Strange unvented DHW problem

[Roger Mills]

The hot water system at my holiday flat has started playing up.

The overall setup comprises an S-Plan system with gas boiler and unvented primary and secondary circuits.

The DHW part is based on a Santon Premier TP120B (120 litre unvented indirect) cylinder and the other components are broadly as per the relevant Wiki -

The cylinder was installed by the previous owner in about 1999.

When I arrived at the flat yesterday, after a period of absence, I turned on the cold water and set the heating/water programmer for DHW as usual, and the boiler fired up as expected.

However, the boiler was still running two hours later - far longer than it should take to heat a tank of water. I then found that water was running into the tundish from the T&P valve - so I was using gas to heat water and simply pouring it down the drain!

The T&P valve is only supposed to open if the temperature exceeds 90-odd degrees, or the pressure exceeds 6 bar - but I'm pretty sure that neither of those was the case. The boiler was cycling on its stat, which was set far below 90 degrees - so the boiler couldn't heat the cylinder to above its own temperature. If I turned a hot tap on, the water coming out wasn't excessively hot - and there was no steam released into the tundish, just hot water. I don't have any means here of measuring the HW pressure, and the only gauge is on the primary circuit - which was reading about 1.2bar cold and 1.4 hot. I did measure the air pressure in the DHW's expansion vessel, and that was about 2.5 bar - so I don't see how the water could have been at a higher pressure.

This morning, I tried heating the water with the immersion heater rather than the boiler. That took a long time because it's only a 1.5kW element

- provided for backup purposes - but that seemed to be ok without discharging through the T&P valve.

Water being discharged usually indicates a failure of one of the control systems - thermostats, etc. - resulting in over-temperature/pressure. Since that doesn't seem to be the case, another possibility is that the T&P valve is faulty. I did open the T&P valve manually quite a few times to try to flush out any crud which may be caught in it, but that made no difference. If it *is* faulty, I would expect it to leak when using the immersion heater - which it didn't. I'll try that again tomorrow in case it was a fluke.

Meanwhile, if anyone has any bright ideas about what to do in order to get an accurate diagnosis, please let me have them!

[John Rumm]

Was that with the water inlet shut off, and the pressure released from the taps so that you were measuring just its charge pressure?

If not, it could be it has actually pressurised and you have lost your expansion room. That could give over pressure flow to the tundish, which would then allow more cold in, cooling the cylinder. As it is reheated the cycle would repeat...

What temp is the immersion set to? is it lower than when heated by the boiler?

[harryagain]

First check that the water has not over heated and tripped off the "temperature" part of the "T&P" valve. Some have to be manually reset. If may be there is a faulty thermostat.

The pressure reducing valve may be stuck (in the open position), it should close when the hot water tap is closed. If it is stuck open, the pressure relief part of the T&P valve will lift discharging water to waste (This is the likely one)

If so, you will have to dismantle the PRV valve and free off . You might try fiddling with the pressure adjustment (if there is one) to get it moving. If they are not regularly "excercised" (By being used) they can stick. Or you might end up buying a new one.

[Roger Mills]

The 2.5 bar was measured with the water hot and under pressure.

Later, I shut off the cold water supply and opened the hot taps to release the pressure. The air pressure in the expansion vessel was then only about 1.5 bar. The charge pressure is supposed to be either 3.5 bar (same as the PRV is factory set to) or to mains pressure if that is less. In any event, 1.5 seems low. I tried to top it up, only to find that the foot pump which I carry in the car is shot! [I have a compressor at home, and rarely use the footpump]. I suspect that I need to make a trip to Screwfix to get an additional mains pressure gauge (I've got one back at home!) and to Halfrauds for a new footpump.

It sounds as if I've probably got less expansion room than I should have due to the low charge pressure but if I'd lost it all - allowing the pressure to rise to 6 bar to trigger the T&P valve, wouldn't I be seeing more than 2.5 bar on the air side?

I don't know in absolute terms. The cylinder has two identical stats - one for the immersion heater and one for the zone valve - both set to the same setting. The immersion heater stat is lower down the cylinder so, if anything, that should result in a higher overall temperature. Both heating systems produce comfortably hot water - probably around 55 degrees.

I ran the immersion heater again this morning, for a shower, without any problem. Admittedly I didn't leave it on long enough to its stat to turn off, but it was on for a hour or so, and the water was hot enough.

By contrast, I tried running the boiler again last night - when the water in the cylinder was still fairly warm - and water *immediately* started dribbling into the tundish. This almost suggests a cross-connection between primary and secondary (e.g. corroded heat exchanger) but if that were the case, surely the pressure in the primary would be more than 1.4 bar?!

