gas fire pilot light won't stay on
Jem Barrett
No matter how long I keep the pilot alight by depressing the control
knob, as a soon as its released the pilot goes out - hence I can't turn
on any of the burners.
I read the mannuals that came with it and it appears the faulty component
may be a thermocouple or perhaps a thermocouple switch. I took a look at
these components and there isn't much too them (how do they go wrong?).
I assume they work by the pilot heating the thermocouple which some how
activates a bimetalic strip that maintains a flow of gas for the pilot.
Can anybody enlighten my a little further?
Jem.
Terry Birch
<jem.b...@barclaycard.co.uk> writes
that you contact Cannon's Service Department on 0541 500 500, as they
have been modifying Coalridge fires (Free of Charge)
Have the fire's serial number to hand when you ring because the girl
will ask you for it, she will also ask if it is a balanced flue model or
not. I heard about the modifictions the week before Christmas, and rang
Cannon to enquire about the mod, and they came out the day after Boxing
Day. (So it must be an urgent mod)
Regards From:
Terry Birch Tel & Fax: (+44) 01332 850 427
30 The Green, Email: te...@curley.demon.co.uk
Castle Donington,
Derby. DE74 2JX.
England
Karel Hladky
<Jo...@gawai.demon.co.uk> writes
>Not a bimetalic strip as far as I know. The thermocouple if my o-level physics
>is still
>with me is a connection of two metals - typically antimony and bismuth (?) when
>heated
>they ganerate a small voltage. Could that be correct? Or do they change their
>resistance. Don't know but I'm sure it's voltage anyway. The circuitry will
>monitor
>this. Too low a voltage from the thermocouple will mean the pilot is not on.
>
Usually the thermocouple will produce enough emf to drive the solenoid
cut-off valve directly without any circuitry as such.
Karel
--
++++++++khl...@khdesign.demon.co.uk++Manchester M14 6BF England++++++++
John Stumbles
> In article <406092...@gawai.demon.co.uk>, JJH
> <Jo...@gawai.demon.co.uk> writes
> >Not a bimetalic strip as far as I know. The thermocouple if my o-level
physics is still with me is a connection of two metals - typically
antimony and bismuth (?) when heated they ganerate a small voltage.
Could that be correct? Or do they change their resistance. Don't know
but I'm sure it's voltage anyway. The circuitry will monitor this.
Too low a voltage from the thermocouple will mean the pilot is not on.
>
> Usually the thermocouple will produce enough emf to drive the solenoid
> cut-off valve directly without any circuitry as such.
Surely not? AFAIK thermocouples produce tens - maybe hundreds - of
millivolts and even from a source resistance of milliohms (two bits of
wire joined together) I'd be surprised if there's enough power to drive a
solenoid. I could be wrong ... ?
--
John Stumbles j.d.st...@reading.ac.uk
Computer Services, University of Reading http://www.rdg.ac.uk/~suqstmbl
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
If it ain't broke, don't fix it.
Matthew Marks
: > Usually the thermocouple will produce enough emf to drive the solenoid
: > cut-off valve directly without any circuitry as such.
:
: Surely not? AFAIK thermocouples produce tens - maybe hundreds - of
: millivolts and even from a source resistance of milliohms (two bits of
: wire joined together) I'd be surprised if there's enough power to drive a
: solenoid. I could be wrong ... ?
You are wrong :-) - amazing but true, except that there is only enough power
to *hold* a solenoid in position, rather than pull it in. That's why these
systems always require you to hold a button in place until the pilot has
ignited.
Matthew mat...@rd.bbc.co.uk My opinions, not Auntie's.
Geoff Love
> In article <406092...@gawai.demon.co.uk>, JJH
> <Jo...@gawai.demon.co.uk> writes
> >Not a bimetalic strip as far as I know. The thermocouple if my o-level physics
> >is still
> >with me is a connection of two metals - typically antimony and bismuth (?) when
> >heated
> >they ganerate a small voltage. Could that be correct? Or do they change their
> >resistance. Don't know but I'm sure it's voltage anyway. The circuitry will
> >monitor
> >this. Too low a voltage from the thermocouple will mean the pilot is not on.
> >
>
> Usually the thermocouple will produce enough emf to drive the solenoid
> cut-off valve directly without any circuitry as such.
