Meter replacement

[JohnW2]

I would suggest that a clue of when something occurs, could help considerably to track down a mysterious consumer.

In some situations, it might conceivable hyelp, but in cases like the OP mentioned ("1.5kw and 1.0kw heaters starting up periodically overnight") I really can't see how 'time information' would help, can you?

Indeed, if it were happening 'periodically' throughout the night (which one might expect if it were, say, thermostatically controlled), then the 'unexpected usage' would probably be spread all over the night (so could not be tied to any particular time window by a 'smart' meter), wouldn't it?

Kind Regards, John

 

[JohnW2]

I reckon you are missing a china watt meter.

I've got several of those, but I don't know what they have got to do with the present discussion,

In fact, if you went round all the potential loads in your house and fed them through such a monitor for a while, you'd be able to discover which one was responsible for 'unexpected electricity usage' without having to have a 'smart' meter, wouldn't you?

Kind Regards, John

 

[DiyNutJob]

I've got several of those, but I don't know what they have got to do with the present discussion,

In fact, if you went round all the potential loads in your house and fed them through such a monitor for a while, you'd be able to discover which one was responsible for 'unexpected electricity usage' without having to have a 'smart' meter, wouldn't you?

Kind Regards, John

My first approach would be to switch off stuff.

 

[Harry Bloomfield]

The same way you bypass pretty much every other piece of hardware. You'd obviously need to extract the firmware, reverse engineer it, update the code, recompile and flash. You could change the meter to return half of your actual usage for example, and no one will know, and any physical inspection won't reveal anything. Not that I recommend it - it is breaking the law, but digital hardware always introduces all sorts of ways you could tamper with it and leave minimum to no trace.

The bit you seem to ignore, is that any sort of tampering with a meter, will be immediately signalled to your energy provider.

 

[JohnW2]

The bit you seem to ignore, is that any sort of tampering with a meter, will be immediately signalled to your energy provider.

Indeed - in fact, that's perhaps one of the few senses in which these meters are slightly 'smart'. They are really little more than a measuring and data-storing device, and probably are less 'smart' (in terms of processing) than is the controller in my washing machine/whatever (and infinitely less 'smart' than the phone I carry in my pocket!

Kind Regards, John

 

[Harry Bloomfield]

I've got several of those, but I don't know what they have got to do with the present discussion,

In fact, if you went round all the potential loads in your house and fed them through such a monitor for a while, you'd be able to discover which one was responsible for 'unexpected electricity usage' without having to have a 'smart' meter, wouldn't you?

You would know the instantaneous load, the average load over a time period, but not when an item was switching on and off. A smart meter can tell you all three, assuming there are no other loads being maintained at the same time.

 

[JohnW2]

My first approach would be to switch off stuff.

Yep, even simpler. That's what I would do (coupled with monitoring total overnight usage) - and it certainly would not rely on my having a smart meter.

Kind Regards, John

 

[JohnW2]

You would know the instantaneous load, the average load over a time period, but not when an item was switching on and off. A smart meter can tell you all three, assuming there are no other loads being maintained at the same time.

It depends upon what one means by 'periodically'. If, as would seem likely for a thermostatically-controlled heating load, it came on for one or two (relatively short) periods every hour or so, then all the 'smart' meter's registers would show would be relatively constant usage throughout the night, and certainly would not seem to give any real clue as to the source of the consumption.

In any event, even if that were not the case and, for some reason. the 'smart' meter showed that most of the unexpected usage' was occurring during one or two 30-min periods, I still can't see how that would usually help one to point a finger at any particular 'culprit' - can you?

Kind Regards, John

 

[DetlefSchmitz]

The bit you seem to ignore, is that any sort of tampering with a meter, will be immediately signalled to your energy provider.

Actually it also ignores that extracting firmware is rarely possible as the micro-controller is usually 'locked'. Bypassing the lock mechanism requires extremely expensive processes and a lot of knowledge in most cases.

