Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead. They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of this
cap, is a small bridge, and there is a 100uF 35 volt cap across its output.
The feed from the + leg of the bridge to the rest of the circuitry, is then
via a tiny inductor. In both units, this inductor was open circuit. No great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw through
the dropper cap. So no drop takes place, leaving huge volts across that poor
little 35 volt cap hanging on the bridge.

I'm surprised that both of them had survived at least violent explosion, but
one of them was badly bulged.

These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly in
this regard. Other than that, they seem to work quite nicely, now that I've
got them both going.

I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the remote
unit, so it's not as though it's there to stop any spurii heading backwards
towards the mains.

Arfa

On 04/04/14 15:23, Arfa Daily wrote:
I have to wonder what exactly the little inductor is for though. The
only 'RF' in the wall unit, is the receiver for the wireless link from
the remote unit, so it's not as though it's there to stop any spurii
heading backwards towards the mains.

It's a low pass filter - so it will probably be augmenting the capacitor
and contributing to providing smooth DC.

But yes, the circuit is clearly not designed to withstand certain
component failure modes. They probably assume it will be binned.

In article , Arfa Daily
writes
Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead. They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of this
cap, is a small bridge, and there is a 100uF 35 volt cap across its output.
The feed from the + leg of the bridge to the rest of the circuitry, is then
via a tiny inductor. In both units, this inductor was open circuit. No great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw through
the dropper cap. So no drop takes place, leaving huge volts across that poor
little 35 volt cap hanging on the bridge.

Sounds like Honeywell, failures are not uncommon.
--
fred
it's a ba-na-na . . . .

In article ,
"Arfa Daily" writes:
Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead. They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of this
cap, is a small bridge, and there is a 100uF 35 volt cap across its output.
The feed from the + leg of the bridge to the rest of the circuitry, is then
via a tiny inductor. In both units, this inductor was open circuit. No great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw through
the dropper cap. So no drop takes place, leaving huge volts across that poor
little 35 volt cap hanging on the bridge.

I'm surprised that both of them had survived at least violent explosion, but
one of them was badly bulged.

These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly in
this regard. Other than that, they seem to work quite nicely, now that I've
got them both going.

I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the remote
unit, so it's not as though it's there to stop any spurii heading backwards
towards the mains.

The series cap acts as a dropper at 50Hz, but any interference on the
mains will see it as a much lower impedance and will not 'dropped' to
anything like the same extent. At the bridge rectifier, mains
interference will appear to be vastly amplified.

Sounds like a crap circuit design though. Is there not even a series
limiting resistor?

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

On Friday, April 4, 2014 3:23:48 PM UTC+1, Arfa Daily wrote:

Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead. They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of this
cap, is a small bridge, and there is a 100uF 35 volt cap across its output.
The feed from the + leg of the bridge to the rest of the circuitry, is then
via a tiny inductor. In both units, this inductor was open circuit. No great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw through
the dropper cap. So no drop takes place, leaving huge volts across that poor
little 35 volt cap hanging on the bridge.
I'm surprised that both of them had survived at least violent explosion, but
one of them was badly bulged.
These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly in
this regard. Other than that, they seem to work quite nicely, now that I've
got them both going.
I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the remote
unit, so it's not as though it's there to stop any spurii heading backwards
towards the mains.
Arfa

Both of them did survive this fault scenario, so perhaps your conclusion that they'd explode is in error. Also the housing might well be intended to contain any exploded debris, its more likely.


NT

Does seem a little strange, a simple zener circuit would have sufficed to
stop issues like that.
Ok it would have taken a bit more current but be much safer I'd have
thought.
Brian

--
From the Sofa of Brian Gaff Reply address is active
"fred" wrote in message ...
In article , Arfa Daily
writes
Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead.
They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of
this
cap, is a small bridge, and there is a 100uF 35 volt cap across its
output.
The feed from the + leg of the bridge to the rest of the circuitry, is
then
via a tiny inductor. In both units, this inductor was open circuit. No
great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw
through
the dropper cap. So no drop takes place, leaving huge volts across that
poor
little 35 volt cap hanging on the bridge.

Sounds like Honeywell, failures are not uncommon.
--
fred
it's a ba-na-na . . . .

"Tim Watts" wrote in message
...
On 04/04/14 15:23, Arfa Daily wrote:
I have to wonder what exactly the little inductor is for though. The
only 'RF' in the wall unit, is the receiver for the wireless link from
the remote unit, so it's not as though it's there to stop any spurii
heading backwards towards the mains.

It's a low pass filter - so it will probably be augmenting the capacitor
and contributing to providing smooth DC.

