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

ote:

ail.com:

ut

atic worked perfectly. I'm still experimenting with resistors to get the c urrent to the button down, and to achieve the correct voltage out for the a rduino - but it's more success than I had on my own. The arduino doesn't r eset on button press, and as described - voltage drops to zero when the but ton is pressed.

sed, a piece flew in her direction and she nearly jumped out of her skin. "Are you trying to burn the house down??" she yells.

was experimenting to determine the minimum current necessary to keep the b ulb lit (and then maybe raise it a little higher to accommodate power fluct uations).. Apparently I started too low and/or held the button too long. My earlier tests, I only held the button for less than a second.

avoid future pyrotechnics? Am I mistaken as to the cause - do I need a mo re robust heat sink, was it heat related? I'm thinking my next attempt wil l be at 240ohm, but I don't want to get yelled at twice for the same thing.

r is

atts

or

tty

ping

it

the

se

insulated jumpers, and I've never had a problem with this breadboard before .

LM317 exploded and most of those came to the conclusion that it was a faul ty unit.

from the LM317. Perhaps I should start high and work my way down until th e lamp lights instead of starting low and waiting for the lamp to go out.

compensation.

low

Backwards, I suppose that's possible. With the heat sink mount facing me ( label facing away) and pins down, (3)vin - (2)vout - (1)adj.

wonghunglow? I have no idea, but if that's a joke - it's hilarious!

The markings are: LM317T CC12CVW MAR423 Some weird glyph on the left and a circle containing e3

ote:

ail.com:

ut

atic worked perfectly. I'm still experimenting with resistors to get the c urrent to the button down, and to achieve the correct voltage out for the a rduino - but it's more success than I had on my own. The arduino doesn't r eset on button press, and as described - voltage drops to zero when the but ton is pressed.

sed, a piece flew in her direction and she nearly jumped out of her skin. "Are you trying to burn the house down??" she yells.

was experimenting to determine the minimum current necessary to keep the b ulb lit (and then maybe raise it a little higher to accommodate power fluct uations).. Apparently I started too low and/or held the button too long. My earlier tests, I only held the button for less than a second.

avoid future pyrotechnics? Am I mistaken as to the cause - do I need a mo re robust heat sink, was it heat related? I'm thinking my next attempt wil l be at 240ohm, but I don't want to get yelled at twice for the same thing.

r is

atts

or

tty

ping

it

the

se

insulated jumpers, and I've never had a problem with this breadboard before .

LM317 exploded and most of those came to the conclusion that it was a faul ty unit.

from the LM317. Perhaps I should start high and work my way down until th e lamp lights instead of starting low and waiting for the lamp to go out.

compensation.

low

heat sink mounting plate facing me, pins down - (3)vin, (2)vout, (1)adj

"wonghunglow"? I have no idea, but if that's a joke - it's hilarious. Her e's a photo of the LM317 I received:

e:

l.com:

ic worked perfectly. I'm still experimenting with resistors to get the cur rent to the button down, and to achieve the correct voltage out for the ard uino - but it's more success than I had on my own. The arduino doesn't res et on button press, and as described - voltage drops to zero when the butto n is pressed.

d, a piece flew in her direction and she nearly jumped out of her skin. "A re you trying to burn the house down??" she yells.

as experimenting to determine the minimum current necessary to keep the bul b lit (and then maybe raise it a little higher to accommodate power fluctua tions).. Apparently I started too low and/or held the button too long. My earlier tests, I only held the button for less than a second.

void future pyrotechnics? Am I mistaken as to the cause - do I need a more robust heat sink, was it heat related? I'm thinking my next attempt will be at 240ohm, but I don't want to get yelled at twice for the same thing.

is

ts

ng

he

sulated jumpers, and I've never had a problem with this breadboard before.

M317 exploded and most of those came to the conclusion that it was a faulty unit.

rom the LM317. Perhaps I should start high and work my way down until the lamp lights instead of starting low and waiting for the lamp to go out.

The bulb current is 24mA

In the upper left hand corner of the schematic Lasse provided you can see the input of the LM317 is connected to 12 volts DC from the supply you have.

Sounds like you might have connected the LM317 to the doorbell AC supply instead. If so, KABLAM! The LM317 will be _very_ unhappy if the supply is AC.

Ed

com:

worked perfectly. I'm still experimenting with resistors to get the curre nt to the button down, and to achieve the correct voltage out for the ardui no - but it's more success than I had on my own. The arduino doesn't reset on button press, and as described - voltage drops to zero when the button is pressed.

a piece flew in her direction and she nearly jumped out of her skin. "Are you trying to burn the house down??" she yells.

experimenting to determine the minimum current necessary to keep the bulb lit (and then maybe raise it a little higher to accommodate power fluctuati ons). Apparently I started too low and/or held the button too long. My ea rlier tests, I only held the button for less than a second.

id future pyrotechnics? Am I mistaken as to the cause - do I need a more r obust heat sink, was it heat related? I'm thinking my next attempt will be at 240ohm, but I don't want to get yelled at twice for the same thing.

