Bernard. I don’t think you have the facts in this thread.
Turn off power for two hours?
- Thread starter Vicario
- Start date
Sponsored Links
I got the promised call back from BG. "We can confirm that we can fit a smart gas meter if space permits. We will also need access to your electricity meter in order to fit the console". But the electricity meter isn't theirs and I thought they couldn't touch it. Does she mean CU, not meter? It was a voicemail so I still don't know why they want to fiddle with my house electrics. ( Sighs deeply)...I'll try to get clarification tomorrow. I almost feel like saying that I'm going to get my gas from my electricity supplier thereby becoming a dual fuel customer.
They can
They do, all of them. It's a mandatory part of the spec. And it's there primarily for demand management - i.e. disconnecting users on a more granular manner when we run out of generator capacity.
I believe there are also functions defined to allow load switching, with the common example being to run the washing machine/tumble drier in the early hours of the morning (as if that makes fire safety sense
)In theory, the meter could report the current price to connected devices and the user set a limit on what they want to pay and the washing gets done when the lecky is cheap enough. However, AIUI the spec doesn't include this, nor what the upcoming rates are going to be. Yes, the "smart" meter spec missed the opportunity for several useful & money saving features
But what we are left with is variable rates, short term pricing to encourage load reduction, and the important remote disconnection - oh yes, and remote switching between credit & pre-pay mode.
Last edited:
I didn't think there was a secondary output from the SM, to power high load devices, which is a shame.
Though, most people won't want to add wiring to their house.
They say they aren't going to use the remote disconnection function, on grounds of safety. i.e there maybe a person with medial needs, or a danger when power is returned.
I guess if our power station situation gets really bad, they could turn off people drawing more than 60A.
And turn it back on after 5 minutes as a warning!
Though, most people won't want to add wiring to their house.
They say they aren't going to use the remote disconnection function, on grounds of safety. i.e there maybe a person with medial needs, or a danger when power is returned.
I guess if our power station situation gets really bad, they could turn off people drawing more than 60A.
And turn it back on after 5 minutes as a warning!
Sponsored Links
Indeed.
I think that many people are probably seriously over-estimating what is practical, either now or for at least a good few decades to come. Remotely turning off substantial 'blocks' of consumers (with preservation of supplies for those flagged as 'high priority' users) is one thing. However, to do anything on a 'per installation' basis, across tens of millions of installations, would be a very-far-from-trivial IT and communications exercise - which I suspect is way beyond the capabilities of the system currently being put together, and probably any other system we are likely to see any time soon.
As times progress, the best way to deal with an excessive demand (in relation to supply) situation would probably be to turn off all EV charging
Kind Regards, John
I guess people need their own solar panels and batteries for the bulk of their power
Quite apart from the issue of the cost of the PV installation (initial and periodic replacement) do we get enough sun to keep an EV charged, let alone more?
Kind Regards, John
Done by radio - there's messages defined in the standards for the SM to signal external relays to turn on/off.
Well they say a lot of things when introducing stuff. Doesn't stop mission creep once the wedge is into the gap.
There's even mention of it (in a roundabout way) in relation to the blackout a couple of weeks ago in the south east. There was specific mention of more targeted disconnections - leaving critical (eg hospitals) or infrastructure (eg trains) or vulnerable users connected. That would need much more fine grained control, the sort of control offered by "smart" meters
In the other thread (IIRC) I did suggest that this might be a limitation in large loss of generation events. I don't know anything about the system being built, but in theory there's no reason it shouldn't be able to disconnect a low of users very quickly - just think about how many SMS messages the mobile networks will be handling at any time. Whether they've actually built (or are building) a capable system is another matter
From a talk I went to a short while ago, it's clear that the whole project is an omnishables of the highest order. There are massive compatibility issues compounded by the fact that they not only don't have interoperability sessions - they have specifically prohibited interested parties from talking to each other to iron out problems. On that basis, it might be a while before the system has anything more than the basics working.
Which is actually one of the proposed mechanisms - the "smart" network will signal to EVs when there's capacity for charging. In theory, you'll tell your EV when you want it ready for, plug it in, and it will arrange to be charged by then. Sounds great in theoryAs times progress, the best way to deal with an excessive demand (in relation to supply) situation would probably be to turn off all EV charging
Well that could be an interesting question.
