Boiler problem: Heating OFF, water ON

[Another John]

Hello all

Well it's happened.. CH heating seems to have gone off today:

- Glow-worm Flexicom HX boiler, installed June 2008.

- The water is heated at the appointed times, but the heating itself doesn't seem to come on.

- the radiators do seem to get warm, but I think that that is due to the water being heated up at the same time: when the water tank gets hot enough, the radiators go cold.

- the room thermostat *seems* to work (or does it?) in that I hear a click when I twirl it round and it passes the desired temperature.

I've never been near this boiler in terms of checking it out, partly because I'm not any kind of heating expert, but mainly because I'm intimidated by the sophisticated case.

Is there anything I can do, before succumbing to the inevitable and trying to find a CH engineer?

Cheers John

[ARWadsworth]

Faulty/sticking mid position (3 port) valve

[Tabby]

It sure would help if you could answer the basic questions you posed.

NT

[John Rumm]

So it sounds like the pump is ok...

Chances are you have a stuck valve. What sort will depend on how you setup is piped, but its probably a Y plan or S plan.

?title=Central_Heating_Controls_and_Zoning#CH_and_DHW_zones:_S-plan> - the room thermostat *seems* to work (or does it?) in that I hear a

Yup plenty, and its not too difficult either. If you can find the valve, see if you can hear that whirr when you get the stat to click.

If needs be, turn the power to the heating off, and take the motor box off the top of the valve. That should reveal a brass tap end that is normally turned by a motor. Check you can turn it freely with pliers. If its a bit still then working it back and forth with a bit of oil on the shaft will usually free it. That may be enough to fix it.

If its free, then with the motor head has had it, or the micro switch that detects the end of movement has failed.

More here:

[The Natural Philosopher]

stuck Motorised valve probably.

Are they internal to the boiler, one wonders..?

[Another John]

Hi All (OP here):

Thanks to TNP and to John Rumm for suggesting stuck motorised valve: I'll have a look: thank you very much. (It's in the airing cupboard btw TNP, thanks for wondering.)

NT/Tabby/Meow wrote:

Sorry NT: I'm not sure what you mean! The only direct questions I posed seem to be "is the thermostat working [just because I can hear it click]?", and "should I go looking for an engineer?". You're a venerable and respected member of this group, and I don't want offend, but I don't actually understand your point, sorry!

I'll have a look at the valve tomorrow morning ... I have suspected it of misbehaving for some time, since the radiators shouldn't get warm in summer, when the water heats ...but they seemed to this past summer.

Are motorised valves notorious misbehavers? I think I've had 3 or 4 replaced in the 25 years we've been in this house.

Thanks a lot for the advice so far!

John

[Roger Mills]

Yes, you have the classic symptoms of a 3-port mid-position valve stuck in the mid position.

The boiler is then controlled *only* by the HW demand, but the radiators get warm whenever the HW is being heated. The valve never moves to the CH-only position, so you get no CH once the HW demand is satisfied, even if there is a CH demand from the room stat.

Does that describe your problem accurately?

Chances are that the actuator is ok and that the 'wet' part of the valve is stuck. The actuator can usually be removed from the valve by undoing a couple of screws. Try doing the following:

1. turn off the whole heating system at the FCU[1] whilst remaining within earshot of the valve (or get a helper if they're not close enough together) Do you hear a zizzing noise, as the valve closes (returns to the HW-only position) under the power of the spring return? If you DO, it's possible that the actuator has simply got confused. In that case, turn the power on again, and re-try the system. It just might work after a 're-boot'.
2. Assuming the valve DIDN'T close, undo the screws and remove the actuator from the valve. Observe whether the actuator then zizzes and returns to a position corresponding to valve closed. If it DOES, it indicates that the valve is stuck in a way which requires more torque than the spring return can muster to move it.

In this case, try to free it by turning its shaft back and forth with a pair of pliers while dribbling some oil (or WD40) down the shaft into the bearing. If it frees up ok, re-fit the actuator, turn the power back on and re-try the system. If you can't move the shaft, it's new valve time - which requires a partial drain down of the system (unless you've got three strategically placed isolation valves surrounding the mid-position valve.

Let us know how you get on.

[Roger Mills]

Sorry, I forgot to add the footnote . . [1] It is important to remove *all* power from the system. Simply turning the HW and CH off at the programmer isn't sufficient because the actuator will *still* be receiving some power from the programmer's HW-off output.

[Andy Champ]

Yes. One of the guys at work had one go last week, and we all knew what it was; and there's a another thread on the same subject an hour behind yours.

