I need a vibration sensor + Tx/Rx modules that do not draw much current

Hello all,

I would like to build a (snail) mail notifier. My idea is to have an ATTiny85 (unless I need more pins than the ATTiny85 has) inside the mailbox with a vibration sensor and a Tx modules, and another ATTiny85 with an Rx module in the house. ATTiny81-M is the one in the mailbox, and ATTiny85-H is the one inside the house.

The way my mailbox works, the mailman has to push the cover of the slot with the mail to put it inside, so I would glue the vibration sensor there, and the rest of the circuit where it's more convenient. ATTiny85-M is in sleeping mode until the mailman pushes the mail inside, the sensor wakes up ATTiny85-M, which uses the Tx module to tell ATTiny85-H that there's mail, and goes back to sleep. Then ATTiny85-H turns on an LED or something (I haven't decided yet).

I already know how to do all this except for one little detail: Having ATTiny85-M in sleeping mode, and only waking up to notify mail and going back to sleep. The idea is to have a a coin cell battery powering the circuit, and make it last as long as possible. I'm using Arduino IDE, and an Arduino UNO board configured as an ISP to program both ATTiny85's.

Bonus: I'd also like ATTiny85-M to notify ATTiny85-H when the battery is running low. I don't know how to do that. I thought of using a solar cell on top of the mailbox and solve all problems, but that would be too conspicuous and might attract unwanted attention.

These are the modules I have my eye on:
Arduino Vibration Sensor
Arduino 315MHz Tx/Rx

If someone can point me in the right direction on how to have the ATTiny85 inside the mailbox to consume as little current as possible and to run off batteries as long as possible, I'll greatly appreciate it. Thanks a lot!

why not use a switch (rather than a vibration sensor? the vibration sensor you are considering consumes power; some sensors do not)
The arrival of the mail in the box triggers the switch that powers the circuit; it sends its information, then turns off by cutting its own power supply.

Thanks for your reply. I thought of using a switch. The problem is that there are many angle degrees in which you can push the slot door to drop mail. For a single letter, there might not be enough force exertion to activate the switch unless. I'm investigating how to do it this way, while, concurrently inquiring about using a sensor.

There are very sensitive switches, whose reaction can be increased with a suitable mechanical assembly. The advantage is that it does not consume anything in standby (nor the circuit which is stopped). A reed relay and a small magnet can also do the trick. Otherwise, the best vibration sensor (to do it yourself) is a piezoelectric disk (the one of the buzzers) associated with a cmos gate (4093) and a transistor to power the circuit. The piezo disk is very sensitive to vibrations, it provides a small voltage to the cmos gate which activates the transistor ; the consumption is very low, and does not affect the battery life.

This has been done already. Make a search for such projects.

Are you talking about a micro switch with a lever? I'll check the inside of the mailbox tomorrow to figure out what size of a micro switch I need. Thanks.

Yes, I actually saw a post about a mail notifier before posting this. But my specific question was about the current draw, not about how to do a mail notifier. Thanks.

Don't search for a complete match of Your project. It doesn't exist. Search for part solutions that You can use.

hi popcalent,
to amuse myself a long time ago, I made this shock detector (vibration) for a house door: when someone knocked on the door (more than 2 times), a sound signal was sent further in the house. My main objective was the reduced consumption, the circuit was powered by a 9 Volts LR9 battery, and should last several months. It could also have been used as a low consumption intruder alarm. I like everything that is low-power, and at the time, I could not find any sensor of very low consumption (I have the impression that today too?)

I used a piezo sensor, with a small weight to increase the sensitivity (you can't see the second electrical connection on the other side of the piezo, it's a thin copper wire like a hair that comes back on the circuit).

The circuit uses 2 NAND gates mounted in flip-flop, with 2 signals : in (B), you should put in a transistor that will activate your ATTiny85, and the activation of (A) by the ATTiny85 cuts its power supply (you have to filter this part, so that the atiny that wakes up doesn't stop the process, while its pins are in transistory states)

The circuit is extremely sensitive ! and its standby consumption is so low that it is not measurable with standard devices, lower than the natural wear of the battery.

Hi PatMax,

Thanks for sharing. I'll try implementing it. I also ordered some micro-switches and I'll try that too. However, I thought of a case that will ruin my plans. When business mail is shoved into the mailbox in a way that sticks out and keeps the slot door open. That would keep the circuit on and eventually drain the battery. I'm considering using a vibration switch. I've never used one before, but I imagine it's not the same as a vibration sensor. I guess it stays open and it closes with a vibration until the vibration is over. Am I right?

For a switch google “ wobble switch” these work over a range of angles and need no power .

A mercury tilt switch would be the most reliable and importantly, weatherproof.

But - are you not using a microprocessor?

It can turn itself off again, and if you are using INPUT_PULLUP or switching a lower value pullup from another pin, there need be no drain except for occasional timeouts to check the switch.

[quote="popcalent, post:10, topic:884951"]
Am I right? [/quote]
You are !
But I kind of forgot about all the possible variations of vibration sensors or switches... it seems to me that there should be a vibration switch, but I'm afraid that it runs on permanent power, which would not be well suited for your project. It's complicated the technique! even a mailbox detector, it's a lot of choices and tests to make :wink:

Thanks for your reply. These seem a little bit expensive, though.

Thanks! I'll order those! Yes, I'm using a micro controller. I'm not sure about how much current is drawn if I put it to sleep, though.

Great, thanks! I've seen modules with three pins (Vcc, Gnd, Out) and single devices with two pins. I imagine the former run on permanent power, and the latter are the ones that work as a switch. Someone recommended me this one.

This is another option I found: a magnetic sensor contact for doors that is normally open.