In summer, 4kW (limit for small install before the rules change significantly) of PV over a long day might well do it - and don't forget that power usage will be lower (non-use of lights and heating), as long as A/C isn't used. But for most people, they aren't at home when the sun's out - they are at work - so that means a not inconsiderable level of storage needed.
Winter is another matter. Not only are days very short, and sunlight in limited supply, but power usage in the EV will be higher (lights and heating) and other demands on the power (domestic lighting and heating) will also be a lot higher. I'm thinking that the answer to the question would be "spell chance"
As I said, it ought to be 'fairly easy' to disconnect lots of users in large pre-defined blocks (defined to avoid 'priority' users) - there's all sorts of ways that can be done. However, I was responding to the suggestion (I suspect somewhat tongue-in-cheek) that the system might disconnect just users with a current demand >60A. That would require two-way communication separately with each of the countless millions of installations - which, as I said, I suspect might be beyond the capabilities (within a reasonable time frame) of the 'first generation system' (if not also the 'tenth generation' one
)There presumably speaks a man who has not sent (or was meant to receive) a text message just after midnight on New Year's Eve (well, New Year's Day!) or at other times of 'exceptional demand' - it can take at least an hour or three (and sometimes longer) - hardly adequate if one is trying to act in response to the sudden and unexpected loss of generation capacity!
It is - and, as above, I don't think existing cellular networks would be up to it.Whether they've actually built (or are building) a capable system is another matter
Although it's not something I like to think about, I'm gradually getting older, and I do doubt that I'll ever see anything beyond 'the basics' (if even them!).
Hmmm. I suspect that if EVs ever do/did become very widespread, the times of day with least available capacity will also be the times of day when many/most people want/need to charge their EVs!Which is actually one of the proposed mechanisms - the "smart" network will signal to EVs when there's capacity for charging. In theory, you'll tell your EV when you want it ready for, plug it in, and it will arrange to be charged by then. Sounds great in theory
Quite so. As you say, Winter is the season when the total demands on the network will be at the greatest, and it's also the season when solar energy is most thin on the ground!Well that could be an interesting question. ... In summer, 4kW (limit for small install before the rules change significantly) of PV over a long day might well do it .... Winter is another matter. Not only are days very short, and sunlight in limited supply, but power usage in the EV will be higher (lights and heating) and other demands on the power (domestic lighting and heating) will also be a lot higher. I'm thinking that the answer to the question would be "spell chance"
Kind Regards, John
Sending commands to smart meters from a control room is a very different process to that of transfering a message from one mobile to another mobile.
Mobile A sends a text mesage to Mobile B
---------------------------------------------
Mobile A establishes a link to a mast
Mobile A sends the message to that mast,
the message is then passed from that mast onto the inbox queue at Mobile A's service provider
the message has then to move up the queue
when at top of queue the message is examined to identify the service provider for Mobile B
the message is then passed to the service provider for Mobile B
The service provider for Mobile B then has to locate which mast(s) Mobile B last linked to
The service provider for Mobile B then has to confirm that Mobile B is still reachable via that mast.
If Mobile B is no longer accesible from that mast the service provider has to search for Mobile B
( mostly this search is performed by waiting for Mobile B to contact a different mast with an "I am here" check in. )
Only then can the message be sent to Mobile B
Control Room sends a message to a meter.
-----------------------------------------------
The link from control room systems to the service provider for the meter is a permanent data link.
The meter is not mobile so the mast that serves it is known and does not have to be searched for.
Also it is very likely that commands to meters would be sent as "group calls". Normally a mobile responds to its own unique number, with group calls several meters receive and react o a single message transmission.
A meter would have a unique number for transfering routine data to and from the energy suppliers account's department.
When a disconnect has to be sent from the Network control room then groups of meters could be controlled by a single group call
987654321 is the meter's unique number, the leading 9 tells the meter to require the full 9 digit number to respond and to treat it as a standard SMS transaction
087654XXX A group call to 1000 meters, the leading 0 tells the meters to ignore the last three digits and accept a command without acknowledgement, Confirmation of discannect ( or other command ) would be by status requests sent to the meters sequentially using the full number ( standard SMS messaging ).l
Very True. However, for a start, on most of the occasions (e.g. NYE) when I have personally experienced very long SMS delays (assumed to be due to excessive demand), it has been a call from a mobile to a landline, so only the first 3 or 4 of your list of 10 steps will have been applicable. Furthermore, on most (but not all) occasions when it's happened, the person who sent the SMS reported that the message had been successfully 'sent' more-or-less 'immediately', suggesting that even the first three steps happened at roughly normal speed. I therefore suspect that the issue is primarily that of the speed at which the system can handle a large number of messages. In any event ....
