Central heating not working...

[JoeJoe]

Have a 5 year old WB combi that is regularly serviced and works fine (I believe).

In the last two days we experienced two problems:

1. The hot water cutting off during a shower. Turning the water off, waiting for a few seconds and turning it back on resumes hot water (until the next time...).

1. Heating that cuts off at random times, and does not seem to respons to demand signal from the programmable room thermostat.

We are using one of these:

I believe is a rebadged Honeywell basic(ish) version. I fitted it

2 years ago to replaced a failed Honeywell 927.

Earlier today I tried to create artifical demand for the heating/boiler by setting the target temp to 30c, but the relay failed to turn itself on (led stayed off). Pressing the manual/override button on the relay box turned the led on and the boiler started.

A few hours later the heating was once again not working as expected, so I repeated the above. This time the relay failed to start - the override button does not seem to do anything.

The display on the controller is bright and clear, and there is no indication that the batteries are low (there is usually a message).

- My gut feeling is that the relay box is dead. Any thoughts?

- Why would the hot water keep on cutting off then? because the boiler receives random requests for heating from the faulty relay during a shower?

- Any other suggestions as to the cause?

Many thanks in advance.

[JoeJoe]

Just a quick update:

Just checked, and the relay box seems to have come back to life. Tried is again after a few hours (both by creating demand by setting very high target temp and by pressing the override button on the box), and it works as it should, and the boiler is turned on.

The faults described above have now happened at least a couple fo times over the last few days, but then, on occassion, everyting seems to work as it should.

[Tim Watts]

Does the boiler have a local indicator that shows "heating demand"? Some WB's do and some don't in my very limited experience.

Either way - the hot water is independent of any heating controls assuming the boiler is getting power. A sudden cutout suggests a problem in the boiler more than in any external(?) controller.

Are you in a position to safety test the output of the roomstat/controller when it claims to be asking for heat (ie is the wire to the boiler live?)

[John Rumm]

A multimeter on the end of the relay box will tell you.

No normally the DHW will take priority over any heating demands.

This sounds more like a over temperature on the primary CH circuit. Could be an obstruction in the boiler (or on the primary side of its plate heat exchanger). Possibly a dodgy flow switch.

[JoeJoe]

Sorry, I am not sure.

Good to know - thanks for that! I already phoned WB and arranged for them to send someone to have a look at the boiler on Monday (it is still under warranty).

I later noticed that the relay/controller were misbehaving (as described above) and thought that perhaps they were the cause, rather than the boiler itself.

I can, but see above - I would probably leave it to the guy coming to look at it.

The wolseley 10 minute drive away has the controller/relay in stock, so I can always get them later if they prove the be the cause.

[JoeJoe]

See my subsequent message - there is clearly something strange with the controller/relay box...

That is what I thought.

So most likely boiler problem then? Good, as WB are sending someone on Monday to have a look (still under warranty).

I just wanted a quick sanity check that it was not 100% the controller/relay that I can easily replace myself.

[Tim Watts]

Could be worth wiring a neon to the output at a convenient time (like a grid module in a single gang box) - very handy for debugging problems.

Good luck - hope it gets fixed :)

[Harry Bloomfield]

John Rumm used his keyboard to write :

These problems are precisely the reason I put together a row of LED's permanently wired to terminals in my heating system's joint box. I can tell at a glance exactly what is happening. Power on, pump running, call for heating, call for hot water, call for boiler to fire up.

It took a while to make it up, but when there is an issue, I can see what is happening with just a glance.

[John Rumm]

There is a lashup you can do with the normal Y plan system and some relays and neons that will achieve this as well (it also saves continuously powering the valve when all calls for heat are satisfied)

[Harry Bloomfield]

It happens that John Rumm formulated :

My LED's give more detail and I would expect them to last rather longer than neons, which in time will dim.

[Brian Gaff]

It might be an idea to check all electrical connections for loose screws etc, and also see if the relay is merely sticking. Other than that I guess there could be some odd dry joint somewhere in the electronics, but it would be a swine to find it. Brian

[DerbyBorn]

I was going to suggest that - the shower will be causing to to run flat out and maybe it is overheating and shutting off a thermal switch.

[tabbypurr]

They don't dim out if you halve run current with a series resistor. Neon life is generally inversely proportional to current to the power of 3. So half the current gets you 8x the usual life of decades.

NT

[Davidm]

How are you powering the LEDs from the 240v ac points you are monitoring? An individual "circuit" for each LED or something cleverer?

[John Rumm]

A diode and resistor in series, or capacitive dropper and a diode will normally do it...

[tabbypurr]

capacitor needs a resistor as well.

NT

[Harry Bloomfield]

Davidm expressed precisely :

An individual circuit per LED, it needs just a diode, 300v capacitor and the LED. You can then connect them between test point and either of neutral, or live - which ever way produces lit for active. I made it up with 8x LED's. I posted full details of the circuit I used on here, maybe a couple of years ago.

I built it up on a long strip of Veroboard, then mounted it in a two gang surface socket box, with a two gang blanking plate, mounted next to the heating joint box.

[Harry Bloomfield]

John Rumm presented the following explanation :

Using the resistor method, the resistor will have to dissipate a lot of watts, much more efficient to use a capacitive dropper. I will see if I can dig out my out circuit, the one I ended up with after a bit of bench testing.

[Davidm]

Thanks Harry :)

[Harry Bloomfield]

Harry Bloomfield expressed precisely :

L ---- 220 -- 0.1 -- 4148 + LED (back to back) ---- N

That is a 220Ohm resistor, a 0.1uF 630v cap, then back to back a 1N4148 diode in parallel with a superbright 6mm LED.

I found the cap value by some trial and error, decrease its value if you use with a lower current LED. 0.47uF blew my test LED after a couple of days on test. With 0.1 they have so far survived 2 years running.

The 220Ohm is intended to limit the inrush current as the cap charges. Almost any small diode with do for the 4148.