I currently don't have a "Whole House" measuring device. In dataMine I can see a one only, circular dial, labelled Watts.
That's confusing - these two statements should be conflicting... The dial showing Watts should only show the "Whole House" device. If you don't have a device configured in the "Whole House" category, then I would have expected this dial to be 0.
I assume it's showing power consumed at the current instant?
Yes, it simply shows the state of the "Whole House" device. For some devices (eg NorthQ) I think this updates every 15 minutes, and for others, it might be every 6 seconds (or anything else!). It simply reflects the value of the device.
Assuming everything in the house was individually metered and from what I have read; I assume a pie chart pops up to the left of the dial based on the categories you have eg heating, lighting, etc?
Exactly. If you take a look in the manual there's an example. It's a donut chart which is effectively 2 pie charts one inside the other. The inner one shows the value of each device, and the outer shows the combined values in the category. I think it works quite well since the colours allow a quick look at what categories are consuming the power now, and you can also see what the actual device is.
Once again this would be an instantaneous reading - noting that depending on polling intervals, it could very easily miss a power consumption event - say like a fridge going on or off or a hot water kettle being used.
Correct - there's no such thing as "real time". So, you'll miss certain spikes etc, but with a reasonable sampling period, it does work very well. My system monitors maybe a dozen or more devices every 6 seconds. It's not perfect, and yes you do miss things, but it does work well, and it gives a good overview of where you're power is going now.
I should also add the my "Whole House" monitor uses a device that looks at the pulse on the power meter, so it is measuring actual power being used, as seen by the meter (and hence is inline with what I'm being charged). This compares actually quite well with the individual appliance monitors that I have (eg on the cooker, washing machine, fridge, computers etc). I don't monitor lights, but Vera does provide a consumption for these based on their state and a user defined power value. Again, this works very well, and ties up very closely with the meter counter.
I'm interested to know if dataMine is just doing instantaneous power measurements or does it integrate power over time to produce energy consumed per day?
Currently, dataMine displays the instantaneous power measurements. For me, these are updated every 6 seconds, so this works out reasonably accurate. There are issues as different things are sampled at different times, but (I think) it does provide a really nice view of your current power consumption. Personally I think this is useful, however I also appreciate that integrating this over time is actually better for some purposes...
So, it’s been mentioned elsewhere, but I have been working on statistics. I actually wrote most of it a while ago, but due to working abroad quite a bit in the past few months, it’s not integrated into the public version yet. I also needed to resolve the database structure, since this will generate a bunch more files! This will produce statistics for each device/variable being logged - this will include min/max, delta (difference between min/max), hourly average… For power, it’s the hourly average that matters - this works out as KWh, and the data will be available for each hour/day/week/month.
Currently, this is as far as I’ve got. I will integrate this into the public version in the coming weeks - once I’ve got the issues out of the current version with the database update. It’s probably worth pointing out that the system retrospectively processes all data, so it will generate this for old data. This also means that if I change the calculations, it can be reprocessed…
Cost is another matter. I might look at this, but it’s a difficult issue as there are different charging schemes (day/night, or $x for first yyy units, then $z per unit, etc) and this makes it inaccurate (in my opinion). Therefore, I think sticking with consumption is best - you can properly compare this across different time periods etc.
I hope that provides an overview of what I do, what dataMine currently does, and what dataMine will hopefully do soon… As always, I’m always open to suggestions… I think once the statistics is working, people will have ideas and we can “easily” refine it if needed…
Cheers
Chris