Anyone understand combi boiler control systems?
newshound
am not particularly keen to replace it at the moment.
Central heating works OK, so pump and fan, and their sensors are fine. Gas
valve expects 3 - 20 volts, starting at about 10. Starts up fine and
modulates up to 18 or so, water gets to the right sort of temperature, no
sign of boiling in the primary HX etc.
With CH off, DHW goes through proper start up sequence, lights burner with
11 volts on gas valve and runs for 10 seconds. Then the voltage on the gas
valve jumps up to 19.8 and shortly afterwards the ignition lock-out cuts in.
This is not the ultimate trip from a thermocouple on the primary HX outlet
pipe, it's the trip which comes in if sensors indicate that the fan or pump
might have failed. An obvious theory, suppose the DHW heat exchanger is
blocked, then the primary water will get too hot, this gets sensed by the
DHW thermistor. I'd expect this to modulate down the gas valve, but perhaps
it has a trip level as well. But, the water in the diverter valve isn't
getting particularly hot in 10 seconds of running; a few seconds after the
trip (stopping the flow as soon as it trips), the thermistor resistance is
down from 10k to 8k, corresponding to about 35 degrees. Not a faulty
thermistor, I have changed it anyway, and both test out OK.
Obviously the Pactrol control board? Except that I have a spare board, and I
get *exactly* the same symptoms with that.
If you switch the CH on and let it warm up, then operate DHW the diverter
valve actuates to give precedence to DHW and it functions OK, modulating at
15 to 18 volts or so on the gas valve depending on what you have set the
temperature on the front panel, and giving a decent flow of nice hot water
(so I think the secondary HX, which I have descaled, is reasonably OK
I'm speculating that high resistance at one or other of the connectors
between the thermistor and the control board might persuade it to give the
gas valve full voltage a few seconds after startup, and then some monitor in
the software is saying "gas voltage at top limit, better trip". It's not
particularly easy to "rewire" this connection because of the multi-way
connectors, but I might have a go at this next. In the good old days there
would have been a preset on the board somewhere for setting these
parameters, but I suppose these days it is done within the software?
Or am I missing something obvious?
Steve
Arfa Daily
"newshound" <news...@fairadsl.co.uk> wrote in message
news:8740g8...@mid.individual.net...
That's extremely similar to a fault that my daughter's combi suffered a
while back, and which Geoff, the expert on here helped me to track down.
It was being caused by an erroneous reading from the PHX temperature sensor.
Measuring at the connector on the PCB where this sensor came
in, you could see a significant jump in the voltage, as the boiler cut back
out. It turned out that the actual connector into the sensor itself was
intermittent internally to the sensor body, and I managed to get a temporary
fix by wedging it with a thin piece of cardboard. Obviously, as you've
replaced the sensor, it isn't going to be the exact same problem, but Geoff
said that it was common for the loom itself to give problems with breaks
at connection points. I managed to prove the point when I was sorting out
the diagnosis on my daughter's one, by disconnecting the sensor plug
at the PCB, and hanging a pot across the pins, set to a suitable resistance
to 'fool' the control processor into thinking that the sensor was present,
and indicating that the set temperature had not yet been reached. Obviously,
you can't leave it running like that for long before you would get a
genuine overheat condition, but long enough to prove that the boiler stays
lit.
Arfa
newshound
"Arfa Daily" <arfa....@ntlworld.com> wrote in message
news:Uo6Pn.16082$c_2.862@hurricane...
Thanks, that's *very* interesting. Oddly enough, another test I tried
earlier was to put a resistance decade box in place of the thermistor and
fire it up with 10k, then drop it 1k at a time in an attempt to simulate
what the thermistor might be doing. But it still tripped in a similar way (I
didn't have the gas valve voltage monitored at the time). All of which seems
to point at a fault at the loom to PCB connection. Time for a bit of
surgery, I think.
Steve
Engineers will go without food and hygiene for days to solve a problem.
(Other times just because they forgot.) And when they succeed in solving the
problem they will experience an ego rush that is better than sex--and that
includes the kind of sex where other people are involved.
Lifted from http://www.unixnerd.demon.co.uk/engineer.html
geoff
<arfa....@ntlworld.com> writes
I thought that the Finest did have a pot to set the DHW temp, but you're
not going to resolve the problem by twiddling with this, you should
leave it alone as it would have been correct before the fault (as shown
by both boards showing the same fault)
The reason it modulates up to full and then down is because the
temperature sensor is after the HE and it needs to be hot enough to give
useful feedback. The NTC should be about 2.5 to 3k at "hot"
are you sure that the CH is working correctly
give individual wires on the loom a tug check for elasticity- the
insulation is crimped as well as the wire, it is easy to miss a broken
wire
What is the flow like?, Is the pump working
The fact that you've descaled the HE doesn't mean that you don't still
have a blockage
--
geoff
newshound
Many thanks for the thoughts. Will hunt for bad wires first (I've done the
obvious check, of course), and failing that, dig out the data logger again
and do some more monitoring!
Steve
>
> I thought that the Finest did have a pot to set the DHW temp, but you're
> not going to resolve the problem by twiddling with this, you should leave
> it alone as it would have been correct before the fault (as shown by both
> boards showing the same fault)
>
> The reason it modulates up to full and then down is because the
> temperature sensor is after the HE and it needs to be hot enough to give
> useful feedback. The NTC should be about 2.5 to 3k at "hot"
My understanding is that the control system should ramp the gas up to full
(or to the appropriate "control" level) over a period of (say) 10 to 30
seconds, presumably because it is taking readings all the time and wants to
make sure everything is working OK. In this case, the gas valve voltage
starts at 11 volts for ignition, and after a few seconds flips straight to
19.8 or something like that faster than an analogue meter can follow. When I
was investigating a previous problem you could see that the ramp wasn't
actually linear, it was a step function going up a few tenths of a volt
every second or so (the numbers might be out by a factor of 3, but that was
the basic shape of the function (as you might expect would be generated by a
digital system).
>
> are you sure that the CH is working correctly
>
Well no, but at least it works. While the DHW doesn't work at all.
> give individual wires on the loom a tug check for elasticity- the
> insulation is crimped as well as the wire, it is easy to miss a broken
> wire
Good point, I will do that check. But I'm almost certain it's worth
by-passing the suspect pair of wires. I'm also slightly suspicious whether
the connectors are gripping the "pins" on the thermistor/NTC and the board
sufficiently: I had a fault before at the thermistor end (on the CH
thermistor).
>
> What is the flow like?, Is the pump working
Well, it's operating the flow switch, and seems to be pumping hot water
round the rads at the normal sort of rate. (I have been tempted to add a
flowmeter to that circuit).
>
> The fact that you've descaled the HE doesn't mean that you don't still
> have a blockage
>
Agreed. But when I can keep the gas alight while running DHW, e.g. after
running the CH for a bit, and leaving it "running", I get a decent flow of
hot water through the DHW. When I had a really blocked secondary HX I was
only getting tepid water, and then the gas modulated down because the
thermistor was sensing high temperature on the primary side.
> --
> geoff