Exactly. As I wrote ...
... and one of the "all sorts of ways" I had in mind was to use a 'master key' approach such as you describe (and one can also have 'sub-masters' as required). HOWEVER, as I went on to say ....
To (hypothetically) identify all those installations with a current usage >60A would seemingly require messages to be received from each and every one of the meters [the outgoing 'request' for this information could be sent by 'group' message(s)], which I still think would take an appreciable amount of time. If that were the case, then it might well be totally 'unfit for purpose', given that most domestic installations probably only ever get near to (let alone over) 60A when an electric shower is in use, and they are generally only used for a few minutes - the data could therefore be very incorrect before it had all been collected!.... However, I was responding to the suggestion (I suspect somewhat tongue-in-cheek) that the system might disconnect just users with a current demand >60A. That would require two-way communication separately with each of the countless millions of installations - which, as I said, I suspect might be beyond the capabilities (within a reasonable time frame) of the 'first generation system' (if not also the 'tenth generation' one
There is, of course, a way in which it could (hypothetically) be done very easily/quickly, with only a very small number of messages having to be passed (in either direction) ... simply by sending 'group' message(s) to 'sufficiently smart meters' which said ...
"IF current total consumption is >60A, THEN disconnect the consumer's installation"
However, simple though such a command is, I'm far from convinced (in fact, doubt) that current 'smart' meters are sufficiently 'smart' to undertake such processing.
Kind Regards, John
I'm not sure group communications is provided for in the comms spec. To start with, it's quite difficult to do securely - and remember that the messages are cryptographically encrypted and protected. It would be possible to pre-define groups and pre-load a group shared key to all meters in the group - but I don't recall seeing anything like that when I skimmed through the spec a while ago.
Absent such a shared key, each message would need to be individually encrypted for each meter, and that means a message per meter.
In any case, that is probably easier to manage than groups anyway. Vulnerable users and protected services could be handled by one or two flags in the meter database - with the load shedding algorithm basically being along the lines of "pick a load of meters, remove protected users from the list, shove disconnect messages for those left into the message queue".
Should the requirement for disconnections become common, then it would also be possible to do "nice" things like selecting meters that have gone the longest since a disconnect command and thus making sure that the pain is spread round equally.
Having said that, the fact that such things are "fairly obvious" probably means that they haven't been/won't be implemented
This is, after all, a government project contracted out to one of the big providers famous for their repeated failure to provide a working/reliable system to cost and time I don't recall seeing anything relevant to that when I skimmed through the spec.
But you highlight another function that the meters could be used for - maximum demand control. In many countries, the standing charge you pay is determined by the rating of a circuit breaker (accessible to the customer so it can be reset). This can be as small as 6A
So there would be the scope for re-introducing the concept of paying a standing charge based on maximum demand as used to happen long ago in this country. But instead of needing a breaker to enforce it (and an engineer visit to change it), it could be software set remotely. But I also don't recall seeing anything like that in the spec - but then it was a while ago and I only skimmed it.
Last edited:
I made it up that's why!
Same here. Whilst, in terms of present-day technology, the processing required would be trivial, as far as I can make out there is no intention that any actual true local 'processing' in that sense is part of the required capability of a 'smart' meter. In fact, they appear to be very 'dumb' - primarily data storage and comms, and not a lot more - but maybe I have misunderstood!
It would not be the first time, here and elsewhere, that I had commented on the pretty unique situation we have with this industry - there is not a lot of precedent for an industry bending over backwards in attempts to restrict the amount of their products/services that their customers buy, and penalising those who attempt to buy 'too much'! In almost any other situation, if it looked as if demand was going to outstrip supply, all the efforts and investment would be directed towards increasing supply (and profits), not reducing demand by 'rationing'!
Kind Regards, John
DIYnot Local
Staff member
If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.
Select the supplier or trade you require, enter your location to begin your search.
Are you a trade or supplier? You can create your listing free at DIYnot Local
Sponsored Links
