Probably trivial, but I'm a programmer - not a ee

I started to write a long screed -- but why? Google "electronic doorbell" and see if something suitable comes up. If it doesn't -- ask here again. If it does, but you don't understand some point or another -- ask here, with a link to the specific design in your post.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

4 servos which actuate 4 mallets on 4 chimes to play a melody. I have the arduino and servos operating as expected, interfacing a standard doorbell button is the trick I just can't seem to figure out.

My goal is to make it a drop-in doorbell, using the same wiring as any doorbell you might buy at home depot. This means I have 5 wires to interface (7 if using a button for back door chime). 3 wires for AC power,

2 wires per doorbell button.

For power, I have a 12v2a/5v6a regulated power supply.

Doorbell buttons come in two flavors, lighted and non-lighted. Interfacing to a non-lighted button is trivial. It's just a regular switch and I can push 5v to capture that input. Lighted buttons normally run on 12vac, but will also light on 12vdc. Unfortunately, the bulb is wired in parallel with the button - so there's constant current. Additionally, the arduino can't take 12v on the inputs - so there's no way for me to use a lighted button without some magic.

I have an alternative that I really don't want to pursue. I could reconfigure the lighted button with a resistor and LED and run the whole thing on 3.3v from the arduino. The reason I don't want to do this is because - as a gift - I want this to be as transparent to the recipient as possible. If at some point the button needs to be replaced, the new button would need to be made compatible.

Another issue I've run into is, when the arduino and button are connected to the same source - pressing the button drops voltage to the arduino - causing it to reset. So the button circuit needs to be isolated from the arduino.

So, if you were forced to use a 12v lighted doorbell button as trigger to a

5vdc device - what would you do?

+++++++++++++++++++++++++++++++++++++++++++++++++++++++=

Use a 12 volt AC (or DC) relay?

Are you sure about the 12 volts? My doorbell uses 24 volts AC.

I've been looking at this for days (and doing lots google searches). Most have wonderful writeups on modifying wireless doorbells to trigger an actio n on an arduino, others detail the steps to convert a 12v lighted doorbell to 3.3v using a resistor and LED. Still others say how easy it is, while u sing a non-lighted button (which is, after all, just a button).

Again, I'm no EE - I'm also not ashamed that I don't know what I'm doing or to ask for help.

Isolating the power, I considered a pair of zener diodes. I also considere d using a relay to switch 5v for the arduino input. I have also considered using resistors as voltage divider taking the 12v down to 4.8v, but I ende d up deciding that wasn't viable because the bulb ensures that constant cur rent is going through the button.... I just don't know which way to go or where to look.

This is the third place I've asked. The first and only response I received to my question was that the light would operate fine on 12vdc.

I would take the easy way out and rewire the button using a resistor and LE D, but that would just be the easy way out for me - the person receiving th e gift would then have to deal with the same problem if they ever replaced their button.

If you can help, or even point me in the right direction - I'm prepared to read your long screed.

I've seen 12, 16, and 24vac doorbells. The gift is for my brother, and his is 12vac.

Since the doorbell button "looks like" a short circuit when it's pressed, yes, you'll need to isolate it from the +12 supply to keep from crowbarring the Arduino. Either a series resistor, or some sort of constant-current limiter (e.g. an LM317 wired up properly) would do the trick. The latter might be a better choice, since you could set the current limit to just a bit above what the button needs to be fully illuminated... this would minimize the change-in-load on the power supply when the button is pressed.

You could then "tap off" the voltage right after the current limiter (before it goes out to the button), and use a two-resistor ladder to attenuate it down to what the Arduino input will accept. Done correct, you should be able to "see" 3.3V at the Arduino input when the button is released, and 0V (or very close to it) when the button is pressed and the output of the current limiter is shorted to ground.

So - one LM317, and three resistors (one to make the current limiter... size it to drop 1.2 volts at the button lamp's current demand... and two to step down the voltage to 3.3 volts to keep the Arduino happy) and you'd be good to go.