I may be stupid, but I'm not THAT stupid :P

It's connected to the 12v side of a 12v2a/5v6a regulated power supply that I purchased for this project.

Den torsdag den 11. december 2014 19.22.16 UTC+1 skrev Michael Richey:

gmail.com:

tput

ematic worked perfectly. I'm still experimenting with resistors to get the current to the button down, and to achieve the correct voltage out for the arduino - but it's more success than I had on my own. The arduino doesn't reset on button press, and as described - voltage drops to zero when the b utton is pressed.

eased, a piece flew in her direction and she nearly jumped out of her skin. "Are you trying to burn the house down??" she yells.

I was experimenting to determine the minimum current necessary to keep the bulb lit (and then maybe raise it a little higher to accommodate power flu ctuations).. Apparently I started too low and/or held the button too long. My earlier tests, I only held the button for less than a second.

to avoid future pyrotechnics? Am I mistaken as to the cause - do I need a more robust heat sink, was it heat related? I'm thinking my next attempt w ill be at 240ohm, but I don't want to get yelled at twice for the same thin g.

tor is

the

Watts

? I

in

or or

retty

er

n?

opping

e it

you

ng the

else

ot a

d insulated jumpers, and I've never had a problem with this breadboard befo re.

wn"

en

for

a LM317 exploded and most of those came to the conclusion that it was a fa ulty unit.

nt from the LM317. Perhaps I should start high and work my way down until the lamp lights instead of starting low and waiting for the lamp to go out.

he

ea compensation.

nglow

(label facing away) and pins down, (3)vin - (2)vout - (1)adj.

-Lasse

I find your description of the pin numbering a bit confusing, but your picture is clear. So are the pins from left to right in your picture like this?

That is what the data sheet shows. It also shows this as a fairly robust device. Hard to blow it up. I stand my my earlier opinion that something was wrong with the wiring although it is possible the part was faulty, just not likely. :)

Was that in the original circuit or on the 12 volt supply?

What was the voltage on the bulb in the original circuit?

So what resistor should you use with the LM317?

from the LM317. Perhaps I should start high and work my way down until th e lamp lights instead of starting low and waiting for the lamp to go out.

I'm working with an off-the-rack/standard lighted button I purchased for te sting and I measured the 24mA with the bulb/button connected directly to th e power supply. There are probably minor variations in current requirement s for the bulbs from different manufacturers.

Voltage for the bulb in a normal doorbell is 12, 16 or 24vac (this bulb is designated 12vac) and I'm building the doorbell for my brother who has a 12 vac system. The bulb lights perfectly well on 12vdc.

rote:

mail.com:

put

matic worked perfectly. I'm still experimenting with resistors to get the current to the button down, and to achieve the correct voltage out for the arduino - but it's more success than I had on my own. The arduino doesn't reset on button press, and as described - voltage drops to zero when the bu tton is pressed.

ased, a piece flew in her direction and she nearly jumped out of her skin. "Are you trying to burn the house down??" she yells.

I was experimenting to determine the minimum current necessary to keep the bulb lit (and then maybe raise it a little higher to accommodate power fluc tuations).. Apparently I started too low and/or held the button too long. My earlier tests, I only held the button for less than a second.

o avoid future pyrotechnics? Am I mistaken as to the cause - do I need a m ore robust heat sink, was it heat related? I'm thinking my next attempt wi ll be at 240ohm, but I don't want to get yelled at twice for the same thing .

or is

he

Watts

in

r or

etty

pping

it

ou

g the

lse

t a

insulated jumpers, and I've never had a problem with this breadboard befor e..

wn"

en

for

a LM317 exploded and most of those came to the conclusion that it was a fau lty unit.

t from the LM317. Perhaps I should start high and work my way down until t he lamp lights instead of starting low and waiting for the lamp to go out.

he

rea compensation.

unglow

me (label facing away) and pins down, (3)vin - (2)vout - (1)adj.

My description is reverse of the picture (where the heat sink mount is faci ng away). I was trying to describe orientation before labeling the pins.

I think I'm not being clear. I am not trying to characterize the door bell button alone. I am trying to characterize the door bell button in the circuit it is designed for.

