Thermocouple measurements.

Hi All,

I had another look at Mums 'old but going to br replaced soonish., the pilot won't stay alight' Ascot / New World (Main?) gas boiler again yesterday and think I found the upper limit cutout / inhibitor (?) had triggered? I pressed what looked like a rest button in the middle of it but as didn't have any real tools / meter with with me (only the trusty Leatherman as always) I couldn't measure anything so shorted the extra stat out (for testing) and tried to light the pilot.

Hold in the contril knob, hit the piezo, nice little pilot flame, hold knob in 30 seconds, release, pilot stays alight (hah huh, so it was overtemp cutout ..?).

Turn the gas knob round to 'On', pilot goes out?

Wait 3 mins, relight pilot, turn knob to 'On' pilot stays on this time, turn mains power to boiler, main burner lights up but then main pilot and burner goes out? Try this a few times with similar (intermittent) results. So, it could just be that the thermocouple was 'marginal' and as they are 'cheap' ....

So, I removed the thermocouple (amazing what you can do with a Leatherman PST) and this morning picked up a replacement. It's not the 'correct' part (long since obsolete) but sufficiently close where it matters to 'do'.

I just did some comparisons between old and new probe using my DMM and the gas stove and noted the following with either (probe, not DMM) in the gas flame (simulating pilot).

New device .. open cct voltage 30mV, short cct current (DMM on 2mA)

250uA.

Old device .. 23mV and 200uA ?

So, from that very crude test the old thermocouple isn't 'dead' but maybe down on performance (especially when not in a 'strong' flame ie only a 'pilot' not a low gas ring) enough to make a difference?

I'll see what happens when I , sorry, 'the Corgi bloke', put's it back later... "You know anything about parachutes ...?" ;-)

All the best ..

T i m

Reply to
T i m
Loading thread data ...

The difference in voltages puzzles me. There are various types of thermocouple, each being composed of two metals (normally alloys) which develop a potential difference, the Seebeck voltage, named after the discoverer. This is characteristic for any pair of metals and I am assuming that you have a very resistive meter. The amperage ought to be proportional to the contact area. Linearity is not perfect, but in a boiler, "hot" and "cold" are clearly sufficient. I think it is type T thermocouples which are used in boilers, but I will gladly be corrected on this. For more information on thermocouples than you want,

formatting link
the links seem to be broken.

Ah, yes, the old joke. BTW a couple of years back I found that simply giving the thermocouple a tighten broke through the oxide film and restored the function.

John Schmitt

Reply to
John Schmitt

Also bear in mind that there are, in effect, juncions between the meter probes and the thermocouple itself.

sponix

Reply to
s--p--o--n--i--x

If the pilot stays on after releasing the button, it sounds as if the thermocouple and flame failure device are working ok. If the pilot goes out when the main burner comes on, it suggests that it's being blown out by the sudden rush of air. In this case, the pilot probably isn't quite powerful enough - either because its height needs adjusting (is there a separate little regulator for this?) or because there's a bit of crud partially blocking its jet.

Reply to
Set Square

The current isn't real. The meter is applying a resistor across the leads so that at its nominal

200mv fullscale, it'll read full current, so 2mA range has a 100 ohm resistor across it.

As the thermocouple has a resistance way under 100 ohms, it's just reading out the voltage.

The voltage ranges can be handy for reading out very low currents (nA) similarly.

Reply to
Ian Stirling

Erm, it's a DMM and I thought even the cheapest of them was 10M ohm / v ?

The amperage ought to be

Understood ..

I think it is type T

Interesting thanks .. shame about the broken links .. ;-(

. BTW a couple of years back I found that simply

And there's little to loose for a diyer .. ;-)

All the best ..

T i m

Reply to
T i m

True, but I had the meter clipped to the very end of the thermocouple, the probe end of the thermocouple in the gas hob flame and was holding the middle bit with my hand (and I'm not a woman so don't have asbestos fingers!) ;-)

I did measure both a few times and there were some small voltage variations depending how hot you got the probe. It seemed to balance out as long as you left it in the flame for a reasonable time (say ~20 seconds).

All the best ..

T i m

Reply to
T i m

Well, I think it DID say on all the time as long as you held the knob in long enough to start with and did nothing else? Once it *seemed* to go out when I put the front back on (carefully), another when I closed the control box lid and another when I turned the gas control knob to 'Full'? But when I tried any of the above again the pilot stayed on?

If the pilot goes out

That's what I first thought but the flame was pretty strong?

Anyway, I fitted the new thermocouple, shorted the 'overheat stat' connectors and lit the pilot .. so far so good ;-)

I turned the control knob to 'On' .. still pilot .. turned the power supply to the boiler on then the boiler itself, up came the main burner and the pilot stayed on ;-)

I tried this a few times and each time it was A ok ;-)

The only 'problem' now is the only replacement thermocouple that I was supplied (that looked like it had the right ends AND the connector for the 'inhibitor') places the inhibitor connectors inside the main boiler (outside the heat exchanger ) area rather than inside the control box? So, I'm going to have to crimp up a pair of longer cables (silicon) to reconnect the inhibitor. I can't see that being a problem as the connectors will be quite low in the casing (alongside the pilot gas tube and the ignitor wire) and will be cooled by the input air flow (balanced flue)?

