Boiler intermittently fails to light

[Bob Smith (UK)]

It's a Ravenheat RSF82E. It's a fanned combi, the kind that lights every time there is a demand for heat (ie no pilot). I have a manual / website for the manual if that would help.

About 1 in 50 times, it fails to light when the CH water cools off a bit. It is part way through the lighting cycle that it fails, the big "clunk" that the gas valve makes. I am not sure what is causing it though, and have

3 theories.

1) The gas valve is sticky (although I hear no failed mini "clunk") and fails 1 in 50 times.

2) The fan is faulty, and does not draw enough air for the boiler to think it is safe to light (assuming the boiler is that smart). The fan does sound scraping / rattly when it slows down. A gust of wind could shift a borderline fan to a "not fast enough" state.

3) The sensor that detects the fan is working is faulty / blocked with fluff (assuming it is that smart).

Is there anything else that could cause this? Any quick and cheap (eg tap gas valve with hammer, hoover sensor out) fixes that cure common problems?

The boiler would work fine if it had a "retry" built into the lightup sequence, but as it is, I keep finding the boiler with a light lit up, merrily blowing cold air past the CH water (the fan comes on before the gas) chilling it. Turning the thermostat down then up, or switching CH off then on restarts the lightup sequence. The proper "reset" botton does nothing (not lack of flow/boiling water problem).

Extra info/hint: Sometimes if I am there when it fails to light, me doing the the on/off thing causes it to try again, but it will fail to light 2 or

3 times in a row, suggesting it is some condition causing the failure that goes within 30 seconds.

Any ideas / suggestions?

Thanks

Bob

[Ian Stirling]

Essentially all boilers do.

The sensor is usually a differential pressure switch. It's got a diaphram inside it, and is connected so that airflow causes a pressure difference. The hoses can get blocked with stuff, though this is fairly rare.

[Bob Smith (UK)]

All boiler's fans sound scrapy, or all boilers fail to light if a gust of wind confuses the "fan running" sensor?

Bob

[Ian Stirling]

Oops, I meant to snip more. All ones 'are that smart'.

[Bob Smith (UK)]

I have had the front off to get at the pressure switch. the 2 flexible pipes are clear and have no splits in them. I cleaned the 3 electrical connections (took them off/on several times).

The pressure switch seems to measure the pressure difference between the back of the sealed box (a connector on a collar on the outer pipe of the flue, open to the combustion chamber), and the front of the sealed box (a 4" pipe pointing towards the front of the case, hanging in space) with no seals in between. It does work, since the boiler exhibits the same lighting up failure symptoms with the front off.

This leads me to believe the pressure switch is borderline, causing the thing to not work occasionally. Do these switches fail? You said clogging of the pipes is rare (no actual air flow = no crud deposits).

If the pressure switch is a likely culprit, I will nip up to Ravenheat and try to get one. I live in Morley, so that is not a problem.

Bob

[Ian Stirling]

If it was me, I'd stick a neon/LED/voltmeter on it, so I can monitor it and see if it actually doesn't work. However, replacement will be harmless. While I've got the cover off, I'd drop a drop of engine oil into the fan bearing (if appropriate) might be an idea. If it quiets it a lot, then it's very likely that the bearings are going, and it might be the fan that's the problem.

[raden]

In message , "Bob Smith (UK)" writes

That's not where the problem would lie - the connections you have to check are the switching connections of the microswitch. You can only really do this with a meter. You need bugger all ohms in both positions. You should check that the diaphragm hasn't split while you're at it

[raden]

That really isn't going to tell you the state of the contacts. The only satisfactory way of testing the electrical state of the contacts is to test the resistance.

I can't off the top of my head remember if the ravenheat has phosphor bronze or ball bearings.

If a ball race DON'T oil the bearings, they run on grease and oil will only serve to wash the grease out

If phosphor bronze, use light machine oil

Your problem will be getting in to the bottom bearing which I think you'll find to be fairly inaccessible

Didn't read that bit before replying, however bear it in mind. The best way to find out the state of the bearings is to spin the fan by hand, you should be able to hear / see / feel if there's any bearing wear or play

[Ian Stirling]

I was only really thinking about detecting if the switch is operating properly, perhaps the pressure threshold has slipped or something, rather than actually checking the contacts.

[raden]

Which will give you a false sense of it working if the contacts are kacked, potentially resulting in time and money wasted looking for the problem when you've decided that it's not the APS

[Ian Stirling]

If observed not to switch on when it should, when the thing does not start, then the fault is obvious.

[John]

I have had intermittent problems with pressure switches with odd causes:

1. Insect in the venturi within the fan discharge which mostly didn't detect but just occasionally did (Sime)
2. Water due to heavy rain getting down the vertical fan flue and blocking the sensing tube until it dried out (Ideal Classic) Worth looking if you can't pinpoint it :-)

[Bob Smith (UK)]

I have been in again, and cleaned the ignition electrode, as well as checking the pressure switch. I also changed the adjustment screw on the switch to make it more sensitive. The boiler still occasionally fails to light.

I can't actually open the pressure switch (It is held together by bent bits of metal, and I would probably damage a seal or something anyway) but found if I blow into one side, it clicks, and when I release the pressure it clicks again. I am pretty sure there is no split because I blocked the tube, waited 30 seconds, then let go and got the click (ie never leaked through the split in 30s).

Been experimenting again, and found a way to duplicate the non lighting problem.

When everything is working fine, it starts up the fan, lights up a neon with a picture of a flame next to it, there is a clunk, the light goes out, and the flames go wumf.

When it fails to light (1 in 50 times, or about every 1/2 hour), it starts up the fan, lights up the "flame" neon then nothing.

With the cover off, it starts up the fan then nothing.

With the cover is half on and leaky, I get the same symptoms as the problem condition every time it tries to light.

Does this mean a new pressure switch would fix the problem, even though it seems to be clicking etc? (I have no meter to measure the resistance etc)

Bob

[Ian Stirling]

It'll normally turn on the pilot light (if present, I'm not sure that there may not be some models with only a single valve, though I doubt it) only after it senses airflow. This is to prevent gas coming out when there is no air, and potentially causing explosions.

If it does not sense airflow, it will not turn on the gas, but may just carry on waiting until it does.

If you disconnect the pipes to the switch, does it do the same thing as it does when it fails?

If so, then I'd guess it's probably the switch, though it could be the circuit it connects to.

Have you blown through the pipes to the switch, just in case there is some dirt inside?

[Bob Smith (UK)]

Ian

Venturi and pipe both clear. Venturi is on the top of a horizontal flue, so no rain gets in there.

Bob

[Bob Smith (UK)]

It will sit there for hours, as though it does not have a retry/wait built into the logic.

Not tried that yet - Will have to be the weekend when I try that one.

If it ware a toss up between circuit and switch, which would be most likely to fail intermittently?

Bob

[Ian Stirling]

In general, anything with moving parts tends to be a lot more unreliable than stuff that doesn't.

Have you disconnected and reconnected the switch wires? An intermittent contact might just cause exactly this problem.

[Bob Smith (UK)]

Took them off/on several times to clean them. No effect.

I am leaning towards the switch being the problem though. I know the circuit could also be faulty, but it is probably more than £30+vat to replace. I could take it out and re-flow all the solder joints, but that would probably take an age...

Bob

[raden]

In message , Ian Stirling writes

Ha, you jest

Sorry, I seem to have missed out on this thread

What boiler are we on about ?

[Bob Smith (UK)]

Ravenheat RSF 82E

Bob