Sampling an AC signal

I have a AC signal coming from an outside source. The signal is 18vac and is triggered with a button. What would be the recommended way to test this signal if it has been triggered with 3v3 micro? Should i just create rectified bridge and zener to max out voltage at 3v3 or 3v... or should i do some other magic?

It depends what the signal is, particularly what frequency it is, and how fast you need to respond to it - both to pressing the button and to releasing the button.

From a minimal hardware point of view a simple series resistor would do the job - basically a mains zero crossing type design, just for a lower voltage (use a lower value resistor).

That helped a ton.. found the H11AA1 which is an Optocoupler, Phototransistor Output, AC Input, with Base Connection .. the info :

DESCRIPTION The H11AA1 is a bi-directional input optically coupled isolator consisting of two inverse parallel gallium arsenide infrared LEDs coupled to a silicon NPN phototransistor in a 6 pin DIP package. The H11AA1 has a minimum CTR of 20 %, a CTR symmetry of 1:3 and is designed for applications requiring detection or monitoring of AC signals.

Click to expand...

Seems like this little 6pin IC is just what i need. Thanks Nigel!

BTW i know itll be pulled up to my 3v3 supply instead

I was going to suggest an AC optocoupler as an easy means. You get isolation from the nasty outside world, and the delay will never be as much as half an AC cycle.

Sounds like a doorbell. https://www.broadcom.com/products/o...n/isolated-voltage-current-detector/hcpl-3760

KeepItSimpleStupid said:

Sounds like a doorbell. https://www.broadcom.com/products/o...n/isolated-voltage-current-detector/hcpl-3760

Click to expand...

Expensive way to eliminate the need for a couple resistors. Those chips are $5+.

It's only 18VAC? Opticoupler is complete overkill. You can use a resistor (one technique is to use a few in series in case one somehow shorts). Most modern IC's have clamp diodes so will protect the pin from reverse or >Vdd, voltage build-up. Use high resistors to ensure the clamp current doesn't exceed 10mA which is generally a maximum.

KeepItSimpleStupid said:

Sounds like a doorbell. https://www.broadcom.com/products/o...n/isolated-voltage-current-detector/hcpl-3760

Click to expand...

Ding ding. It is

Visitor said:

Expensive way to eliminate the need for a couple resistors. Those chips are $5+.

Click to expand...

Less than a buck:

ACharnley said:

It's only 18VAC? Opticoupler is complete overkill. You can use a resistor (one technique is to use a few in series in case one somehow shorts). Most modern IC's have clamp diodes so will protect the pin from reverse or >Vdd, voltage build-up. Use high resistors to ensure the clamp current doesn't exceed 10mA which is generally a maximum.

Click to expand...

Seems not to overkill if i plan on selling these. I rather it protect the user 1000x more than a resistor

Visitor said:

Expensive way to eliminate the need for a couple resistors. Those chips are $5+.

Click to expand...

Just to be clear, I meant the Boardcom chip ($5+, over $11 at Mouser), compared to an AC optocoupler and two resistors.

Visitor said:

Just to be clear, I meant the Boardcom chip ($5+, over $11 at Mouser), compared to an AC optocoupler and two resistors.

Click to expand...

Ah ok lol sorry yeah.. thats crazy.

Doorbells are getting wierd though and causing issues with other products:
Say you want a Ring Doorbell to activate a real solenoid doorbell
(Ring does power stealing) -or-
You have a lighted doorbell - LED inside of the standard doorbell.
-or-
You have this chime doorbell that draws next to nothing.

You ask a manufacturer about their doorbell mute interface for their radios and they don;t know anything about it.

Another "doorbell interface" senses the change in magnetic field of the chime. What chime on the synthesized speaker thing

I eventually want to do some interfacing to a doorbell.

Eventually, or I did want, to use an alarm quality wireless transmitter for a back door doorbell. That was before RING or August Lock. Maybe, it's OK? There is no wiring to the back door. It was almost ready to go.

I made a home made lighted doorbell for the front door. It died. A solder joint failure. I haven't fixed it yet except replaced the button. Now people knock on the door because they don't see the button. So, it's knock on outer scrren door, knock using a knocker or ring the doorbell that you can't see because it's unlighted.

In reality, I'd like to mute the dumbest "Smart TV" from Samsung. It uses an IR/bluetooth remote and you can only hae one real one. A cell phone app probably works. Then there is "Smartthings".

Reality, I want to mute the TV when lots of things happen:

phone rings
You make a phone call
front or back doorbell
"nurse call" gets activated.

I've been wanting to play with the HDMI-CEC bus, but haven;t had the time. Now having the TV come on when something like the ring doorbell is used would be awesome.

I've been trying to figure out a way to play "audio" throught the TV speakers. I though that would be easy,.

KeepItSimpleStupid said:

Doorbells are getting wierd though and causing issues with other products:
......

Click to expand...

I know what you mean.Especially about that Audio stuff. I love that idea. All my tvs are ROKU with web interface so i can actually do volume control via web app and mute all tvs in house... Look into it its pretty cool. TVs are cheap now also.

I need a new phone. I have like android 2. Nothing works. Muting all of the TV's is pretty cool.

There's lots of things I want to mute automagically, but there is no interface yet. Doorbell isn;t easy. Neither is nurse call.
I do have a land line phone. Belman and Symphon has a device that sits on your cell and can activate theor system, but I think it's a contact closure. It detects the screen lighting up when you get a call.

What really pisses me off is the audio stuff especially whoever invented HDMI. The TV spits in my face when I want to play audio only. It even won't work through the AV inputs. Why not audio with a "black raster" or better yet "AV" moving to various positions on a black screen.

So, yea trying to figure out how to play audio only on a TV without regard to fidelity. I do have a plan.

Some time ago I was looking for a phone indicator that would illuminate if the phone was off hook and not talking for 15 minutes. I had to settle for off-hook, check voice mail (don't have) and ring (also wanted)

Some cordless phone will answer your cell via bluetooth which is cool.

If your looking for something else to design how about the ring/off-hook thing. You would probably have to use a DAA.

and for those playing around, detecting ring and or a message with some sort of device would be cool.
Maybe message via RS232 or the AT type commands. and Ring and CID from the AT type commands.
The hard stuff like BT to phone would be done by you.

I've just done a doorbell interface for someone, to an auxiliary input on a security system.

I ended up with one of these from Amazon; less than the cost of the Omron relays we keep in stock..

The input is opto isolated and rated for 12V so a small bridge rec is all that is needed to feed it from around 12V AC from a doorbell solenoid.
It is retriggerable so holds in as long as power is applied to the opto circuit, then drops out after the set delay. I have the time delay set for about one second.

Probably overkill for a direct input on an MCU based device, but perfect for many other types of interface or control (like the radio mute someone mentioned).

rjenkinsgb The problem would be in my case the doorbell always sees 1/2 wave rectified 16 VAC. Why? the doorbell switch has a LED across it.

KeepItSimpleStupid said:

rjenkinsgb The problem would be in my case the doorbell always sees 1/2 wave rectified 16 VAC. Why? the doorbell switch has a LED across it.

Click to expand...

Not when the button is pressed. The LED is shorted out.

So, your putting the sensor across the dool bell switch and not the doorbell coil?

Now take a chime or the "Ring doorbell" using an external bell.

Ah yes, I stand corrected.

When the button is pressed, the coil will see full-wave AC. When the button isn't pressed, there will be a current-limited (by the LED resistor) sine wave. But what will the voltage across the coil at that point?