£20 a year for a web interface? Hmm, not convinced.
Simply measuring and monitoring energy usage helps people conserve, and helps avoid unexpected high bills while 'leccy prices are rising as quickly as they are now. Consumption monitors save cash and the planet, which is why UK.gov is keen on getting smart meters in every home, although these are probably rather less polished …
I'd be more likely to consider something like this if the "hub" stored the data and provided an internal web interface. £20 a year for that data to be sent up to their servers (where it's out of your control) so you can use their pretty web interface doesn't sound too appealing to me at least.
Its in-phase - ie resistive load - you get billed on as a domestic customer. The power factor is assumed to be 1.0 in a normal domestic house, although these days that's unlikely to be true - it won't be far off anyway.
Bigger customers will have a modifier applied depending on how inductive a load they are likely to be running. REALLY big customers get billed on the actual load presented - but they invariably have a substation on-site. Once you get to low PF then you can run into current problems on high power installations so estimating is no longer good enough.
They charge you for actual energy consumed, i.e. watts. For industrial users they measure VA to encourage them to correct their power factor.
Author explains nice and clearly the limitations of metering with a simple current transformer rather than a plug-in device, which can measure voltage and power-factor.
I can understand why this thing has a web interface - it's so no-one needs to faff around with dynamic DNS, DMZ's or port forwarding on their router to be able to access their data - and that's all well and good.
But £20 a year? For basically use of a (hardly complex) website?
Not to mention the fact that if they go out of business, that website goes down and you're left with useless hardware and no access to historical data, plus any privacy implications there may be.
You briefly mentioned:
"AlertMe can track energy use for the whole house online, and I can turn stuff on and off from my phone or laptop while I'm out to confuse would-be burglars."
How's that work, then? You didn't mention that in any further detail at all. Obviously that can't be achieved via the clamp sensor thingy - does it mean that if you have the plug-in monitoring widget, you can turn whatever is connected to that on/off remotely?
Clamp meters are really only a rough guide. As you found, they measure inductive loads rather badly... Just as they're banned all those nice resistive filament lamps too.
A lot of us hobbiests who play with ATmega based board, like the JeeNode, have hooked various detectors onto the front of electric and gas meters. Modern meters are easy as they blink a light for every watt/hr you use. Older ones with the spinning disk can also be reliably read with an IR detector.
Then we have fun hacking the communication protocol used by those remote sockets so we can control them with our own software.
It usually works out cheaper, and more reliable than these consumer devices - not to mention much more fun :-D
I'm not an electrician but you state you have Solar-PV panels and pump excess back into the grid.
I'm not exactly sure how your house is wired but, having looked up some info on this, couldn't you put the clamp on anywhere before it goes into the main consumer unit or fuseboard?
So, it measures consumption for the whole house and you can't break it down by applicance?
I just got myself a plug-in meter for £20 (http://www.amazon.co.uk/Plug-In-Power-and-Energy-Monitor/dp/B000Q7PJGW) and it does the job much more accurately and usefully. (Not on comission here, btw.)
Not only does it let you see how much individual applicances use, it also measures the power factor (PF), VA, watts etc. so you can see the difference between what you're being billed for and what is "useful" energy, real/reactive power and all that geeky goodness.
Fair enough, you can't access it through the web or save stas after you switch off, but for £20 I'm not complaining!
... but its nowhere close to 0.5%. Its much much much worse - as all these cheapo meters are. I checked it out against a calibrated megger multitester and the rms voltage was 6% out, the PF was 4% low etc etc. At one point it was assuring me that my mains voltage was 243V when in reality it was 256V (decent meter and a UPS showed me the same figure). The high voltage was traced to a cabling insulation fault (L->N) on the pavement outside which melted some tar as well as cable sheath in the middle of a Champions League semi-final.
Hang a fan off the end of the meter and see what the PF is - it'll be a fair bit lower than you'd expect.
It costs peanuts - there's no way its going to be accurate to 0.5%. You have the IDENTICAL model to me and I can assure you that once you take into account the cumulative inaccuracies it isn't even close to 5% accuracy.
Its a useful cheap tool. Its not accurate to within 10% but what does that matter? 10% of bugger all is nothing mmm? 10% of "fuck me it uses how much?" is still loads.