I find this hard to believe. The current required to operate a solenoid is
quite high and must be way above that produced by a thermocouple.
--
Geoff Love
(e-mail: ge...@gdes.demon.co.uk)
Jem Barrett
>: Surely not? AFAIK thermocouples produce tens - maybe hundreds - of
>: millivolts and even from a source resistance of milliohms (two bits of
>: wire joined together) I'd be surprised if there's enough power to
>:drive a solenoid. I could be wrong ... ?
>
>You are wrong :-) - amazing but true, except that there is only enough
>power to *hold* a solenoid in position, rather than pull it in. That's
>why these systems always require you to hold a button in place until the
>pilot has ignited.
Well I originally took the fire apart for a good look and couldn't work
out how it operated. A wire goes from the thermocouple to a microswitch.
Another wire comes from the microswitch and goes to the main gas
distribution widget. On the microswitch is a switch that is depressed
(when the fire is off) by a cam attached to the control knob. When you
turn the knob to ignite the pilot the switch becomes open.
My thinking is the current generated by the thermocouple recloses the
microswitch which allows the current to flow to the gas distribution
widget. How this maintains a flow of gas for the pilot i haven't a clue.
It can't just be the fact that the switch is closed because I held it
closed with fingers and the pilot still refused to stay alight.
Anybody want to enhance my theories?
Jem.
Matthew Marks
: It might be useful if you could describe to us what the thermocouple looks
: like and also the wire from it. AFAIK, what the gas industry calls a
: "thermocouple" is not a "thermocouple" in the electronics/electrical
: industry! In my Gas boiler, the "thermocouple" is that thing looking
: like a stubby metal pencil with a copper tube coming out of it. The tube is
: invariably too long and so is coiled before it enters a gas valve.
: The pencil end sits in the pilot flame. I understand it to be entirely
: passive (mechanical), and behaves as a flame detector. At the gas valve
: end, it either conveys heat to a bimetallic-operating valve, or there
: is a small pin which protrudes when the bulb end is hot, so acting directly
: on the valve. I suppose that you might have one acting directly on a
: microswitch.
No! After stating what I thought was the case, I have checked with someone
who has actually done work on these things, and it *is* a single-junction
thermocouple. It is not quite so remarkable that this can hold in a
solenoid when you consider that:
1) as I said, it only has to hold the magnetic circuit closed, not actuate
the solenoid;
2) the *voltage* might only be a few millivolts, but it's a low resistance
solenoid and apparently the current is of the order of 60mA. It's hard to
measure though, because the internal resistance of a meter will probably
be too high.
If you don't believe me, try applying a small voltage to the thermocouple
connections on the gas valve and see if it stays actuated.
I don't know why it's in a copper tube: could be to keep the resistance low,
to stand the heat at the hot end, or so that plumbers don't shy away from it
because it looks like something electrical!
Rather squares with the description of the gas fire having a microswitch
doesn't it? But you'd have to take care that the resistance of the contacts
is very low!
: It is highly unlikely that your valve is operated by a dissimilar-metal
: millivolt-producing device. You would need an amplifier to bring it
: into the real world and heaters are simply not that sophisticated!
Think of the current, not the voltage. The current through a solenoid is
what matters.
There are many remarkable electrical things: the notion of an electric clock
going for a year on a single battery was quite remarkable to me when it
was first invented. And those air freshener things which keep a fan running
continuously whenever washrooms are light are also almost unbelievable.
Incidentally, another method of flame detection, as used on Potterton Netaheat
boilers, is to pass a current through the flame. The Netaheat uses the same
electrode to provide the ignition sparks as to determine when the flame has
ignited. Very neat!
Rick Hughes
failure detect is a thermocouple, when hot it generates a voltage.
This voltage feeds into the flame failure valve coil on the gas valve.
So no heat = no voltage and the current in the coil will be insufficient
to keep open the gas supply ( which is what you do when you hold the dial
open when you first light up the boiler.)
If after the 30 secs or so when you leave the valve go the pilot goes out
then the odds are the thermocouple is naff .... go to your local trade
plumbing outlet and they only cost around £3.
They are a simple threaded fitting.
