Viessmann boiler failure

[Mike Watterson]

Even 20 year old ones had thermal cut out as well as the thermostat.
In theory the expansion pipe for HW is safest to empty outside and NOT into any header tank, then you see the hot water outflowing.

[Mark Hennessy]

I dare-say the expansion of the water in the header tank might've caused some output from its overflow, but that's beside the point. That family had only moved into the rented property that day - in amongst the stress and hassle of unpacking and settling down their children, they can be forgiven for not spotting water flowing from the overflow - assuming any did, and assuming the overflow had been placed in a sensibly conspicuous place.

But the problem occurred because the header tank in the loft - the tank containing all the cold water that feeds into the hot water cylinder - was made of plastic. As most are. The immersion heater stuck on, which led to the water in the tank in the loft heating up, and at some point, that caused the plastic tank to fail catastrophically, releasing all those gallons of stored water in an instant.

Perhaps I am using the term "header tank" incorrectly? I mean the cold water storage tank that provides water (and "head") to the hot water cylinder. Some references suggest that the term "header tank" might refer to the small tank that non-sealed central heating systems used. Either way, it's completely clear from the context from just a perfunctory skim of the news story.

The trouble with immersion heaters is that they can have long service life, especially where the water is "kind" to them, so there must be plenty of 20+ year old ones out there. Certainly, the one that failed was a type that lacked a separate cut-out. I'm not sure what type I have in my DHW cylinder - it must be around 20 years old. We have gas, so it's only for backup. I was glad of it when the boiler failed 5 years ago - it meant that I could take my time deciding how to go about replacing it, rather than just making a "panic buy". During that time, I was monitoring the current it took for safety and energy assessment reasons (I concluded it was cheaper to leave it on 24/7, once I'd satisfied myself it was reliable).

That poor family though. And others, it would appear. The story certainly touched me at the time, with a 2 year old, and another baby on the way.

[Mike Watterson]

There should be two separate overflow pipes: One for cold tank in case the ballcock sticks. One from the top of the HW cylinder (which is where the HW is taken off). The main cold tank should be connected at the bottom end of the hot tank. So the ONLY way the cold tank can fill with hot water is if the HW overflow / expansion pipe empties into it instead of outside, or if there is no expansion pipe at the top of the HW cylinder.
Heating systems can be different and can be pressurised or have a header tank and an expansion pipe from boiler and run at atmospheric pressure.

A terrible tragedy and it's not impossible that a plumber / builder or surveyor (when they bought it) is guilty of negligence.

[boater sam]

Domestic hot water heating system with cold water storage tank and non pressurised cylinder heated by electric immersion heater and other heat source.

The cylinder fills from the cold storage into the base of the cylinder.
The hot water is taken from the top of the cylinder along with an expansion pipe which discharges into the cold storage tank BUT ABOVE  THE MAXIMUM WATER LEVEL IN THE TANK EVEN UNDER OVERFILLING CONDITION.
So if the cylinder boils for whatever reason, steam will vent over the cold tank water surface not into the water. In this condition the cold water tank continues to keep the cylinder full of water at all times. The water level will be kept topped up by the mains cold feed into the tank. In this condition considerable steam at zero pressure will be created but the system is fail safe even if the cold feed is interrupted. In that event the immersion element will eventually melt and fail when it is no longer immersed, disconnecting the power.
The only way that the storage tank can be heated to any degree is if the expansion pipe end dips into the water and a circulation is achieved, the steam being condensed so that there is no cold water makeup.
In this condition the cold storage will eventually boil, weakening a plastic tank considerably.

So if properly installed, the system is fail safe.
If the hot water cylinder is a sealed pressure cylinder, a whole new set of dangers arise necessitating precision thermal and pressure controls to maintain a safe operating condition.

[Mark Hennessy]

Michael, the house was a council house, as you should have spotted from just a perfunctory skim of the news story.

As Sam says, the take-off from the top of the cylinder is taken up to the cold water tank, well above the water level. The cold water tank obviously has an overflow to the outside world, and by design this has to be in a conspicuous place so that it's noticed - this is for spotting a failed ballcock, as you say. But these don't always remain conspicuous. For example, in a house belonging to a relative, an extension means that water discharges from the overflow pipe and rolls across the new flat roof and into the gutters. You ain't gonna see that happening from the garden!

In this instance, some discharge from the overflow might well have occurred - like I said earlier - but as this happened at 2.30AM, it's highly likely that during the build-up to this, it simply wasn't spotted. It was late November, so dark at around 4PM, and the family had only just moved in - literally that very same day - so weren't familiar with the property and wouldn't have had a chance to work out all this sort of stuff. Indeed, beyond the likes of us, would a "normal" person make a point of locating overflow pipes on day 1 in a new house, whilst unpacking and trying to get their 2 kids bedded down in their new room? Indeed, the father was asleep but still dressed at 2.30AM when it happened, which suggests that they'd been so busy that they'd not had a chance to set up a bed in their own room yet.

