On Sunday, January 17, 2016 at 11:12:17 AM UTC-8, Micky wrote:
On Sun, 17 Jan 2016 02:38:08 -0700, Don Y
wrote:

On 1/16/2016 9:27 PM, Kate wrote:
I have replaced my doorbell button five times in a year. It is a wired one,
but the light keeps going out. It has a diode in it. The Westminster chimes
sound great.

Are they mechanical chimes or electronic chimes?

Very important.

The third time the light went out, I hired an electrician, and it still would
not light two months later.

Do they operate even though the bulb is toast?

Very good question.

The electrician told me to get a doorbell without the little wire inside the
doorbell, called a diode.

You shouldnt' have to get a new doorbell, ever, but certainly not when
it's a fancy one like yours.

Did it ever work right? 5 buttons in a year, but what about before
then? If it was new 18 months ago, and even if it wasn't, you could
call the manufacturer. A) they should know the problem and the
solution, B) if it does require replacing the bell and the problem
started within the warranty period, they may well replace the bell.
Some places will even go beyond a warranty period if they know it's
their product that failed.

OTOH, if your transformer is bigger than it was supposed to be, that
would not be the bell company's fault. Again, how long did the bell
work right.

My house came with a normal chime and a normal transformer. I think I
put in a lighted button. Then I couldn't hear the bell when I was in
the basement so I added a clapper bell in parallel in the basement.
Then I found a nicer bell on sale and put that in the hall and the old
one in the basement. Then the transformer wasn't strong enough to
ring both of them, so I bought a bigger xformer, probably higher
voltage.

Everything was fine for years until someone told me that everytime I
got a package, my burglar alarm went off. Apparently this had been
happening for years, and I figured out my glass/wood breakage
detectors were tripping because the doorbell was too loud (The
delivery man used to ring the bell then, or more likely the mailman.)

Rather than lower the sensitivity of the glass detector, I lowered the
voltage of the transformer by inserting a resistor in series with the
bell. This might be too much for you, because I have a variable
resistor whose knob I can turn until I get it working right, and then
a collection of various sizes of resistor I could find one that was
the same size as the variable one was when it worked.

Still, if you had even one or two resistors, you might well get lucky.
YOu can use little jumper wires, with alligator clips on each end, to
put a resistor in temporarily.


Assuming he is competent, then he's determined that the chime unit
requires the diode to allow power to continue to flow into the
unit AFTER you have released the button.

The bulb doesn't care about whether or not the button is pressed or
not. It also doesn't care about the diode or the chime unit.

Sort of.

The bulb has to be sized (electrically) so that the power that it
draws from the circuit isn't high enough to let the chime unit
think the button is being pressed (when it isn't). A button that
requires lots of power effectively looks like a short circuit...
ACROSS the button (so, the button appears pressed).

The bulb DOES care about the open circuit voltage across the
button. This will be determined by the characteristics of
your chime unit (those will be printed ON the chime unit!)
and the transformer chosen to drive the circuit.

The diode subtly changes what the bulb "sees".

Your transformer delivers AC (Alternating Current) to the
circuit. (batteries are Direct Current) Think of these
as waves on the ocean -- they have peaks and troughs.
But, the *average* "water level" of the ocean is somewhere
between (halfway!) the peaks and the troughs.

The bulb (switch) "sees" the range from peak to trough
all the time -- while the button is not pressed. The size
of these peaks/troughs -- the range between them -- is set
by the transformer. E.g., a 24 volt transformer has a
range between peaks and troughs that is twice what the range
would be for a 12 volt transformer.

The diode alters this. It lets ONLY the peaks go through
it. So, the peaks BYPASS the light -- taking a shortcut
through the diode! Instead, the light bulb only "sees"
the troughs. As the troughs are just as LOW as the peaks
are HIGH, this means the light "sees" half of the "voltage"
that the transformer produces.

View this sideways:

Normal electricity from transformer:
/
(
\
\
)
/
/
(
\
\
)
/


Electricity that takes a shortcut through the diode:
|
|
\
)
/
|
|
|
\
)
/
|


Electricity that the lightbulb "sees":
/
(
\
|
|
|
/
(
\
|
|
|

I lay my head down to look at these and I fell asleep. I feel rested
now. Very good explanation.

Bad illustration, but hopefully, you can see that the difference
between the leftmost (trough) and rightmost (peak) seen by the
light is LESS than the difference between the trough and peak
delivered by the transformer.

WITHOUT THE DIODE, the light sees what the transformer delivers.
So, it operates at a higher voltage. If not designed for that,
it burns out faster.

Also, if the diode fails ("open"), a bulb ends up seeing twice
the normal voltage that it would have with the diode functioning
properly. (diodes have ratings just like every other component)

Can someone please educate me. From what I am finding out, all wired doorbells
must have a diode or else the doorbell won't chime.

Well I don't know if any have one in the bell, but probably only fancy
ones like yours that play more than 2 notes.

See my response elsewhere this thread for a rough explanation.
Illustrations would be great, here, but too tedious to upload to
a hosting site, etc. (sorry)

If I really don't need a diode, for a wired doorbell, can someone tell me that
brand and model number of the unit? I am not interested in a battery operated
unit, but surely, there must be a doorbell button out there that just keeps on
working.

Call the manufacturer. First look on their website. If you dont'
know who made it, call any manufacturer and bluff your way. It won't
hurt them to spend 3 or 4 minutes on you. After all, all doorbell
makers are brothers.

Again, the *button* appears to be working -- but not the light
(which, presumably, is only illuminated when the button is NOT
being pressed?)

This cannot be rocket science.
The electrician I had moved, so I need to start all over again.

It's relatively easy to create a modification to an existing button
to "fix" this problem. I've installed LEDs in the illuminated
"doorbell" buttons for the garage door opener (one inside the house,
the other inside the garage). They'll be there until long after
the opener gives up the ghost! :-/

Many thanks!

HTH

Great advice. I will call the manufacturer and hopefully they can help me.