 

[omph]

The bit you seem to ignore, is that any sort of tampering with a meter, will be immediately signalled to your energy provider.

Reading some docs floating online, there is a case switch which is triggered when the case is taken off, again not impossible to bypass if you know where it is placed and have a spare case lying around. The interesting thing is after 65554 triggers it goes back to 0, a potential point a hacker could exploit.

I appreciate you might see it as impossible but in hacker communities most of the issues you describe as impossible are actually not as difficult as they seem, and tools are usually developed (electronic/software) which can speed up the process. The issue would be how you'd bring that to a mass audience. But some malicious entities could offer this as a service.

Now, when hacking, you obvious don't test on live systems, you build a lab, reverse engineer, and then take it from there, and it's obviously for experts, i.e. software and electronics engineers.

 

[omph]

Actually it also ignores that extracting firmware is rarely possible as the micro-controller is usually 'locked'. Bypassing the lock mechanism requires extremely expensive processes and a lot of knowledge in most cases.

Depends on the system, but newer computers do employ secure boot, TPM controllers so yes it is more difficult. I don't know if SMs employ such features though. In such cases you can either see if you can get the firmware through some other means and overwrite, write your own firmware, or modify an open source one, or take a dump of memory. It's very likely they are modifying open source software for their hardware needs. These are embedded systems, so constrained resources, less complexity and thus easier to work with.

 

[DetlefSchmitz]

Depends on the system, but newer computers at least do employ secure boot, TPM controllers so yes it is more difficult. I don't know if SMs employ such features though. In such cases you can either see if you can get the firmware through some other means and overwrite, write your own firmware, or modify an open source one, or take a dump of memory. These are embedded systems, so constrained resources, less complexity and thus easier to work with.

Whilst I do not know which processor a typical smart meter uses, I believe this to be wishful thinking. Most hacking on smart meters has been done on the external data and on the US systems, admittedly with some interesting results. However, I seriously doubt that the firmware in a UK smart meter can be dumped in any realistic way. If it can then someone has made a serious cock-up.

 

[omph]

Whilst I do not know which processor a typical smart meter uses, I believe this to be wishful thinking. Most hacking on smart meters has been done on the external data and on the US systems, admittedly with some interesting results. However, I seriously doubt that the firmware in a UK smart meter can be dumped in any realistic way. If it can then someone has made a serious cock-up.

Bugs are an eternal problem within software. How do you think you can get pirated encrypted ROMs if it's impossible to bypass ? Just because someone hasn't bothered hacking a SM's inner systems, doesn't mean it's impossible. They are still pretty new to the market. Look at the CPU speculative execution scandal in 2016 - that tech had been around for a very long time and no one picked up on how insecure and dangerous it was - and that's just one example of many.

Also, there are a lot of assumptions being made about the architecture of the system - it depends on the system and how it's been designed. There are different types of firmwares too, some manage HW only, others are more software driven, others may point to boot loaders and/or load other subsystems. There are many points of entry here. Higher level firmware is easy to breakdown, but you must remember, in order for the CPU to read the firmware it must be decrypted, and operational code for more complex systems is almost always thrown into RAM. As long as you have the hardware, you can find a way to bypass it.
There is currently a project going on in the linux community whereby a team of hackers are reverse engineering the Apple M1/2 so they can boot *nix OSes. That is FAR more complex than a SM and is still doable.

 

[JohnW2]

I know I'm hardly a person who should be saying anything about threads 'which go off on tangents', but this tangent seems to be a particularly silly one - which, although quite possibly of considerable interest in some circles, surely has no place in a "DIY electrics" forum, does it?

Kind Regards, John

 

[DetlefSchmitz]

... but this tangent seems to be a particularly silly one - which, although quite possibly of considerable interest in some circles, surely has no place in a "DIY electrics" forum, does it?

It doesn't interest you so it has to be stopped? By all means make your request with the mods. Or perhaps stop reading it?