Now maybe its been re-engineered since I bought mine (altho they said 0.5% too) but don't expect much for a couple of tenners.
Not worth making a fuss about - useful cheap tool. Not that accurate but as an indicator of usage it works.
Of course the clamp (OWL etc) will be much coarser resolution than a plug-in meter. It has to measure current induced from a single-turn coil and then multiply by a stored estimate of the line voltage (which may be only settable in 10V increments). This means that the resolution limitations of the original ammeter are exaggerated to a large degree.
The plug-in (as mentioned previously) can see the actual voltage and power factor in real-time.
It would be good if the AlertMe system was capable of using the voltage reading from the plug-in to improve accuracy of the house consumption (clamp). Unfortunately from the review this doesn't seem to be the case...
I think you are mixing the two up. Even with poor accuracy, it is useful to have good resolution because small changes can be detected: think analogue meter pointer vs digital display. Displaying a resolution to the individual watt and yet only having large steps can only be explained by a crappy ADC with insufficient bits (or bits being thrown away somewhere along the line). A 16-bit ADC, hardly expensive nowadays, would be able to measure up to 65kW with 1W resolution, which gives loads of headroom on a domestic supply. The display only updates every few seconds so plenty of filtering can be done to reduce noise.
Not sure why you are concerned about the single turn of a current transformer. Only the primary of the current transformer/clamp is a single turn; the secondary can have as many turns as you like, and as the OWL's is some 40 ohms (or more, as it may include a load resistor to protect the coil when not plugged in) I'd guess that's quite a few. (The coil is not clamped around the meter tail; the magnetic core is instead.)
I would also hazard a guess that the plug-in meters use current transformers too. I would have thought it would be much easier to get reasonable accuracy out of those without calibration than an extremely low value resistor (even 6 milli-ohms would dissipate 1W at 13A, and as far as I know you can't buy a small off-the-shelf 6 milli-ohm resistor, so it would be down to the vaguaries of the dimensions of a PCB track).
Apologies if you know all this already but I inferred from your reply that you were misunderstanding.
I have always assumed that the main reason why whole-house metering systems (such as the OWL and this one) don't measure voltage is that a consumer-accessible source of voltage (i.e. a mains socket) is often not available near a meter tail (think external meter cabinets).
As the author states, measuring the current waveform with a current transformer (clamp) without reference to the voltage waveform makes it impossible to tell which direction the current is flowing in.
Moreover, you have to put a clamp on a single-core cable, i.e. a meter tail as opposed to the twin-and-earth which goes round the rest of your house, or the flow and return currents will cancel and you will get a reading of zero.
Some solar installations are connected to a separate consumer unit so you have two pairs of meter tails which join: one carries the current consumed by the house and one carries the current generated by the panels. In that case, you can put the clamp round one of the tails which emerge from the main consumer unit, and you will then measure the house current.
Other solar installations (like mine) are connected to a circuit breaker within the existing consumer unit. You then have two, fairly impractical, possibilities for measuring the house current:
1) Split all the twin an earth cables emerging from the consumer unit apart from the solar feed, and put one of each core through the clamp to sum the individual house loads. (You have to make sure they all go through in the same direction or you will subtract instead of sum.)
2) Split the solar feed twin and earth cable and feed one of the cores through the clamp in addition to a meter tail. If you get the phases right (choice of core and direction) you will subtract the solar current from the grid current and get the house current.
The second is more practical, provided you can get them both in the clamp, but still requires fiddling about in a way only an electrician would be qualified to do. The job can be made easier by using more than one clamp, but only one is provided in this kit.
If this device managed to save me 10% (doubtful) of my current annual bill (Circa £500), I'd save enough to cover the £50 to buy the main device. With the extras and subscription, even if I did achieve 10%, it would still be 2-3 years before the saving met the original and ongoing costs relating to this device. By which time it would be the 8086 of the 'digi-meter' world and probably need to be replaced.
If this was a £20-30 one off outlay, I'd consider it. At these prices only someone with more money than sense would buy it, which kind of defeats the point! When certain energy providers are giving away energy monitors (under certain conditions) this although an advance, is probably going to be received as well as Betamax.
