Grenfell Tower – an engineer’s take

This is from the inimitable Microdave:

You will (I’m sure) remember me talking about working on various electrical problems at my friends chicken farms – well that got me thinking the other day when I had a look at the Grenfell Tower Residents Action Blog.

I noted the reports of “Power Surges” in 2013, and it reminded me of a specific fault at one of the two farms. This manifested itself when several of the emergency vents didn’t work properly during power cuts. The chickens in large commercial sheds aren’t as hardy as the ones you might see in small paddocks and the like – they keel over very quickly if the temperature gets too high, so various emergency measures are provided to try and keep some natural ventilation going if the power goes off, and the genny doesn’t kick in (it happens!).

I discovered the motors which operate these vents were showing a short circuit on the input, in one direction (but not in the other). I found that a simple diode inserted got them going again, which saved several replacements at £600 each!

Further investigation found (to my horror) that said emergency genset was putting out in the order of 270 volts instead of the usual 235-240. How long this state of affairs had persisted nobody knows, but it could have been getting worse gradually since it was installed some 4 years previously. I was able to make adjustments to restore normal voltage, but it was replaced by a newer unit not long afterwards.

Now to the reason for this message – we have been told that the fire started in a fridge/freezer, and then spread, with devastating consequences. Reading today that the model in question – a Hotpoint FF175BP fridge freezer – was made between 2006 and 2009, had me wondering if the same excessive voltage scenario could have been the cause. Due to its age, this model would have to have been present in the flat when these “Surges” occurred.

The blog didn’t go into much detail about the nature of these, but reported that monitoring equipment was installed in various flats, and that many appliances had suffered catastrophic failures. It also reported that the problem was traced to a burnt out cable. Now at risk of getting technical, I am going to postulate that the cable damage had resulted in a “floating neutral” situation.

All mains generation and distribution is done with 3 phase equipment – it’s far more efficient than single phase. Generators are built with 3 sets of windings spaced at 120 degree intervals, and all the main distribution wiring and transformers have 3 cables – effectively one for each “phase”. Only at local distribution level will this be separated into individual phases for residential and small commercial use.

In this case, the substation transformer has a “Centre Tap”, where one end of each winding is connected, and this is brought out in addition to the 3 phase wires. Such overhead 4 wire distribution is visible in rural areas (but buried in towns/cities) and the normal practice is for the individual phases to be fed to properties in turn, to (try) and balance the overall load. The centre tap (known as “neutral”) is also earthed at the source, and is what allows RCDs to protect you in the event of a damaged cable.

There is a mathematical relationship between the phase to phase, and phase to neutral voltages – in the UK you will normally see 415/240 volts talked about, in any situation where three & single phase supplies are involved.

This all works fine UNLESS the neutral ever breaks between the transformer and the connected equipment(s). Then any imbalance of the loads from one phase to the next will cause the neutral voltage to “float” above that of ground. This, in turn, will lead to individual phases going both below AND ABOVE the standard voltage. I had to do “Vector Diagrams” and calculations at skool 45+ years ago, but I would be struggling now.

It’s the higher voltages which do the damage – voltage swings from below 150 to above 300 are quite possible, and in extremes could approach the phase to phase voltage of 415! Clearly any electronic domestic appliances and lamps etc would simply “blow”, but motors are pretty robust, and can often survive in such situations.

In view of this I am going to postulate that the reason the fridge/freezer caught fire is not necessarily due to any inherent fault, but that it could very well have suffered some damage during the surges, which didn’t initially affect its operation – specifically, the motor starting capacitor. All domestic fridges have hermetically sealed motor/compressor units, and the refrigerant gasses are usually very similar to that used in fire extinguishers, so even if it “blew up” internally, it’s unlikely to start a fire.

BUT the capacitors are usually mounted outside the motor body, and all such devices have a maximum voltage rating – usually 10-20% over normal rated supply. The insulating layers inside can be strained by excess voltage, but not necessarily fail there and then. A fridge/freezer of that age would have a simple mechanical thermostat – no electronics, and the motor & capacitor would only be connected to the mains when called upon by the thermostat.

If this fridge was running when subjected to high voltage surges, the capacitor could have suffered, even though the actual motor windings escaped unscathed. Then some years later the capacitor finally says “enough” and ruptures – only now the fire is not contained, and is directly underneath the expanded polystyrene insulation, and plastic liner of the fridge… The plug-top fuse would be of no help, as this needs to be a fairly high amperage rating to cover the start-up surge of the motor.

I stress this is only my theory, and there will probably be the usual “knee-jerk” reaction, condemning perfectly serviceable appliances, but it’s worth bearing in mind. The GTRAB doesn’t appear to have any contact details, or I would be sending them this.