Thanks to all who helped / advised / questioned on this one and an extra point for those who said 'thermocouple' ;-)

All the best ..

T i m

p.s. Now Mum is talking about where to fit the new condensing boiler!

Reply to
T i m

Doh!

But if the thermocouple is generating a voltage then there will presumably (unless it's output impedance is infinity) some current generated?

So (confused here) the meter was supposed to be measuring the current generated by the thermocouple (all be it into the input resistance of the meter rather than a real load). The meter itself does this by measuring the voltage applied across it's internal shunt resistor and using various scale modifing restors to calibrate the mV (uV?) display to read (display) Amps? Could you explain (and I'm an aging electronics engineer remember) how / why that would be different from reading any current in a cct (on a DMM) please?

Hmm .. I think even I would throw *that* thermocouple away!

All the best ... ;-)

T i m

Reply to
T i m

Of course.

Sorry, I was not being clear. Of course it's not different. My point was that it's not a 'real' measurement, it's affected by the test instrument to a great degree. On 2mA, it'll read .25mA, on 200uA, 30uA, and on 20mA 20mA. (or so).

True. But I've found it really handy in the past to measure leakage currents, when the uA range won't quite cut it. A 3.99 meter with a 1M input impedance will resolve down to 100pA when set on voltage. (care does need to be taken that you're getting accurate measurements.)

Reply to
Ian Stirling

Could you explain (and I'm an aging

Cheers Ian ...

T i m

Reply to
T i m

Hmmmm. Yes and no. There is movement of electrons from one metal in a differentially heated thermocouple to the other, so Yes there is a current. I expect this varies with the imposed resistive load and tops out somewhere, so meter readings may vary depending on their resistance.

Believe it or not my physics teacher had a gas-powered radio. It was Russian-made (or at least the notation was in cyrillic) and had a bank of thermocouples over a gas flame and a heat sink at the back to keep the junctions at different temperatures. The radio used thermionic components, naturally, and the case looked rather like the radio had been involved in a shoot-out at Stalingrad. It was certainly a good way of impressing a class of teenagers. The next week a gas-powered refrigerator was our object lesson in thermodynamics.

formatting link
a useful detour possibility if you are in the midlands.

John Schmitt

Reply to
John Schmitt

An open circuit voltage of 25mV would indicate a working t/couple. The gas valve solenoid+t/couple might typically total 1 Ohm thus giving about 25mA current.

If there is some doubt about the t/couple and it gives 25mV when heated, I'd reckon on it being OK.

Reply to
Ed Sirett

That's what I think I discovered when I compared old with new? I thought I was going to see a vast difference (considering how old it could have been) but apparently not?

As I mentioned on the OP, the results were intermittent with the pilot staying on for 10 mins (before we did summat that upset it) or not at all?

In hindsight I don't like the condition of the spade connectors between the thermocouple and the overtemp stat (something I can resolve when making the new longer cables to fit the 'different design' replacement thermocouple that I have tacked in (and and seems to work?)).

I might even pop the origional one back in and see what happens (it's all easy to get at).

All the best ..

T i m

Reply to
T i m

In article , T i m wrote: [snip]

Very similar symptoms to those we had with our factory hot-air heater, trying to relight it after being off all summer. Perched on the top of a 12ft ladder in my case.

Press and hold button, light the pilot, wait, release the button, looks ok. Turn on the main burner, looks ok..... then the whole lot goes out. Repeat, and repeat, and repeat, etc..... until it finally works.

After a few years of this I finally cottoned-on to what was happening. The main feed pipe had been shut off at the meter, for about 6 months, and air had crept into it. The solution was to simply purge the pipe beforehand.

Reply to
Tony Williams

At least you found a solution ;-)

So, I popped into Mum's again this morning with the new spade connectors and silicon wire to make some new leads to go between the over-temp stat and the thermocouple connector. I made one lead and first used it to short the thermocouple connectors just to make sure all was still ok .. it was. I then went to 're-dress' the thermocouple neatly when I felt it 'give' where the thin copper tube joins the bottom of the thermocouple itself (doh .. don't you just hate it when that happens ..). I'm normally very carefull with such things (nice easy bend radisus etc .. not sure what happened there ..) ;-(

Anyway, I ripped out the new thermocouple and re-fitted the old one (still not convinced that the old one was 'that bad') shorted the inhibitor connector with one of my new leads and all was still ok ;-)

Re-fitted it all neatly, ran the two new leads between inhibitor connectors to 'over temp' / inhibitor device and everything was sill aok. So, I think it must have been bad (hr) connectors?

We put back all the covers / panels etc and ran the bolier properly till it had fully heated the cylinder and all was still fine..

So, fingers crossed I suppose on this one .. will it now last till it's replaced with the new bolier soonish ... (I feel it will)?

All the best ..

T i m

Reply to
T i m

Of course the meter is measuring the current in the circuit accurately, but the measurement is not a reflection of anything useful.

Reply to
Ian Stirling

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.