For a drop-in system, leave the door button(s) connected to the 12 VAC and use an optical coupler (one for each button for separate tunes) in place of the existing doorbell. You can parallel the optical coupler with a resistor so a button light's current would not trigger it. A series diode between the resistor and the coupler would provide DC if the coupler requires it. A series resistor can be used to adjust current.

Fred

Den torsdag den 4. december 2014 21.40.38 UTC+1 skrev snipped-for-privacy@gmail.com:

t have wonderful writeups on modifying wireless doorbells to trigger an act ion on an arduino, others detail the steps to convert a 12v lighted doorbel l to 3.3v using a resistor and LED. Still others say how easy it is, while using a non-lighted button (which is, after all, just a button).

or to ask for help.

red using a relay to switch 5v for the arduino input. I have also consider ed using resistors as voltage divider taking the 12v down to 4.8v, but I en ded up deciding that wasn't viable because the bulb ensures that constant c urrent is going through the button.... I just don't know which way to go o r where to look.

ed to my question was that the light would operate fine on 12vdc.

LED, but that would just be the easy way out for me - the person receiving the gift would then have to deal with the same problem if they ever replace d their button.

o read your long screed.

+12V

I was about to ask Dave Platt if he could diagram his suggestion for me, but it sounds a lot like what you've diagrammed here.

I'll get my hands on a few LM317 and give this a try.

Thank you, and David Platt for describing a solution.

One more question, regarding the LM317

How much heat can I expect that to put out? Can I get away with fixing it to the thin aluminum sheet of the electronics enclosure, or will I need to consider a more robust heat sink and ventilation?

Hey! Something like that was going to be part of my long screed! Except I wouldn't have thought of using an LM317, which certainly reduces the amount of work required.

I would be inclined to add a transistor inverter at the junction of R2 and R3, to isolate the processor input more -- but what's there should work fine, and would definitely be what you'd find if the thing came from China.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

Post your results for the next guy, when it's working?

I didn't mean to be unhelpful -- I just figured that someone should have done it before, and posted information in excruciating detail.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

If they did, I couldn't find it. The closest I've found is the resistor/LED rewire.

I'm planning on a lot of pictures, diagrams, links to my resources (for bell creation/tuning), the 3d printed hammer hinge I designed (on shapeways) and now the schematic provided here (once I figure out the necessary resistors).

The ultimate goal is to have an artistic piece (so his wife will allow it to be installed) which plays a tune he's fond of (instead of the standard ding-dong or westminster that everyone has).

When complete, I'll post it on the instructable website.

R2 would already limit the current into ESD diode, though it isn't spec'd how much is allowed

-Lasse

If the wiring to the doorbell is long, transients can be picked up which can capacitively couple past the resistor into the input. Maybe this is not a significant effect, but using a transistor that can handle the 12 volts will also protect the input from an open R3. The ESD diodes are often pretty limited in the current they can handle.

--

Rick

Den torsdag den 4. december 2014 23.50.11 UTC+1 skrev snipped-for-privacy@gmail.com:

depends on how much current the bulb needs

assuming 200mA and 1.5V dropout = 0.3W, an LM317 in to220 it shouldn't need any heatsink

when you push the button the full ~12V*200mA = 2.4W is dissipated in the LM317 That would need heatsinking but since it is only for a short time and the LM317 will turn off if it gets too hot it would probably work with out

-Lasse

That gives me what I need to know. I'll go ahead and attach it to some aluminum channel, attached to the case. Small enough that it won't take up too much space in the case - large enough to dissipate some heat during button push.

Thank you again for your help!

When it's finished, I'll post a link to the instructable here.

Den fredag den 5. december 2014 01.31.58 UTC+1 skrev snipped-for-privacy@gmail.com:

just remember that the tap on an LM317 is connected to the output

-Lasse

good to know!

That's why they invented mica insulators and nylon bolts.

--

Rick