I'm not sure the overflow pipe would have needed to have been dipped in the water in the tank, but could certainly see how that would greatly accelerate the process. With a faulty thermostat, the water in the DHW cylinder would have been bubbling away really quite violently, after all. And if the pipe runs were short, there would have been a reasonable amount of conduction via the water (I'm surprised at how warm my pipes are, even when they appear in the loft).

I can only presume that the previous tenants only ran the immersion heater for short periods of time.

There is one puzzling aspect:

John Williams, chief housing officer for Taunton Deane Borough Council, said that safety inspectors were not expected to examine attics of houses - where the immersion tank was - "because they don't have ladders".

It's not impossible, but it's really unusual for an "immersion tank" to be in the loft, assuming he meant "DHW cylinder". I'm assuming this is someone being unclear in their use of language, and that the cylinder was in the usual airing cupboard. Assuming that's the case, and the only thing in the attic was the cold water storage tank, then an inspection of that would have revealed nothing unless the immersion heater had been running for a decent amount of time when the inspection was carried out. Well, perhaps it might've revealed the overflow pipe being too close to (or in) the water? So that sounds like a poor excuse to me, sadly.

Regarding the question of negligence, inspecting an immersion heater sounds like a difficult task. How many hours would you have to hang around on site for? How would you ensure it's going on and off? One of those "clever" clamp meters that can estimate current in a cable (rather than a single core) might do the job, but those are rare, and the only one I've ever seen is no longer made. So you'd have to disconnect the heater and put an ammeter in series - or run it with the cover removed if there was access to the post-switch heater circuit. Or perhaps leave a logging thermometer? Really, I think this relies on a tenant reporting strange sounds or scalding water, but that's a whole other minefield!

From what I've read, some councils and housing associations decided to replace certain types of immersion heaters after this tragedy. Preventative maintenance.

Anyway, what happened happened, and it's happened on more than one occasion. Greater awareness of the risk can only be a good thing, and I'm sure the families concerned would wish that no-one else has to experience what they have. Ultimately, not all installations are perfect, and not all installations remain perfect (e.g. pipework can move when people throw things in their lofts). I remember taking great care in the design of my system to ensure that the pipework couldn't "slump" in future, but then I spent many weeks installing my new water tank - not a luxury available to the average contractor doing work for a local council who would have screwed the price right down.

[ppppenguin]

I've seen 3 or 4 installations where the DHW cylinder was in the loft. Unusual but nothing wrong with it. The cold water cistern has to be raised on a platform to get it above the cylinder. This method gains the space that would have been used for an airing cupboard etc but of course you don't get a nice warm airing cupboard. However well** you insulate the cylinder and its pipework the airing cupboard will always be warm.

**I suppose you could make a superhuman effort and insulate it supremely well. But perhaps not in this lifetime.

[Mark Hennessy]

The only one I saw was in a lean-to kitchen. To fit, it was laid on its side! The cold water storage tank was in the main loft, which was a story higher, and it had been put on a raised platform to give a good shower without the need for a pump in the upstairs bathroom.

The previous owner was a plumber and did all the work himself, so I guess it was OK

But a maintenance nightmare, as there was no loft hatch in the kitchen!

[ppppenguin]

Back to my boiler. I've just added another litre of Sentinel X100. When I let off the pressure (via the Magnabooster) I got a few fragments of black iron oxide. So either the system has been making some or there were some hidden residues that the DS40 failed to clear out. There was also a bit of a surprise. the first bit of water was a dirty grey colour with the fragments of black magnetite, I sort of expected this. Then it ran clear, then it went reddish brown, like rusty water though the colour may not have been quite right. To get red iron oxide you need oxygen in the system, something that I really doubt is present in a pressurised system. If it really is red iron oxide (haematite) then radiators can rust through very quickly indeed. Then it ran clear again, then a bit more reddish brown, then clear.

[boater sam]

As I have  pointed out in the past, flushing a system without removing and flushing each radiator both ways is not always going to achieve a clean system. Even with a clever power flush pump, its a waste of time for little result and the sediment will still be there. Acid flush such as DS1 will not dissolve all sediments found in central heating systems. Its fine for cleaning a new system when it is first completed but an old system requires some physical effort.
A complete new system installed with plastic pipe and fittings, new clean boiler and good quality clean radiators will only require a flush to remove  any "bits" and oils, as there will be no flux or sealing compounds.

[ppppenguin]

Each rad was flushed both ways using mains pressure. Rather than removing each rad I turning off all the others and let the entire mains flow go through each rad in turn. i rigged a system of valves so that I could flush both ways.