3 port valve question

[T i m]

Hi all,

I got called over to the elderly lady across the road today as she said her central heating was coming on along with her hot water.

I replaced the timer a good while ago (the plastic had gone brittle on the old one and the internal battery had failed) so I was reasonably sure it wasn't that.

Having a play with the 3 port valve I think there could be an issue with that and a quick Google to faults on them suggested a couple of things to check and I did:

I removed the head (two screw Honeywell type luckily so no water loss) and I could move the actual valve with my fingers. However, it doesn't seem to go though that many degrees end-to-end, but does seem to stop fairly sharply at each end?

If operated by hand with the system calling for heat (I turned the HW stat up), when you turn it one way it seems to do so (diverting the water) nearly silently, whereas when turned to the opposite extreme it seems to do so with a 'bonk' (as if the water flow pressure is flapping it shut, like a lock gate). ;-)

With the head off, if you gently reverse drive the motor with the side arm, it allows you to push it to the end of the slot and then comes back slowly on it's own under spring pressure (as it should)?

However, with it fitted to the actual valve, it only seems to come back half way?

So, is the valve itself unable to rotate fully for some reason? Something caught in there (some calcium?) and should I put an adjustable on there and give it a firmer 'nudge' in the hope it might shift it?

Cheers, T i m

[Roger Mills]

Have you at any point removed *all* power[1] to the system? If not, try doing that. These valves can sometimes get confused, and need a "re-boot" to fix them.

[1] Turn it off at the fused spur, or wherever it's powered from. Simply removing the demand from the programmer isn't enough. With all power off, you should hear the spring return moving the valve to the HW position.

[T i m]

<snip>

Yup. The main isolator is in the airing cupboard along with this MV and the timer. ;-)

Understood.

As I mentioned though (although not that the power was off admittedly, even though it was etc) the manual override arm does return to the bottom of it's slot when it's not connected to the actual valve head?

So, either the spring tension isn't sufficient to overcome any friction in the valve or the valve is sticking somewhere?

But looking at how little rotation I could get from the valve with my fingers / Leatherman, it did feel like the latter. That said, the 'valve' seemed to come to a neat stop at both ends of it's rotation but that rotation (angle) didn't feel like it was proportionate to the arm movement when the head was disconnected?

This was all the rotation I could get (without forcing anything anyway).

Running the system with the MV head off and turning the valve manually wasn't that revealing as the system was already hot and once hot it didn't cool that quickly.

Cheers, T i m

[www.GymRatZ.co.uk]

On 02/04/2019 08:33, T i m wrote: <snip>

Did you see the following page?

I recently had to replace 2 Honywell 3-way valves in the shop which had been in service for around 17 years! Replacement valves were the same length and wiring but the threads were different so had to cut olives off. Partially cutting through the olive so as not to damage pipe then putting a flat-blare screwdriver in the saw cut and rotating fractured the remaining bit of olive without risk to pipe.

I'd have thought the spindle should rotate significantly further than the few degrees in your picture.

Might be worth changing the valve anyway if it's been in used for 10 years or more, and I'd suggest changing nuts/olives anyway to ensure everything mates up nicely. The old nuts on mine screwed on 2 or 3 rotations onto the valve body before jamming which initially looked like they matched.

[John Rumm]

The spindle is connected to a small arm with a ball on the end of it. The rotation just needs to shift the ball from the opening at one side of the valve to the other - hence it does not take much rotation since the arm only needs move through a small arc to work.

The motors (if working) have a fairly decent amount of torque.

[Harry Bloomfield]

I don't know about the Honeywell, but all of the 3-ports I have come across can be freely rotated through 360. With all power off, the valve and actuator should spring return all the way back, if they don't then either the actuator or valve is sticking/jamming.

Spring return 3-ports actuators are notoriously unreliable, in my experience. Two of the positions, require the motor to be powered, but stalled to hold the position. Better and more elegant are the Momo type, which motor to position, then switch off completely. Which is the type I have now fitted, after decades of problems with the spring return types.

[T i m]

<snip>

That makes much more sense than me thinking it was like an isolator ball valve as found on a washing machine feed tap. ;-)

Is it just me that would have this picture on the wall. ;-)

Understood.

So, from what she said, the CH is (also) coming on when it's set for HW only.

This would suggest that the CH outlet isn't being (fully?) blocked when in HW only.

If I understand it right, the default / un powered position of the valve is normally providing HW and therefore means the CH outlet is (supposed to be) being 'blocked' by the ball valve (using spring tension alone)?

So, it's either not being driven correctly by the timer (still / also feeding power as if it's calling for HW and CH) or just isn't seating properly?

Cheers, T i m

[T i m]

If you can rotate these 360 the valve would have broken off! ;-)

Agreed. In the un powered position I believe it is HW only and so should have the CH port blocked. If it isn't you would get 'some' CH when in HW only and if it wasn't particularly cold out, could appear to be full CH.

I think when in full travel the motor is just mechanically stalled but when in the central position it is being stalled / held as you say.

And can they be wired straight into a Honeywell Y Plan (I think it is) with no other modifications do you know?

Cheers, T i m

[John Rumm]

Perhaps, or for some reason it is also "seeing" a CH demand, and hence goes to mid position.

See the circus diagram:

Yup:

Yup, those seem the obvious options.