Odd!

[Roger Mills]

I'm as sure as I can be that the water is not getting too hot. What comes out of the hot tap feels 'normal' and there's no steam in the water being discharged into the tundish.

This has to be a possibility, and I'm just off to Screwfix to get a pressure gauge, which I can attach to the washing machine tap (which is fed from the outlet of the PRV). I don't (yet) know what the mains pressure is - but I doubt whether it's enough to lift the T&P valve on its own. [I can check it at an outside tap once I get the pressure gauge].

I sincerely hope that this is not the case, because the PRV is very difficult to get at. It's behind the sink unit - with just a small hole in the carcase to access the knob. [No, *I* didn't install it!]

[Onetap]

DEfective T&P Relief valve, IMHO. I've come across the exact same thing, the T&PRV starts discharging at a lower temperature than it should.

If the discharged water isn't 90 degC (or whatever), the valve is bucked. Buy a new one; don't plug it, something may have caused it to start discharging.

The valve manufacturers swore blind they'd never heard of such a thing, lying bastards.

You should have a G3 ticket to work on these, but you're a big boy responsible for yourself. I used to have one, but haven't renewed it.

It is worth mentioning that such holiday flat water systems can harbour legionella and you should take care to pasteurise the contents of the cylinder, pipes and fittings before using it (i.e., just get them very hot).

There is an urban myth that you can't get legionella from domestic systems.

The reality is that you can and people do. Infections from domestic systems aren't covered by the relevant H&S legislation.

There is an increase in infections that corresponds to the end of the summer holidays.

[Roger Mills]

According to the Wiki, these things use non-standard threads to make "plugging it" very difficult if not impossible.

Indeed!

How hot is "very" hot?

[John Rumm]

Depends on how much volume that 1.5 bar is taking - you may find that even squeezing it to "nothing" - the reduction in volume is not significant enough compared to the non squeezable spaces in the valve and pipework to actually raise the pressure that much.

Is there not a blending valve on the outlet anyway?

Say there was a hole in the indirect coil, then you would expect cylinder pressure water to force its way back into the primary boiler circuit (i.e. the opposite of what would normally happen on a vented cylinder). So the resting pressure on the CH primary would rise to the PRV set pressure. That does not sound like what is happening here...

It does sound a tad "odd", although I think I would be inclined to get the charge pressure on the cylinder expansion vessel up first and try again before trying to do too much more investigation.

[John Rumm]

Since when has something being difficult stopped a fool on a mission? ;-)

All water cylinders are included in G3 of the building regs now...

(although I expect there is still specific certification for training on unvented systems - not sure if they have changed the name of it though since calling it G3 no longer makes sense)

There is a time / temperature trade off - so if you get it over 62 deg C, and then hold it for 30 mins, the that should be adequate, however you only need 15 secs at 72 deg C - basically you can keep going up and shorter.

Some boilers have an option of an "anti legionella" cycle, where they heat the cylinder to a higher than normal temperature once a week. On mine it ups the primary flow temperature toward the end of the charge cycle, and takes the cylinder up to 70 rather than its normal 60 (measured at the sensor pocket - which is a quarter of the way up the cylinder - so probably enough to mean the whole cylinder is over 62)

[Roger Mills]

Not as far as I'm aware - unless it's hidden behind a cabinet somewhere. I've always assumed that the HW outlet is connected straight to the taps.

Yes indeed - as per my comment about expecting to see more than 1.4 bar.

Yes, I'm working on it! I've bought a new foot pump but, because the EV is quite high up, I can only operate it by hand - and I'm really struggling to get the charge pressure above about 2.2 bar. I've now measured the mains pressure at an outside tap - about 3.8 bar - and downstream of the PRV - about 3.2 bar, so I guess that the charge pressure should be about 3.2 I ran the boiler to heat the water this afternoon after increasing the charge pressure a bit - and it ran for over half an hour before the T&P valve started dribbling - which seems to be a bit of an improvement. I need to find a way of increasing the charge pressure a bit more - maybe a 12v compressor?

[Roger Mills]

I've been known to by-pass safety devices on occasion - e.g. guards on circular saws - but this is one I *wouldn't* muck about with!

According to the Santon literature, the stat is adjustable between 43 and 70 (and is currently set somewhere in the middle) - probably around

55-60. According to my IR thermometer, the boiler flow temperature is currently in the low 60's - so I'll have to tweek that up if I want to raise the DHW temperature. But I need to fix this discharge problem first!