Andy

[John Rumm]

They can be. Especially on vented systems with lots of crud floating about in the works. A good flush and then maintenance with corrosion inhibitor helps in these cases.

[Dave Liquorice]

Not all three port valves have spring return. I pretty sure that the one in my flat didn't. It made the same motoring noise between all positions and would stay in the last position it was in if you powered the system down. Don't ask me make/model I moved out of that flat about 15 years ago...

[Roger Mills]

That is true. I thought I had read that it was a Myson 322 - which definitely *does* have a spring return - but now realise that that was in a different thread!

Spring return valves are the most common though - so it's very likely to be one of those. We'll see what the OP reports back when he's checked it.

[Another John]

OP here: progress report:

First - *Thanks* to Adam, John R and to Roger for directing me to the

3-way motorised valve.

Second: Well I've found out how to take the cover off! :-) I had to use a mirror to locate the screw, and then it took me 20 minutes to realise that the screw simply locates the cover, and actually taking it off involves springing the sides away. You now have an impression of my competency.

Having got the cover off, it was easy to remove the motor head, as described by the advisers.

What I SHOULD have asked is: how much movement on the spindle am I looking for? It does move, but only about 20-30 degrees. I'm aware that that may be the full movement required, so I'm not willing to force it further!

Also: it moves, but fairly stiffly. However, I'd expect this with a brass fitting moving inside a water pipe.

Also: the motor head does have a spring, which actuates when the head is removed from the valve. Placing the valve back in place, I can push the spring firmly side to side, but when the power is back on to the system, every thing goes "loose" -- there's no resistance on the lever, either way.

Finally, the pump doesn't now seem to switch on at all when heating is selected -- previously, the pump came on and stayed on, I presume trying vainly to find some hot water to shove round the system.

So in short: how freely should the spindle on the valve move, and through how many degrees?

Thanks so much for the help so far!

John

p.s. pics at

(front view) and (back view)

[Roger Mills]

I'm not 100% sure about the Honeywell 3-port valve (having only owned other makes) but I would expect it to have a total movement of 90 degrees (45 degrees either side of the mid position).

The motor and spring are not *that* powerful, and if the valve is too stiff, they won't be able to move it. If you've got strong fingers, you should be able to rotate the shaft by gripping it between finger and thumb - or certainly by using just *light* pressure with a pair of pliers. If it's stiffer than this, it needs working to and fro until it frees up.

I'm not clear from what you said whether the actuator moved back to the HW postion under spring power as soon as you took it off the valve. Did it?

The lever only moves the valve to the mid position - not all the way to the CH position - to facilitate filling the system. When the valve is at the mid position or beyond, the lever goes slack - this is normal.

You can check whether you are getting the full movement of the valve, as follows. With the actuator fitted and all power off, the valve should be in the HW position, and it should require force to move the lever.

Move the lever over its full travel, and hold it there while you remove the actuator from the valve. The valve will then be in the mid position, and you should be able to turn it beyond that position by the same amount that it moved in getting from the HW to the mid position.

[John Rumm]

Don't worry - we have all been there!

I expect the wet bit of the valve is something like a butterfly valve - so you would certainly get less than 90 degrees of rotation, perhaps 60 or a bit less.

It ought to be movable with your fingers - but perhaps stiffly, and relatively easy to move with pliers.

While you have it open you may as well oil the shaft and work it back and fourth a good few times. That ought to free it a bit and also let you feel where the end stops are.

When the motor drives the valve to one end, the spring is compressed and hence you won't feel it. So how it feels will depend on demand.

If you run the valve with the lid off (warning - mains voltages present!), you should be able to see it move in response to the demands of the room stat and the cylinder stat. (note the demands from the various stats will only come into play when the programmer has that particular channel turned on).

Do you have access to a multimeter or a decent volt stick (note not a neon screwdriver)? If not a small digital meter will only cost you a fiver and is invaluable for working our what is going on.

Have a look at the way these things are normally wired (note we even have your model of valve!)

the pump and boiler to fire, you need volts on pin 8 (or equivalent

- you may have a neat wiring centre board someplace where the terminals numbers match the diagram, or more likely you have a box full of chocky block terminals!)

If you set your meter to read AC volts on a range high enough to cope with 250V, stick the black lead on a neutral somewhere, you should be able to follow through the CH demand and work out where the fault is. First turn the HW off at the programmer so we can focus on just CH. You should see 230V on pin 4 going to the room stat. Twiddling the stat up so you hear it click and start calling for heat ought to make volts appear on pin five, which is connected to the white wire into the valve. That should cause the valve to romp over to port A. In doing so it should hit a microswitch, which causes the orange wire to go live, and its this that kicks off the pump and boiler. Note with a more modern boiler like yours the pump will probably be fed power by the boiler rather than as shown on this diagram[1]. However its the same signal that causes the boiler to fire.