[tabbypurr]

I'd expect the most likely problem to be a stiff shaft. We know the motor runs, as it's not stuck on HW only. Next most likely failure AIUI is the shaft sticking in the valvey bit (not the motorised head). So you've got a motorhead with a stuck shaft.

NT

[T i m]

<snip>

I was thinking of slackening of putting a couple of washers under the two screws that hold the head to the body and making up a wire pointer that can extend past the outside of the body. That way you can see exactly what state the valve is in (for testing)?

Yeah, I got all that from GymRatZ link and associated video etc.

I was just wondering if there was a 'common' failure mode that gave those symptoms?

Cheers, T i m

[T i m]

<snip>

Except I can easily turn it between end-stops with my fingers?

TBF we don't, we only know that the rads also get hot when the HW is called?

Covered above.

The manual arm thing isn't directly connected to the valve arm / spindle so you get a distorted idea of what is *actually* happening in there.

Reading around, with the power off, it suggested that if you pushed the manual lever to the far end against the spring, it should slowly come back under spring tension. With the head disconnected it does but with it connected it doesn't (but the valve can be easily actioned with your fingers). It also seems it may also depend on 'other' things but I haven't worked out what they are yet?

Cheers, T i m

[tabbypurr]

The valve moves under its own spring power to the HW only position. I'm thinking if it's not in that position, the shaft has to be too stiff for the spring to move it back. Your account seems at odds with that I know. Why don't you move the thing manually to the HW only position, switch CH off then power it up, check its position. Then turn CH on as well as HW, turning the room stat up, and see if it changes its position.

NT

[Cynic]

At the end of the day carefully perusing eBay yielded me a new, unused, old stock clearance genuine Honeywell valve complete sub £20 A couple of months ago. You might be wasting a lot of time and effort on an old valve with worn drive parts in the head.

[T i m]

On Tue, 2 Apr 2019 17:36:23 -0700 (PDT), [e-mail] wrote: <snip||>

Check (it should / might be).

Well I can confirm the shaft isn't in any way seized as I can *easily* flop the valve backwards and forwards with my fingers.

Well, you can't easily see where the shaft is pointing when it's all in place and the manual override arm isn't a direct translation of that either, so, the fact that there seems to be a difference between how the arm reacts with the head disconnected from the valve may not indicate a / the fault.

As I mentioned elsewhere, I think I'll try to space the head off the valve slightly with some washers and leave a wire pointer in there to give me a visual indication of the actual valves position at any time.

No room stat mate. There is only a cylinder stat and thermostatic rad valves so relies on the return temp on the boiler to manage the CH.

What I can do is with the system cold, take the head off, put the valve in HW only and just run the system up. If the CH gets hot as well then it *has* to be the physical valve. ;-)

Cheers, T i m

[T i m]

Ok.

Oh, indeed. I'm not particularly fussed to repair the MV, just that I might if it's easier (not having to drain down etc), once I know where the issue is etc!

Cheers, T i m

[Roger Hayter]

Many years ago I had a problem with one of these valves where the rubber ball that actually occludes the ports had swollen up and didn't fit any more. I don't know if this stll happens - or whether it was an unusual fault at the time, but it made the valve very leaky.

[T i m]

Ahhh ..

This is the sort of thing I'm wondering Roger, as I learn more about how these things work.

I understand the ball is designed to be allowed to rotate so that (in theory) a fresh 'face' can be presented to the valve(s) ... assuming it can still and wants to rotate etc?

If the ball is distorted and can rotate then in theory the 'fault' (leakpast) will still happen but would be random. However, I think there may be a stronger closing force applied by the motor (so that would be the CH only position) than the return spring might be able to apply in the HW (only) position, meaning the fault more of that cause is more likely to see unwanted HW than CH?

When I was playing with the valve spindle with the system (boiler / pump) on, turning the shaft to the right (that I think might be HW) definitely caused the ball to seat with a 'thump' (like water hammer) whereas to the left it was hardly noticeable [1]. Maybe the valve being forced shut (in that direction) could have distorted it?

Time will tell. ;-)

Cheers, T i m

[1] Before learning the 'valve' was actually a ball on a stick (like a ball c*ck) rather than a ball valve (like a WM isolator), it 'felt' more like a simple metal flap like the old hot / cold diverter flap / control in car heater / blowers but I couldn't see how that would seal very well. A rubber ball makes more sense.

[Chris B]

And I had similar symptoms with this type of 3 way valve (CH coming on in summer when HW only called for) - Taking the valve apart after it had been changed (as you do - just to find out how it worked) showed that 25 years of use had worn away a significant groove on the rubber ball and permitted leakage. (I also suspect that the rubber was somewhat less flexible than when new which wasn't helping)

[T i m]

Ok ...

Of course! [1]

Hmm, I thought the ball was free to rotate and so *should* spread the wear around (like a very early version of SSD wear leveling <g>)

And when you consider that it is in cold water one minute then very hot water the next ... and for many years ... hardly surprising? ;-)

I have been up to my neck since I posted and it's been a bit colder so she may well have the CH on in any case. I hope to try the manual override (turn the spindle with my fingers) with it in the HW position from cold and see what I can feel. Luckily it's all *very* easy to get at.

Cheers, T i m