In order to not burn out the transformer when the door bell button is pushed, the transformer must have some resistance to limit the current. I believe the door bell coil is also in series with the switch and has significant resistance. So when the light is lit, it is not at the full

The way the LM317 works is to drop some of the voltage across the input to output pins to limit the voltage across the resistor to 1.25 volts. So in your circuit the door bell button will not see the full 12 volts either. It will see the 12 volts minus the 1.25 volts across the resistor minus whatever the minimum drop out voltage is on the regulator. Looks like at room temperature you will see more than 1.5 volts across the LM317. So the doorbell will see at most 9.25 volts provided the LM317 is still in regulation which is not strictly necessary until the button is pressed.

The point is that the doorbell button works at some voltage less than 12 volts in the original circuit. In your circuit it will work at some voltage less than 12 volts. It would be useful to try to match the two voltages or at least find out if the voltage the LM317 will provide will light the lamp enough. Easier to do on paper than by blowing up regulators.

r testing and I measured the 24mA with the bulb/button connected directly t o the power supply. There are probably minor variations in current require ments for the bulbs from different manufacturers.

is designated 12vac) and I'm building the doorbell for my brother who has a 12vac system. The bulb lights perfectly well on 12vdc.

I'm grateful for the help, and if you can describe a scenario to determine that voltage using a multimeter - I'll be happy to get that information.

Go to the house where you plan to install the doorbell and measure the voltage across the button. Maybe this isn't really important and I am being a bit obsessive about it. I would like to know the details before I start on something like this. But experimenting is not a ridiculous way to do it.

for testing and I measured the 24mA with the bulb/button connected directly to the power supply. There are probably minor variations in current requi rements for the bulbs from different manufacturers.

lb is designated 12vac) and I'm building the doorbell for my brother who ha s a 12vac system. The bulb lights perfectly well on 12vdc.

has

ll

'm

ll

ine that voltage using a multimeter - I'll be happy to get that information .

The problem with that is, the currently installed doorbell is using a 12vac power supply that won't be present in the one I'm building.

My system needs 12vdc (button light + arduino) and 5vdc (4 servos)

I understand that. My point is that the light bulb likely *isn't* working at 12 volts in the present system and that is the light bulb you will be designing your new system for, no? But like I said, I'm probably being a bit compulsive about this and the brightness of the light isn't so important as long as it lights.

You do understand that your current limiting circuit will only drive the bulb at something like 9 volts, right? That likely will still be visible at night.

Funny, if you replace the bulb with an LED the whole problem goes away. A simple current limiting resistor goes between the switch/LED and the PSU so not much more current will flow in either case, button pressed or not. You also won't need to do anything to interface it to the rPi since the voltage swing will only be about 2 volts to ground. But I know you said you want to stick with standard door bell buttons for a very good reason. Heck, you could even drive it with 3.3 volts rather than 12.

:-) I didn't have stupid in mind - just a simple mis-wiring. When you have a bunch of wires, it's easy to mis-wire to the wrong one. Ask me how I know. :-(

From the description of the LM317 blowing up, that kind of mis-wiring seemed likely.

You provided additional information: the bulb is rated 12v at 24 mA. To get ~24 mA out of the LM317 use a 51 ohm resistor between LM317 out and adj pins. You won't get 12 volts on the output pin, but you may get enough to light the bulb. If it glows to your satisfaction, you're done. If not, read on.

Regulators like the LM317 require what is often called "headroom". Headroom is a voltage at least 2 to 3 volts higher at the input than the desired output voltage. In your case with 12 volts input, you can't get 12 volts output, and it's possible that whatever voltage you do get isn't enough to light the bulb to your satisfaction.

If you're using a 12V regulated supply and a 12V, 24 mA bulb that doesn't glow bright enough, you don't need - or want - the LM317. Replace it with a 10 ohm 25 watt resistor:

Will the doorbell button handle 1 amp of current without eventually pitting and burning out? I expect these are the world's cheapest switches.

I expect the resistance of the doorbell coil in the existing installation is much more than 10 ohms so that the lamp does not see anything like the full 12 volts. But even at 9 volts the lamp likely lights enough for a door bell button. So the dropout of the regulator is not likely a factor. In fact, if he designs it for the current the bulb passes at 12 volts, the regulator will be out of regulation because of the lower current and so the drop out voltage will be even less than the spec 1.5 volts. That's ok since it only needs to regulate when the button is pushed.

I think I am being overly cautious to even worry about this. But we will see how his circuit goes.

----------. about 14V 22mA 12V | --> - AC .-->|--+-->|--+---------[LM317]--[56R]--. . . . | . . . | | | | . --- . ---|------. | `------------+----+-----+--o o--+---. | | | | . | | . | +--|

There are only two ways to kill an LM317: (1) input overvoltage, or (2) large reverse current, due e.g. to shorting the input when there's a big cap on the output.

Otherwise they're essentially bulletproof. To make one disassemble itself like that, you probably had 1000 uF on the output and shorted the input, or something like that.

Cheers

Phil Hobbs