As a temp workaround you can often get the boiler to light by providing a
flame from a blow torch onto the thermocouple, this will get it hot enough
to allow the boier to work, there seeme to be hysterisis built in, and it
will continue to work O.K. untill the mains gas is turned off.
Then it will not relight.
Cheers
Rick
In article <406092...@gawai.demon.co.uk>, Jo...@gawai.demon.co.uk wrote:
> In article: <4f502k$2...@gatekeeper.barclaycard.co.uk>
jem.b...@barclaycard.co.uk
> (Jem Barrett) writes:
> >
> > I've got a Cannon Coalridge gas fire which developed a fault yesterday.
> > No matter how long I keep the pilot alight by depressing the control
> > knob, as a soon as its released the pilot goes out - hence I can't turn
> > on any of the burners.
> >
> > I read the mannuals that came with it and it appears the faulty component
> > may be a thermocouple or perhaps a thermocouple switch. I took a look at
> > these components and there isn't much too them (how do they go wrong?).
> > I assume they work by the pilot heating the thermocouple which some how
> > activates a bimetalic strip that maintains a flow of gas for the pilot.
> >
> > Can anybody enlighten my a little further?
> >
> > Jem.
> >
> >
> Not a bimetalic strip as far as I know. The thermocouple if my o-level
physics is still
> with me is a connection of two metals - typically antimony and bismuth
(?) when heated
> they ganerate a small voltage. Could that be correct? Or do they change their
> resistance. Don't know but I'm sure it's voltage anyway. The circuitry
will monitor
> this. Too low a voltage from the thermocouple will mean the pilot is not on.
>
> I do hope that is correct, any way it is something like that.
>
> Try replacing the thermocouple, it is usually this that goes and they
aren't expensive.
> >
> --
> JJH
Jem Barrett
>It might be useful if you could describe to us what the thermocouple
>looks like and also the wire from it. AFAIK, what the gas industry
>calls a "thermocouple" is not a "thermocouple" in the
>electronics/electrical industry!
>In my Gas boiler, the "thermocouple" is that thing looking
>like a stubby metal pencil with a copper tube coming out of it. The tube
>is invariably too long and so is coiled before it enters a gas valve.
>The pencil end sits in the pilot flame.
My thermocouple is the same as what you describe. The copper tube does
not go directly to the gas valve but to a switch. Another tube leaves
the switch and goes to the gas valve. The tube appears to be nothing
more than some method of protecting the insulated wire that runs inside
it.
I tend to agree with Mathew that electricity is involved somewhere, but
I'm still at a loss as to how the gas valve performs the gas shut off. Is
there another kind of switch inside the valve?
Jem
Matthew Marks
: I tend to agree with Mathew that electricity is involved somewhere, but
: there another kind of switch inside the valve?
I don't really know what you're getting at here: there is a spring-loaded
valve which allows gas to get to the pilot and the main burner. There is
probably some mechanical arrangement to prevent the main burner igniting
when you are holding down the pilot valve: e.g. you can't turn the main valve
on while the button is depressed. When the pilot is lit, the thermocouple
will generate enough oomph to hold the spring-loaded valve open. When you
turn the whole thing off (ie open the microswitch) the circuit is interrupted
and the valve will close under the action of the spring, shutting off the gas
supply to everything. If the microswitch closes again, even while the
thermocouple is still hot, the valve will not open because the solenoid is
incapable of moving its armature, only holding it in position.
Does that help?
||
vv
Jem Barrett
says...
>
>I don't really know what you're getting at here: there is a
>spring-loaded valve which allows gas to get to the pilot and the main
>burner. There is probably some mechanical arrangement to prevent the
>main burner igniting when you are holding down the pilot valve: e.g. you
>can't turn the main valve on while the button is depressed. When the
>pilot is lit, the thermocouple will generate enough oomph to hold the
>spring-loaded valve open.
OK, first the electricity issues. We're sure its electrical because the
wires that come out of the copper tubing are connected to the microswitch
by electrical connectors (like in a car's electrics).
So the thermocouple generates a small current that closes the
microswitch. Does the current then flow down the other wire out of the
microswitch that leads to the gas valve. When the current gets to the
gas valve it then activates the spring valve? Am I reading you
correctly? It sounds feasable. When you switch off the fire the
microswitch is forced open (or is it closed??) by a cam on the knob.
