttons.
Is it possible to fake that kind of thing, or would I be looking at 3 transistors or optocouplers to make use of the existing hardware?
I guess that the caps stretch the button signals, i.e. the 'light' will stay on for some time. You can test yourself, by shorting the pins.
But these caps defeat the use of optocouplers, because they supply a high discharge current. In so far I'd second the idea of a 4-relay board. But the caps also will melt reed contacts, so the relay contacts should be rated for high current (1A) and mains voltage, to definitely exclude reed contacts!
I'll just clear these 2 points up, but anyway in the meantime you've had good answers from Dr.Diettrich
INTP:
. . .
I really need you to clarify how you see a bigger relay as a simpler solution, and what you mean with this "existing environment" understanding detail etc
. . .
INTP:
. . .
How is a relay a 'simpler solution' than a transistor? Is there more to using the transistor than I am aware of?
"existing environment" in this context meant the existing electronics of the garage door control unit.
I consider the relay to be simpler because there is a one to one relationship between relay contacts and the micro-switches you are emulating/replacing. The transistor is naturally smaller but the integration of a transistor circuit in an existing circuit requires a considerable knowledge of the characteristics of that circuit and then you have the problem of driving the transistor from an Arduino and, therefore requiring additional components to isolate it.
If the system is being designed from scratch, then transistors would be preferable to relays. If you are integrating with an existing (unknown/undocumented) system and want the least risk approach, I still would choose the relay.
Incidentally, the relay modules in the example link claim to have 10 Amp contacts.
Okay, I understand that you don't know me or the simple circuit involved (literally the simplest commercial pcb I have ever seen, only a single layer of traces), and are advising according to being super cautious and doing things in a roundabout way as a result.
But if I left out the context of a big mean scary garage door opener machine, and simply asked how to control a SPST switch of 16VDC @ 30mA with the Arduino, would you still be pushing the whole relay getup, even when just the tiny optocoupler on that board is all that is needed, or again a transistor could work if circuit isolation is unnecessary?
The transistor option would leave out the cap routes, I have no need to turn on the garage lights when I'm not there =D
But purely for academic understanding, what kind of spike are we talking about?
16VDC to a 1uf cap and momentary button in series. Pushing button briefly passes current as cap fills up to cut off that current, thus timing the signal to a short pulse. Then what happens to that cap charge?