Well, yes, it's in the position of a choke-input filter, but it is honestly
so tiny that I'm sure its value must be small enough to have a negligible
effect in conjunction with a cap of just 100uF, at the sort of ripple
frequency we're talking here.


But yes, the circuit is clearly not designed to withstand certain
component failure modes. They probably assume it will be binned.

I was thinking more of the potential for it to do something really nasty,
like catch fire ...

Arfa

"Brian Gaff" wrote in message
...
Does seem a little strange, a simple zener circuit would have sufficed to
stop issues like that.
Ok it would have taken a bit more current but be much safer I'd have
thought.

Indeed

Arfa


Brian

--
From the Sofa of Brian Gaff Reply address is active
"fred" wrote in message ...
In article , Arfa Daily
writes
Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead.
They
are powered directly from the mains via an X-rated series cap working as
a
watt-less dropper. Strung between mains neutral and the 'south' side of
this
cap, is a small bridge, and there is a 100uF 35 volt cap across its
output.
The feed from the + leg of the bridge to the rest of the circuitry, is
then
via a tiny inductor. In both units, this inductor was open circuit. No
great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw
through
the dropper cap. So no drop takes place, leaving huge volts across that
poor
little 35 volt cap hanging on the bridge.

Sounds like Honeywell, failures are not uncommon.
--
fred
it's a ba-na-na . . . .

"Andrew Gabriel" wrote in message
...
In article ,
"Arfa Daily" writes:
Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead.
They
are powered directly from the mains via an X-rated series cap working as
a
watt-less dropper. Strung between mains neutral and the 'south' side of
this
cap, is a small bridge, and there is a 100uF 35 volt cap across its
output.
The feed from the + leg of the bridge to the rest of the circuitry, is
then
via a tiny inductor. In both units, this inductor was open circuit. No
great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw
through
the dropper cap. So no drop takes place, leaving huge volts across that
poor
little 35 volt cap hanging on the bridge.

I'm surprised that both of them had survived at least violent explosion,
but
one of them was badly bulged.

These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly
in
this regard. Other than that, they seem to work quite nicely, now that
I've
got them both going.

I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the
remote
unit, so it's not as though it's there to stop any spurii heading
backwards
towards the mains.

The series cap acts as a dropper at 50Hz, but any interference on the
mains will see it as a much lower impedance and will not 'dropped' to
anything like the same extent. At the bridge rectifier, mains
interference will appear to be vastly amplified.

Fair point


Sounds like a crap circuit design though. Is there not even a series
limiting resistor?

Didn't see one, but I wasn't looking all that closely, so it's a
possibility. It would have to have been a pretty low value though, otherwise
the dissipation would have been enough to make it physically quite large,
and there was certainly nothing much bigger than a 1/8th watt size that I
remember seeing

Arfa


--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

wrote in message
...
On Friday, April 4, 2014 3:23:48 PM UTC+1, Arfa Daily wrote:

Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead.
They
are powered directly from the mains via an X-rated series cap working as
a
watt-less dropper. Strung between mains neutral and the 'south' side of
this
cap, is a small bridge, and there is a 100uF 35 volt cap across its
output.
The feed from the + leg of the bridge to the rest of the circuitry, is
then
via a tiny inductor. In both units, this inductor was open circuit. No
great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw
through
the dropper cap. So no drop takes place, leaving huge volts across that
poor
little 35 volt cap hanging on the bridge.
I'm surprised that both of them had survived at least violent explosion,
but
one of them was badly bulged.
These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly
in
this regard. Other than that, they seem to work quite nicely, now that
I've
got them both going.
I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the
remote
unit, so it's not as though it's there to stop any spurii heading
backwards
towards the mains.
Arfa

Both of them did survive this fault scenario, so perhaps your conclusion
that they'd explode is in error. Also the housing might well be intended
to contain any exploded debris, its more likely.


NT


Nope, no error. When I first looked at the board and how this part of the
circuit was configured, I hooked it up to a variac, and stuck a meter across
the cap. I stopped winding the variac up when the volts across the cap went
past 70. Bear in mind that this was a 35 working volts rated cap ... I'm
sure you must have seen electros explode from excess volts across them.

The cap in the other board *had* bulged and split open, so clearly, it had
done this to register its discomfort at being so abused ... :-\

I'm sure that the case would contain any explosion, but I was thinking more
in terms of fire, although the plastic that it's made of could, of course,
be fire retardant.