If you think the AlertMe setup is too expensive, just try digging into the numbers behind the national smart meter rollout. The initial estimates were giving a payback period of about 15 years. That clearly looked a bit crap, so OFGEM has managed to fix it by doubling the estimated savings that we're all going to make (YAY!). That's in raw cash terms - completely ignoring the opportunity costs involved.
Here's a request for the Reg energy desk - please take some time out from moaning about windfarms and do some digging into the real costs of the proposed smart meter rollout. After all, at 11.3 billion quid it will be on a par with that other government-backed IT boondoggle, the NHS IT system.
A nuke - because electricity was going to be too cheap to meter thanks to nuclear power. Remember that promise?
Well, as an long time employee on one of the largest metering companies, I sense a bit of bias in the beginning of your article:
"smart meters [...] are probably rather less polished than AlertMe's kit".
What!? The smart-meters of today can give you immensely more accurate data and the in home units (IHU's) coming online are on par with any other such kit. As for the functionality is miles better too as they can combine multi-energy and water in their monitoring scheme.As for reporting and web interfaces, those are available already but utilities want to sell them on, much like mobile comms did with SMS's.
Add to that, that the entire combo is sold to utilities at less than £30 (although I'm sure they'll sell it on a £££'s), and then you get an idea of the fast buck everyone's trying to make.
Color me unimpressed.
I have a number of measuring devices including real power company meters, a Sentron PAC3200, two of the clamp on devices and two of the plug in units. Only the Sentron seems to be highly accurate but at a list price over $1k, you would expect it to work well. The real electricity meters are within about 1 percent of the old dial type. One RF device is about 10 percent and the other is closer to 20 percent. The plug in types are very good with resistive loads but of the power factor isn't leading or lagging, then they are very wrong. CFLs have bad power factor that neither leads or lags. My new high efficiency reverse cycle A/C heater with its variable compressor is an example of a load that most of these devices can't cope with very well.
For most meters, the best trade off between cost and decent data seems to be by using the relay contact on many meters which can closes a contact every 1 watt hour or so. A one-wire counter can give you very accurate data based on how fast you care to read it within reason.
Interesting you say that clamp meters can't measure accurately your air-conditioning unit, as I would assume that if it is the inverter type (you say it has high efficiency) it would have power factor correction like mine does. But I also assume that by the use of $ you are in the US, so perhaps regulations don't demand that there? The PFC circuit in my aircon unit is on a separate PCB and so looks like it is an "optional extra" for some markets.
Having said that, my OWL monitor when connected to my PV system is pretty erratic at low outputs, so I guess it's being affected by harmonics from the grid-tied inverter (or the inverter is lying about the stable nature of the output it is producing).
So all we need is out-of-phase appliances and we'll be laughing all the way to the bank, as the utilities can't read what they really use. Don't get me going on CFLs, we either pollute the planet with mercury, or carbon if we don't switch to them. My Maplin plug-in wattmeter (about a tenner 5 years ago) is reading 25 as I type this on an HPG70 laptop, and wanders between 0.5 and 1 when nothing is plugged into it.
Clamp on meters are fairly useless on domestic lines as the power factor can vary a great deal.
Plus there's a much bigger problem. The device doesn't seem to have any usable interface. There's a GUI application which looks like it was made by and for idiots. One useful application would be to find out when certain appliances have finished. You can find that out via the power consumption. However since you cannot interface yourself with the device, there is no chance to do anything with the data.
Home automation never took off in the days of building yourself a computer with a soldering iron, and it isn't going to now. First off AlertMe attempted a half-baked and expensive self-install burglar alarm, which was expensive, unreliable and wasn't suitable if you had pets, but which allowed users to turn their lights and tvs on and off by remote control. It certainly didn't seem to be a big seller and, judging by its support forums, even the beardie-weirdies who appear to have been the only purchasers weren't happy with it.
When that failed, Alert-Me decided to try to ride the green wave by repositioning the same flaky technology as an energy management system, with the useless home automation capabilities as its unique selling point. But you can buy the useful bit just as cheaply elsewhere, and ever since the days of the Nascom ordinary people haven't wanted, needed or bought the ability to turn their lights and televisions on and off by remote control.