Just wondered what you think?

JH adding this – saw it on Twitter:

8 comments for “Grenfell Tower – an engineer’s take

  1. Ed P
    June 29, 2017 at 1:45 pm

    BS stands for British Standard, not bullshit, so EuroNorm 13501 must have been signed off in the UK after being issued in the EU.
    But allegedly the cladding was not compliant with this (lax) regulation, so it seems the contractors rather than the law are implicated.

    • Errol
      June 29, 2017 at 6:46 pm

      These would be the people who said that the sort the council had chosen was not sufficient? That it shouldn’ be used at those heights?

      No, while ‘British standard, the EU defines it. It tells us what we will accept, then we’re told to ram it into law.

      BTW, the EU also signed off those panels. Again, the EU is at fault but no one will accept it.

  2. Mudplugger
    June 29, 2017 at 4:11 pm

    Little attention seems to be paid to the root-cause of the fire, everyone’s being distracted by the ‘cladding issue’, which was merely a consequential effect.

    Following on the dodgy-fridge theory, one additional factor may be that, in many such dwellings, the residents choose to bring in propane appliances (cookers, heaters etc.), being preferred to electricity for cost and control reasons.

    If a propane-gas cooker was located next to the suspect fridge, with the rubber connecting-hose to its cylinder passing behind the fridge, then when the fridge had a major overheating event as described, that could easily have ignited the ‘live supply’ of propane behind it, creating a significant explosion.

    Of course, many other residents in the 120+ flats will also have had such propane appliances on site, with their associated gas-bottles, so there was plenty of internal accelerant available on each floor to help the blaze to spread so quickly and comprehensively.

    In this context, the external cladding hysteria may be considered secondary.

  3. June 29, 2017 at 4:47 pm

    Could be.

  4. Voice of Reason
    June 29, 2017 at 11:06 pm

    Given the low quality of the electrical motors coming out of China and installed in household appliances, this weakness would not surprise me.

  5. Hereward Unbowed.
    June 30, 2017 at 5:35 am

    I noted the reports of “Power Surges” in 2013

    Mike, could you possibly speculate on the causes of these “Power Surges”are they specific to West Kensington and pert to Grenfell or, are they common across the country?

    imho all of this is quite vexing:

    In this case, the substation transformer has a “Centre Tap”, where one end of each winding is connected, and this is brought out in addition to the 3 phase wires. Such overhead 4 wire distribution is visible in rural areas (but buried in towns/cities) and the normal practice is for the individual phases to be fed to properties in turn, to (try) and balance the overall load. The centre tap (known as “neutral”) is also earthed at the source, and is what allows RCDs to protect you in the event of a damaged cable.

    There is a mathematical relationship between the phase to phase, and phase to neutral voltages – in the UK you will normally see 415/240 volts talked about, in any situation where three & single phase supplies are involved.

    This all works fine UNLESS the neutral ever breaks between the transformer and the connected equipment(s). Then any imbalance of the loads from one phase to the next will cause the neutral voltage to “float” above that of ground. This, in turn, will lead to individual phases going both below AND ABOVE the standard voltage.

    btw, thank you for the run down, vexing but nevertheless an interesting and a cogent post.

    • microdave
      June 30, 2017 at 11:50 am

      “Mike, could you possibly speculate on the causes of these “Power Surges”are they specific to West Kensington and pert to Grenfell or, are they common across the country?”

      They could occur anywhere in the country – 3 phase distribution is universal. I would speculate that this could have been nothing more than a bolted connection becoming loose, or a large, crimped terminal failing. Most electrical conductors and terminals are made from copper, and being rather soft it can easily be crushed during installation. Good practice is to go round afterwards and re-check everything.

      The difference with the neutral failing, as distinct from one of the individual phases, is that the latter will only result in a loss of voltage, NOT the risk of an increase. In theory, in a perfectly "balanced" system, the centre tap would carry no current, but in practice there is always a degree of imbalance when single phase loads are involved. The exact location of the reported burnt out cable is not mentioned, so it's possible that only certain parts of the estate suffered directly, but sudden arcing under heavy loads will also result in very short duration "spikes" elsewhere on the system. These tend to damage electronic equipment more than conventional loads.

      As an aside, if anybody notices their lights suddenly getting appreciably dimmer or brighter than normal TURN OFF THE MAINS AS SOON AS POSSIBLE, and call the supply company.

      • Hereward Unbowed.
        July 1, 2017 at 2:10 pm

        Thank you for that, my physics only extends to ‘O’level GSE but the explanation certainly fits.

Comments are closed.