Arfa

On Saturday, April 5, 2014 3:24:19 AM UTC+1, Arfa Daily wrote:
wrote in message
...
On Friday, April 4, 2014 3:23:48 PM UTC+1, Arfa Daily wrote:

Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead.
They
are powered directly from the mains via an X-rated series cap working as
a
watt-less dropper. Strung between mains neutral and the 'south' side of
this
cap, is a small bridge, and there is a 100uF 35 volt cap across its
output.
The feed from the + leg of the bridge to the rest of the circuitry, is
then
via a tiny inductor. In both units, this inductor was open circuit. No
great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw
through
the dropper cap. So no drop takes place, leaving huge volts across that
poor
little 35 volt cap hanging on the bridge.
I'm surprised that both of them had survived at least violent explosion,
but
one of them was badly bulged.
These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly
in
this regard. Other than that, they seem to work quite nicely, now that
I've
got them both going.
I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the
remote
unit, so it's not as though it's there to stop any spurii heading
backwards
towards the mains.
Arfa

Both of them did survive this fault scenario, so perhaps your conclusion
that they'd explode is in error. Also the housing might well be intended
to contain any exploded debris, its more likely.

Nope, no error.



When I first looked at the board and how this part of the
circuit was configured, I hooked it up to a variac, and stuck a meter across
the cap. I stopped winding the variac up when the volts across the cap went
past 70. Bear in mind that this was a 35 working volts rated cap ... I'm
sure you must have seen electros explode from excess volts across them.

Yes, but where I think you might be in error is in not considering energy. Caps explode when offered high current and are unvented. The supply cap limits the energy flow, perhaps to a nonexplosive level.


The cap in the other board *had* bulged and split open, so clearly, it had
done this to register its discomfort at being so abused ... :-\
I'm sure that the case would contain any explosion, but I was thinking more
in terms of fire, although the plastic that it's made of could, of course,
be fire retardant.
Arfa

It takes a certain amount of power to get fire-hot. Does the psu provide that much? Generally series cap psus provide only minimal power levels.


NT

On Fri, 4 Apr 2014 15:23:48 +0100, "Arfa Daily"
wrote:

Electrician chum recently brought me two wireless central heating
thermostats that he had had fail, to look at. I got round to pulling them
apart today to see what was wrong. Both of the 'wall' units were dead. They
are powered directly from the mains via an X-rated series cap working as a
watt-less dropper. Strung between mains neutral and the 'south' side of this
cap, is a small bridge, and there is a 100uF 35 volt cap across its output.
The feed from the + leg of the bridge to the rest of the circuitry, is then
via a tiny inductor. In both units, this inductor was open circuit. No great
shakes, you might think, but the thing is, with that inductor open, there
can be no current draw from the bridge, and hence no AC current draw through
the dropper cap. So no drop takes place, leaving huge volts across that poor
little 35 volt cap hanging on the bridge.

I'm surprised that both of them had survived at least violent explosion, but
one of them was badly bulged.

These units are from a large and generally well-respected manufacturer of
such items, so I am astounded that they can have been designed so poorly in
this regard. Other than that, they seem to work quite nicely, now that I've
got them both going.

I have to wonder what exactly the little inductor is for though. The only
'RF' in the wall unit, is the receiver for the wireless link from the remote
unit, so it's not as though it's there to stop any spurii heading backwards
towards the mains.

Arfa

I assume you just bridged across the "tiny inductor" to get the units
working. How do you know it's an inductor and not a fuse?
--
Dave W

On Sat, 5 Apr 2014 03:11:51 +0100, Arfa Daily wrote:

Well, yes, it's in the position of a choke-input filter, but it is
honestly so tiny that I'm sure its value must be small enough to have a
negligible effect in conjunction with a cap of just 100uF, at the sort
of ripple frequency we're talking here.

Prseumably this is some form of digital device, with a
microprocessor, clock, LCD display etc. It could be a small choke to
stop RF getting out and onto the house wiring.

--
Cheers
Dave.

towards the mains.

Arfa

I assume you just bridged across the "tiny inductor" to get the units
working. How do you know it's an inductor and not a fuse?
--
Dave W


Because I've been in this game a very long time ... d:-)

Definitely an inductor. The two give-aways are the silk screen component
designator, which is "L1", and the fact that it is a tiny 'cotton reel'
ferrite bobbin, wound with laquered copper wire.

No. I did not short them. I replaced them with an encapsulated inductor of a
few hundred uH

Arfa

On Sun, 6 Apr 2014 17:48:59 +0100, "Arfa Daily"
wrote:



towards the mains.

Arfa

I assume you just bridged across the "tiny inductor" to get the units
working. How do you know it's an inductor and not a fuse?
--
Dave W


Because I've been in this game a very long time ... d:-)

Definitely an inductor. The two give-aways are the silk screen component
designator, which is "L1", and the fact that it is a tiny 'cotton reel'
ferrite bobbin, wound with laquered copper wire.

No. I did not short them. I replaced them with an encapsulated inductor of a
few hundred uH

Arfa

Fair enough.
--
Dave W