View Single Post
  #44   Report Post  
Posted to alt.home.repair
Robert Green Robert Green is offline
external usenet poster
 
Posts: 4,321
Default Testing dollhouse circuits and bulbs

"Jennifer Murphy" wrote in message
wrote:
"Jennifer Murphy" wrote in message


My multimeter arrived today. I cannot get to the wiring underneath the
dollhouse because there is paint and glue drying. ;-)


Welcome to the wonderful world of electrical troubleshooting. I have

about
20 meters (maybe 30) because they are indispensable for electronics work.
Some hook to PC's for recording work, some are small and lighted that I

keep
in my car. I keep one in my camera bag, one connected to the battery
powered sump pump and even have a AA-A battery charger that reads the
voltage of each cell as it charges. My wife says it's a sickness but I
believe you can never have enough multimeters, flashlights or

rechargeable
batteries. (-:

I tried to get a voltage reading at the light sockets inside the doll
house. I could not get a consistent reading. The space is very cramped
and the sockets very small (3/8").


For testing like that I might make up a test prod from a dowel about the
size of the bulb with a screw in the center to simulate a bulb, but

that's
probably a bit much for you just starting out.


I don't have the tools for that ;-(


That's too bad, but understandable. After thinking about, I would break one
of the dead bulbs, cut a small jumper wire in half and solder the bare wire
ends to the bulb's contacts. Jumper wires have alligator clips on each.
Cutting one in half leaves a bare end and an end with a metal clip that you
can clip to your meter probes. I could fabricate such a "tester" in less
time than it took to type this message but I have the tools. I'm aware that
you don't. I'd be happy to make up a socket tester for you gratis but it
wouldn't happen until well after Christmas. I can't imagine testing the
sockets any other way. It's hard to hold the probes correctly even with the
socket on the benchtop. But when mounted upside in a the tiny rooms of a
dollhouse? Sounds like torture from the Spanish Inquistion! And nobody
expects the Spanish Inquistion (if I didn't say it, at least 5 Monty Python
fans would have).

It takes a delicate hand to
read a micro-socket when it's on a test bench. Mounted in a doll house
might be much harder and close to impossible without the right probes.

One
the backside of the socket there are often screw terminals. That would

be a
good place to check the socket voltage.


I can't get to the back side. The sockets are mounted in the ceiling. On
the first floor, it's about 1/2" thick and completely finished. On the
second floor, there is an attic, but it's sealed. I don't want to start
taking it apart.


I don't blame you. Once upon a time when I first started rebuilding Brit
sportscars, I came across this humorous comment in a "how to" book:
"Carburetor" is a French word that means "leave it alone." The same could
be said of your antique dollhouse's electrical wiring.

I seemed to get a better reading on
DC than AC. The DC readings jumped all over the place, but it looked
like between 0.9 and 1.1 v. The AC readings jumped around even more.

The
most consistent reading was 6v at one socket.


Is there a way using this meter to tell if the current is AC or DC?


That's a good question. Measuring an AC voltage using a high quality,
accurate voltmeter, that set to measure DC volts, should, by definition
result in a reading of zero volts. The reason you read a small residual

DC
voltage when measuring an AC voltage might be because of small

inaccuracies
in your meter. Any true AC voltage has absolutely no DC component. I

would
test it using a diode - even an LED would work - that will pass current

in
only one direction. But that's little advanced. I suppose you could test
the circuit by reversing the probes. If it's DC, the + and - signs

should
reverse themselves when you switch the probes. With AC, it shouldn't

have
any effect.


I thought I got a +/- change one time, but it's so difficult to get at
the sockets, that I can;t be sure I'm getting good contact and not
shorting the probes.


I totally understand. It's a job for a specially made test probe.

Next, I did a continuity test of the bulbs themselves. Of the 12 bulbs

I
removed from the house, only one passed the continuity test.


Excellent work! With an old dollhouse and the possibility that the bulbs
are getting more voltage than designed for, it's very likely the only
problem is burned out lightbulbs.


I ordered a bunch of 2.5v #14 bulbs. They should be here tomorrow.


That's where the rubber meets the road. I suspect you'll get more than a
few working. Who knows? If you clean the sockets and cycle the switches,
they might all work.

I then tried a resistance test on the bulbs. The one that passed the
continuity test showed 11 ohms of resistence. The rest showed infinite
resistance.