Between the thermocouple and the gas valve there is only the microswitch.
They are joined by insulated wires partially enclosed in copper tubes (at
the thermocouple and valve ends).
Jem.
Matthew Marks
: mat...@rd.bbc.co.uk (Matthew Marks) wrote:
: >No! After stating what I thought was the case,
:
: My server doesn't have any copy of your posting so maybe that explains
: why you are shouting at me.
Clive, will you please calm down. I was not shouting at you, and you have
lowered the tone of this discussion by responding in such a manner. If I was
telling you things that you already know then you'll just
have to forgive me: it's inevitable in this medium.
:
: >I have checked with someone
: >who has actually done work on these things, and it *is* a single-junction
: >thermocouple. It is not quite so remarkable that this can hold in a
: >solenoid when you consider that:
: >
: >1) as I said, it only has to hold the magnetic circuit closed, not actuate
: >the solenoid;
:
: How much power would that take?
No power: no work is being done. You are only holding against a spring. Of
course, there is current flowing so work must be done to overcome resistance.
>2) the *voltage* might only be a few millivolts, but it's a low resistance
>solenoid and apparently the current is of the order of 60mA. It's hard to
>measure though, because the internal resistance of a meter will probably
>be too high.
:
: So. Nothing to do with the dissimilar metals of the probes and thermocouple
: wires then?
I never said that wouldn't have an effect, but these are cold junctions: after
all, there has to be a cold junction somewhere. (For the benefit of other
readers, the dissimilar metals have to join at two places in the circuit,
and the voltages generated at each point will oppose, but if the joints are
at different temperatures then there will be a net current flow because the
hot junction will "win" oner the cold junction.) The effect of any third
metal of the meter leads will be cancelled anyway, as the connections to it are
all at the same temperature.
: >If you don't believe me, try applying a small voltage to the thermocouple
: >connections on the gas valve and see if it stays actuated.
:
: Have you tried this?
Rhetorical question I assume: you haven't tried it to disprove my assertion.
: If, as I suspect in this case, the bulb contains something like
: a wax, its melting will be mechanically amplified by the small bore tube and
: produce a mechanical displacement at the far end. This could actuate a micro-
: switch.
But where does the current come from to drive the circuit that the microswitch
is part of?
: > But you'd have to take care that the resistance of the contacts
: >is very low!
: The micro-switches contacts? They need to handle the _power_ as well as the
: arc on inductance disconnection.
The arc would be insignificant I would think; the resistance would have to be
very low because you've only got about 12mV to play with and a high current.
: >Think of the current, not the voltage. The current through a solenoid is
: >what matters.
:
: Yep. And the voltage produces the current.
And with a very low resistance you don't need much of it.
: Lets get in the ball-park! An iron.constantan thermocouple
: will produce 20mV at 400C. Assuming a circuit resistance of 0.1 Ohm that
: is a power of 4mW. Accounting for lead + coil resistance etc, will bring
: the power right down. I don't know of any magnetic circuits that operate
: down there. Do you?
Yes. A gas pilot valve. And perhaps that fan motor:
: >And those air freshener things which keep a fan running
: >continuously whenever washrooms are light are also almost unbelievable.
:
: The wonders of modern science.
: >
: >Incidentally, another method of flame detection, as used on Potterton Netaheat
: >boilers,
<snip>
: Same as in most auto-igniting gas ovens so not exactly novel.
I never said it was. Please don't take things so personally. This is a
public forum and the information is of general interest.
: You may guess that I'm not convinced and you'd be right!
Well perhaps I'd better ask the engineer here who has actually done quite a lot
of work on these things to contact you. He's no fool.
Karel Hladky
<jem.b...@barclaycard.co.uk> writes
>My thinking is the current generated by the thermocouple recloses the
>microswitch which allows the current to flow to the gas distribution
>widget. How this maintains a flow of gas for the pilot i haven't a clue.
>It can't just be the fact that the switch is closed because I held it
>closed with fingers and the pilot still refused to stay alight.
>
>Anybody want to enhance my theories?
Obviously a safety feature so that if you accidentaly aim a blowtorch at
the thermocouple nothing will happen unless you also have the control
knob in the on position :-)
Or are the microswitch contacts open when the switch is depressed ?