The are two common factors in both business models. The first is a greedy attempt to extract an overpriced recurring fee for little more than a web interface. I am sure the Cambridge Angels love what the recurring charge does to the long-term revenue projections, but others are, and will be, less greedy and offer more. The second is an astonishing ability to raise funding using the latest fads as a cloak for what is really still the same old home automation concept. You won't buy their stuff, but if you have a half-baked start-up and want to raise cash these are the guys to call.
... try buying one to measure the draw for each circuit inside a power panel (say, 12-24 breakers with snmp interface - ~4000squid!)
The nasty spiky load of cheap CFLs is just like the kinds of loads you used to find on switchmode computer PSUs until they were forced to fix things up due to the high neutral currents that'd flow in multiphase office installations. How many other cheap, low power switchmode units are as bad as this?
The stupid thing is that clampons can be very accurate if they used hall effect transducers and a wire to monitor the volts.
My neighbours have professionally installed security systems, which presumably cost upwards of £1000. Whenever the alarms go off, we all ignore them because from previous experience we know that they were set off by the cat or by electrical malfunction. These alarm systems seem to do nothing else but ring a loud bell. The owners of the houses know nothing about the alarm until they get home and pick up their answerphone message from one of the neighbours telling them that their alarm went off. I'm guessing that the alarms aren't wired up to the police station - and even if they were, I'm not sure our local constabulary would turn out.
I have an Alertme system which cost about £150. If the alarm goes off, it too rings a loud bell inside the house. But it also sends me an email and a text message. I can check on the house webcam and if all looks well I can reset the alarm remotely over the web with my phone. It can also tell me whenever certain people (children for instance) enter and leave the house.
The Alertme system always emails me if it can't see my hub and emails me again when the hub re-connects. It also emails me if it hasn't seen certain detection devices for a while.
It's true, the web interface is clunky, slow and not hugely well designed.
The telephone support is absolutely fantastic.
I hope you'll be reviewing other options, because as far as I know, AlertMe are the only vendor that requires you to pay a subscription fee to use their kit.
The company is also painfully slow at bringing new products to market - it took an incredibly long time for them to bring the actual energy /display/ to market - and when it did, it looked nothing like the original design which they claimed they'd been 'testing' - /that/ unit seems to have been just sold to British Gas (one of their sugar daddies), who seem to get a lot more of their time and attention than their regular consumer customers.
Plenty of other kit out there - why not do a group test of some of the others like OWL, Wattson etc - CurrentCost are a good one too, they just released a meter reader that works by reading the LED pulses of many modern meters, offering much higher accuracy.
Electricity meters have access to, and can intercept, live and neutral and measure watts, as has already been pointed out in these comments. The induction type (spinning disc) has two coils, one with many turns, across the mains supply (i.e. measuring volts) and one with few turns, in series with your house (i.e. measuring amps). Electromagnetic magic multiplies these together instantaneously, so not only do they measure true watts, but will rotate backwards if you are feeding energy back into the grid. (Hence the mad scramble by your supplier to change to an electronic type when you tell them you have a microgeneration system.) The electronic types are designed to give the same reading as the induction type, and so these too sense both voltage and current.
While plain old spinning disc type meters can be reliable and accurate for much longer than electronic smart meters (40+ vs. ~15-20 years) the reason behind the scramble to substitute them is two-fold:
1) Utilities have long requested as a standard that, mechanical registers positively count current flow in either direction (before micro-generation, this was a simple way to "save" some money by re-wiring the cables)
2) As mechanical meters wear out over time they lose accuracy. This penalty is small for the utility when supplying a home but is augmented when measuring current in the other direction. (i.e. a ~2% loss in a customer's bill is a small matter but a 6+% over compensation from erroneous reading is adding up to too much of a burden)
As a long time engineer in precisely such a company, I can assure you this is not the case.
That would be plain illegal. Meters nowadays are covered by the MID (measuring instruments directive) and this has plenty of positive provisions for the consumers too.
Furthermore, utilities have many more legal tricks to fleece their customers that resorting to this pettiness would be laughable. (Why risk this when you can just tell ofgem that the rollout of smart meters won't pay for itself through efficiencies (which it will) but will require constant margin hikes to compensate for their "investment"?)
AC for obvious reasons
Biting the hand that feeds IT © 1998–2019