[dennis]

You have de-pressurised the hot water before you try and inflate the bladder? If you don't you wont have the expansion tank full of air when you finish.

You should be able to get it to 3.5 bar with a cycle pump.

[dennis]

Isn't there enough chlorine left in the water to keep it clean anyway? I can't find a reference as to how much chlorine is in there or if its enough.

[Roger Mills]

Yes, I've done that!

That's a possibility.

[Onetap]

60+ will help do it. Just be aware of the risk of leaving a cylinder of tepid water brewing away for several weeks/months.

Having a shower, and inhaling the droplets, on arrival and before the cylinder has had time to fully heat up is likely to book you a bed at a hospital.

My system has a pateurisation program that heats the hot water cylinder to 70 or 75 degC once a week.

ISTR the PRV has a lower pressure setting than the T&PRV, so it's not likely to be pressure related.

When the T&PRV operates correctly, due to over-heating, it discharges a few litres of steaming hot water at full bore, then shuts as the incoming water drops the temperature. The cycle repeats every few minutes.

[Onetap]

No, the chlorine (or whatever they use in the mains) won't kill all the legionella. The concentration required would make the water toxic. It does restrict its activity, though.

Most of the chlorine is released when the water is heated.

[Roger Mills]

Indeed - as long as they're working as intended. The PRV is factory set at 3.6 bar, and the T&P valve is supposed to operate at 6 bar (and/or

90+ degrees).

There's no steam coming out of mine, just a steady trickle of hot - but nowhere near boiling - water for a couple of hours of so, with the water meter turning very slowly as it feeds in the replacement cold water. If, while this is happening, I open the T&P valve manually, I get a very high flow for a second or two, and it then reverts to a trickle when I let go of it.

It eventually stops - which which time the water in the cylinder has cooled considerably, and is only just hot enough to wash with.

If I heat it up again using the boiler, it starts all over again.

If I heat the water with the immersion heater to a comparable temperature, the valve doesn't open.

If I run the boiler to heat the radiators but not the HW, the T&P valve doesn't open - which of course it shouldn't, but I thought I'd better check in view of its odd behaviour.

I'm tending to think that the T&P valve is duff. If it were triggering on an over-pressure, then surely there could be no cold water inflow from the 3.5 bar mains? [Even if the PRV had failed open, the static mains pressure is only about 3.8 bar]

Also, there's no way that the temperature can be above 90 degrees. There's no evidence that it's over hot, and the boiler flow temperature (as measured with my IR thermometer when running the radiators this morning) is only about 72 degrees.

I suspect that it's erroneously triggering on temperature - but I can't figure out why it opens when I heat the water using the boiler, but not when I use the immersion heater.

[dennis]

Its sealed and under pressure so where would it be released to?

[Onetap]

I'd missed that detail and am unable to explain it. The only thing I can su ggest is that the heating coil is usually in the lower part of the cylinder (to minimize the amount of legionella that survives in the coolest base of the cylinder beneath any limescale debris). It will heat more of the water than an immersion heater mounted higher up, so the cylinder pressure will be higher due to the extra expansion.

I think so too, maybe assisted by the higher pressure when heated by the bo iler.

The behaviour is otherwise very similar to the defective one I saw; I'd rep lace the T&PRV on that basis and not try to solve the mystery. It may be a leak into the cylinder from the heating, but I doubt it.

When it works properly, the wax capsule thermostatic actuator goes from sol id to liquid at the set temperature and the valve goes suddenly from closed to full-bore open and vice versa.

The defective one started to trickle once the water was tepid, 45 or 55 deg C or so.

[Roger Mills]

UPDATE: This morning, it discharged when I used the immersion heater!

Neither the temperature (about 55-60) nor the pressure (no more than 3.5 bar, even when hot) are high enough to open the valve if it's working properly - so it's almost certainly duff.

Sounds like what mine is doing.

Any gottchas to look out for when replacing it? The main problem in my case seems to be that there's a plastic cover over the primary circuit connections which will prevent the valve from being unscrewed. See

The cover is loose but was fitted *before* the connections were made, and cannot be removed without a major un-plumbing exercise. Have you come across one like this? Any suggestions? I'm wondering whether I can use a Fein Multimate to cut the plastic cover into two parts across the nearest pipe hole, and then somehow stick it together again. Either that, or somehow enlarge the pipe holes in situ enough to lift the cover far enough to be able to rotate the valve under it.

All constructive comments appreciated.