Matthew Marks
: OK, first the electricity issues. We're sure its electrical because the
: by electrical connectors (like in a car's electrics).
: So the thermocouple generates a small current that closes the
: microswitch.
No - see below.
: Does the current then flow down the other wire out of the
: microswitch that leads to the gas valve. When the current gets to the
: gas valve it then activates the spring valve? Am I reading you
: correctly? It sounds feasable. When you switch off the fire the
: microswitch is forced open (or is it closed??) by a cam on the knob.
I think you are confused about the microswitch: it's just a switch like any
other, which allows current to flow between its terminals if it is not opened
by the cam. To open the gas valve in the first place you need mechanical
pressure, because the thermocouple cannot generate enough current to pull the
valve open, only to hold it there. To hold the valve open once you release
your finger requires the thermocouple to be hot (to generate a current) and
the microswitch to be closed (to complete the circuit and allow the current
to flow through the solenoid and back to the thermocouple). If the pilot goes
out the thermocouple will cool and the valve will close under spring pressure;
if the microswitch is opened the same thing will happen because the circuit
has been broken. If the microswitch closes again then the valve will stay
shut because there is never enough current to open it, only to hold it open.
Alternatively, the microswtich may be connected in parallel with the solenoid
and thermocouple, in which case when it *closes* it will short out the
thermocouple and the valve will close.
Richard Herring
: Clive Elsmore (Clive....@jet.uk) wrote:
: : It might be useful if you could describe to us what the thermocouple looks
: : like and also the wire from it. AFAIK, what the gas industry calls a
: : "thermocouple" is not a "thermocouple" in the electronics/electrical
: : industry! In my Gas boiler, the "thermocouple" is that thing looking
: : like a stubby metal pencil with a copper tube coming out of it. The tube is
: : invariably too long and so is coiled before it enters a gas valve.
: : passive (mechanical), and behaves as a flame detector. At the gas valve
: : end, it either conveys heat to a bimetallic-operating valve, or there
: : is a small pin which protrudes when the bulb end is hot, so acting directly
: : on the valve. I suppose that you might have one acting directly on a
: : microswitch.
: No!
Well, this may actually be (nearly) true of Clive E's boiler: read on.
: After stating what I thought was the case, I have checked with someone
: who has actually done work on these things, and it *is* a single-junction
: thermocouple. It is not quite so remarkable that this can hold in a
This is also true of some boilers.
[more true facts about solenoids and thermocouples snipped]
: Incidentally, another method of flame detection, as used on Potterton Netaheat
: boilers, is to pass a current through the flame. The Netaheat uses the same
: electrode to provide the ignition sparks as to determine when the flame has
: ignited. Very neat!
Almost. Maybe it's different with later models, but on mine, the electrode
detects the presence of a flame and indeed turns the spark-generator off - but
that's all it turns off. (Circuit available on request; it's quite clever)
The presence of pilot flame then heats a liquid-filled
bulb at the end of a capillary tube which dangles in the flame; the pressure
of the vapour which this produces then operates a mechanical bellows in what
the trade calls the "capillary valve", which turns on the main gas flow. No
bimetallic strips, no solenoids.
(Though the pressure gauge which turns the gas flow and the spark generator
on in the first place operates a solenoid gas valve which controls the flow
to both the pilot and the capillary valve. Pottertons are packed with
interesting bits of control and feedback technology.)
The confusion arises because the Potterton capillary tube filled with fluid
looks identical to the coaxial electrical cable used by other manufacturers
to connect a hot thermocouple to a solenoid valve.
The difference soon appears when it goes wrong - instead of spending a few quid
on a new thermocouple, you are looking at not much change from a hundred for a
new capillary-valve/tube/bulb assembly, as you can't get the bits separately.
--
Richard Herring | richard...@gecm.com | Speaking for myself
GEC-Marconi Research Centre | Not the one on TV.
Leo Hendry
>
>My server doesn't have any copy of your posting so maybe that explains
>why you are shouting at me.
'!' is an exclamation mark. Nothing to do with SHOUTING.
- Leo
Matthew Marks
: In article <DMK1A...@unixsoft.demon.co.uk>
: Leo Hendry <l...@unixsoft.demon.co.uk> writes:
:
: > '!' is an exclamation mark. Nothing to do with SHOUTING.