That sounds right. Infinite resistance means the filament is broken and

the
bulb is dead. You should be able to confirm that with a magnifying

glass.
If they were run at double their rating, the filament damage should be

very
obvious and might include blackening of the bulb's interior.


The filaments all look OK. A couple are somewhat blackened.

I then took the good bulb and tried it in several sockets. After
fiddling with the button switches, I was able to get it to light in
several sockets. It was very dim.


You may also be experiencing problems with dirty switch contacts and

dirty
or corroded bulb sockets. The center button connector of the socket

should
be nice and shiny in each socket. I bet they're not.


The switches are definitely at least a little suspect. They are the
push-button type. They are a little loose in the housing and if I jiggle
them, the lights flicker. I can't see into the sockets, but the house
has been stored in garages on the east coast (humid) for many years. I
might go get a mirror on a stick and see if I can see into the sockets.
But if they are corroded, there ma not be much I ca do without taking
the while thing apart, which is too much for this Christmas.


You can use the rubber eraser end of a small pencil inside the socket to
clean off the oxidation. Make sure to blow out the rubber particles with
canned air or a little makeup brush.

When everything dries, I will try to get under the house and see if I
can get better readings.


I don't think you have to do a complete rewire - if it were 110VAC

passing
through those wires you would have seen the bulb go supernova and

possibly
pop. There's got to be a transformer somewhere.


Yes, I'm quite certain that it's not 110 and probably not AC.


I would guess that the AC line cord goes into a small transformer that
outputs from 2 to 5 VAC. Lightbulbs don't care whether they run on AC or DC
so I don't see any reason for the designer to have a rectifier in the house
that converts AC to DC. It would be an extra and unnecessary step for the
builder and I'm guessing if he omitted strain relief on the line cord, he
was on a real budget.

By getting the bulb to
light in several sockets you've conclusively proved they can not be wired

in
series (in a chain like old Christmas lights where if one bulb went dead,
they all died). That's important to know.


Yep, I was glad to discover that.


That's what leads me to believe there's a small AC transformer at the end of
the line cord. It should be heavy enough to be detectable in some way. It
might even put out a hum that's detectable by a close-by AM radio.

Trace where the AC wires comes into the dollhouse and find the

transformer
(apologies if you've already done this - I just started reading this
thread.). If it's well wired to the cord and the cord's in good shape, I
wouldn't rewire the unit. Switches can often be "decrudded" by operating
them a few dozen times to scrape off the oxidation.


The AC cord is one of those really old fabric-covered types. The place
where it goes into the base of the house is just a hole in the wood with
no fitting or stress relief. There is no obvious wear that I can see,
but it all seems a little unsafe to me.


You're right. I would replace the cord with a more modern one and add
strain relief if it's not too hard to access the point of entry. If it is.
I would probably use some low-temp hot melt glue to create one. Not pretty,
but it's held when I've had to do it. You could cut the cord a few inches
from the hole and slide protective tubing of some sort to protect the wire's
insulation. Then either solder a newer line cord, the old line cord or an
in-line power switch to the bit left sticking out. You could also wrap the
cord that sticks out of the hole with a enough layers of black electrical
tape and then push that bulge into the hole to create strain relief of
sorts. Old cloth covered wire without a strain relief bothers me. I've
rebuilt items with similar cords that looked intact but that crumbled in
short order when they were flexed just a little bit. On the other hand,
most of the original wiring in the my house is cloth covered. I've since
added new circuits with 12/2 romex to take most of the load off the old
wires where I could. But I digress . . .

I think the next step is to clean the sockets, cycle the switches,

install
the replacement bulbs and take it from there.


I managed to get under the house. The whole bottom is covered and glued
shut. I'm starting to think that this wasn't that well made. I'm tempted
to cut the cord and go without lights for now. If the little girl really
gets into it (right now she's more into Buzz Lightyear), I'll consider
taking it apart and installing LEDs and a battery pack.


That sounds like serious deconstruction of the house would be involved. I'd
wait and see how it goes over first. Rebuilding using LED's and batteries
or a small wall wart (the plug in transformer-rectifier that are used for
chargers and such) It might be a perfect teaching opportunity to involve the
youngster with. You could even buy some cheap solar garden lights and turn
it into a solar powered doll house. (-:

Anyway, thanks for the thoughtful and patient help.


That's why we're here. Well most of us. A few seem to think it's open mike
night at their local comedy club, but that's Usenet.

--
Bobby G. - It's 58F degrees at night in December in DC. Maybe global
warming theorists are right!