[1] The boiler tends to control the pump on more sophisticated systems to allow it to have pump overrun - i.e. to keep water circulating after the boiler has stopped firing. This circulates the hot water out of the boiler and stops the heat being wasted by venting out its flue. It may also include anti "short cycle" logic in the boiler - hence it may not jump into life the moment you hit it with a demand - especially if its not long since the last demand.

[Another John]

OP again:

Thanks again to all, in particular John R and Roger M.

John: I'm' afraid I'm no good whatever on electrics. I did languages at schoo, and The Arts at Uni ... when you start talking electrics, it's as though you suddenly switch to talking in Mandarin! Sorry -- not very good for a DIY-er I know!

However, piecing together everything that you've kindly sent to this thread, I conclude that the head of the valve has to be replaced.

I can move the spindle pretty easily between two extremes -- it's not

90degrees, but there's a definite stop at either end, which feels like it might go bang if I force it.

When the power is off to the head, the spring return is evident, and I can move the lever -- against resistance -- to the other end of its travel.

As soon as power is back on, the lever goes slack, indicating that the valve is at the mid-point (right?), from which it does not move.

I do not think -- having gone through all the advice, that I can fix this, so I thought I'd go down to the local Plumb Center[sic] tomorrow and see if they have a head, if not I'll have to order one from Screwfix or someone.

Finally: I know a lot more now, thanks! I've also discovered (what lots of folk here knew) that the Honeywell 4073A is probably the most popular, and most venerable, bit of kit in British CH systems!

Cheers for now: in the interests of the community, I will report back again when/if I get this problem fixed.

John

p.s. We're not entirely without heating: we get it when the hot water heating comes on!

[John Rumm]

Fair enough ;-)

Having said that - this sort of diagnostics is about as simple as its going to get. With a meter set to read AV volts, you can clip one probe on a neutral, and use the other to see if the wire in question is live or not (as these are the only two states that matter with these valves)

Perhaps we should do a basic fault finding with a multimeter article for the wiki, you can tell us if it makes sense!

The fact that the motor goes to the mid position suggests that its not totally knackered. However it is possible its switch has died and so the signal to fire the boiler is not being passed on when the valve achieves the right position.

Good. That suggests the wet bit is ok.

Yup that moves it to the mid position.

If you turn the hot water "off" at the programmer, and the CH "on", then the valve should go from sprung to slack and back again as you turn the room stat up and down. If it stays sprung all the time, it might suggest a non working room stat.

Not a bad idea, even if it turns out something else is broken, having a known working valve will make finding it easier.

Yup, please do.

No harm in extending the hot water heating period at the programmer for now. Once the cylinder is hot, it won't take any heat out of the CH water, and the rads ought to work ok even if a bit slower due to getting less flow than normal.

[Roger Mills]

Except that cylinder stat will turn the boiler and pump off when the HW is hot.

The way to fudge it would be to remove the head from the valve (but leave it wired up), manually turn the valve to the CH position, and turn the cylinder stat up to maximum. That way, the boiler and pump will run and will only heat the radiators. In this situation the room stat won't do anything of course, so control is purely manual - needing to turn everything off when the house is hot enough - not forgetting to re-fit the head and adjust the cylinder stat when HW is needed again.

Not ideal, but . . .

[Another John]

Bugger: I knew this wouldn't be easy!

I went to Plumb Center, and blow me down if they didn't have the exact component in stock and at a good price: replacement head 4000-3916-002

I get down to replace it and find: the old valve head has FIVE wires (and also a circuit board) whilst the new one has only THREE wires (blue brown and earth) (and no circuit board).

The five wires in the old one are orange, white, grey, blue and earth(green). They connect into a box which contains a very orderly rat's nest: pic at

'm now scouring the web for clues as to how 3 into 5 might go...

John

[Another John]

Replying to my own progress report ... (Look away now if you're bored with this, I wouldn't blame you) ...

The label on the back of my valve says it's a V4073A[1039]; it quite specifically states "Use replacement head 4000-3916-002".

I find, going to the site uk-plumbing , that there are different versions, and that

4000-3916-003 is the one needed for V4073s (-002 is the one needed for V4044s, it says).

Plumb Center are now looking to see if they have a 003, and it remains to be seen if that has five wires.

I can't understand why my valve has a label with what is apparently explicitly wrong data on it.

John