:
: And nor do capital letters.
I think you'll find that it is generally accepted on the net that writing in
capitals constitutes "shouting" - I very often see posts telling people not to
shout in response to a message where someone has left their caps lock on
(or is connected to the net with a telex machine :-).
Steve Hunt
> '!' is an exclamation mark. Nothing to do with SHOUTING.
And nor do capital letters.
-- Steve
--
Steve Hunt - Online Media - Cambridge, UK
st...@omi.co.uk +44 1223 518563 IRC: Daff
Andy the Pugh
>
>The presence of pilot flame then heats a liquid-filled
>bulb at the end of a capillary tube which dangles in the flame; the
pressure
>of the vapour which this produces then operates a mechanical bellows in
what
>the trade calls the "capillary valve", which turns on the main gas flow.
No
>bimetallic strips, no solenoids.
>
>The difference soon appears when it goes wrong - instead of spending a few
quid
>on a new thermocouple, you are looking at not much change from a hundred
for a
>new capillary-valve/tube/bulb assembly, as you can't get the bits
separately.
>
I noticed that the one on an old gas cooker we had had been crimped with
pliers. Onm consideration I decided that this was to vary the internal
volume (which is presumably why the spare has to be coiled up). Crimping it
a little more reduced the time for which the ignition button had to be held
in for from 5 minutes (tedious) to 10 seconds (much better)
If you are feeling brave you could try carfully sqaushing the tube.
(Insert stabdard disclaimer about gas and safety devices and calling in the
professionals)
--
ap
Andy the Pugh
>
>Therefore the contacts in the thermocouple circuit
>have to be good. I presume that the thing in Jem's case that looks
>like a microswitch is actually an elecro-mechanical relay. If I remember
>right, the Total Magnetic Flux is a product of the number of coil
>turns and the current. For a reed relay, you might have 1000 turns
>of thin (resistive) wire at 10mA. You'd get the same flux if you had
>50 turns of thick (low resistance) wire at 200mA. The thermocouple
>produces 20mV and if you a circuit impedance of 0.1 Ohm you could
>actually generate the 200mA, so it might just work, if you can get
>the required flux concentrations in the switching element.
>(It could also be a solid state "hall-effect" relay - in which case
>forget my calculations!).
>
Of course, the very act of drawing a current from a thermocouple cools it,
so there is a limit to the power you can draw from a thermocouple.
(Reverse Peltier effect I think, but that is just off the top of my head)
--
ap
Karel Hladky
<Clive....@jet.uk> writes
>Humble Pie time! Further to my earlier postings about "thermocouples",
>I have to come clean and admit that I was wrong about the one fitted
>in my boiler.
This one should go in the FAQ, it's at least the third time it's cropped
up since this ng started.
BTW the 'thermocouples' in gas cooker ovens do actually work on
expansion since they run at a much lower temperature and the generated
emf would not be enough to switch anything. Hence the far greater cost.
Matthew Marks
: Matthew Marks (mat...@rd.bbc.co.uk) wrote:
: : Incidentally, another method of flame detection, as used on Potterton Netaheat
: : boilers, is to pass a current through the flame. The Netaheat uses the same
: : electrode to provide the ignition sparks as to determine when the flame has
: : ignited. Very neat!
:
: Almost. Maybe it's different with later models, but on mine, the electrode
: detects the presence of a flame and indeed turns the spark-generator off - but
: that's all it turns off. (Circuit available on request; it's quite clever)
It must be different for later models: mine has two solenoids and only the
electrode dangling in the flame.
: The difference soon appears when it goes wrong - instead of spending a few quid
: on a new thermocouple, you are looking at not much change from a hundred for a
: new capillary-valve/tube/bulb assembly, as you can't get the bits separately.
Just as well that I have the later model then!
Alan J. Flavell
>I think you'll find that it is generally accepted on the net that writing in
>capitals constitutes "shouting" - I very often see posts telling people not to
>shout in response to a message where someone has left their caps lock on
>(or is connected to the net with a telex machine :-).
curiously, German telex machines (the ones I've seen, anyway) have all
lower case letters, i.e no capitals.
As we say on c.i.w.a.html, this is a "browser issue". ;-)
---
Alan