so a couple of months ago, i installed a TED5000. i was not able to capture the whole house, just the sub-panel. so i miss the garage lights and outlets, and also the Blink. but that turns out to be good for modeling different leaf charging scenarios.
i wrote a perl script to take the TED5000 output (hourly) and compute the E1A and E9A costs for my electrical use. using the blink's statistics i hacked the leaf consumption back in and added in a fudge factor to pick up the usage that's not caught by the TED.
the result is a little counter-intuitive. first of all, with my current consumption (around 800kwh/month) the E1 scenario is cheaper than any E9 scenario i can come up with. furthermore charging the leaf at peak times seems to be marginally cheaper than charging at off-peak times. on average i'm using about 4kwh/day on the leaf. E9A does not become cheaper than E1 until i start using about 7kwh per day on the leaf.
i'm thinking that either the script is buggy, or my continuous load of 300-400kw just drives the E9 cost into the stratosphere.
so the question is - is it legitimate to compute the tiered usage on a day-by-day basis? in other words, the script starts at midnight with daily usage 0 and adds up the usage for each hour. using the daily baseline quantity and the time of day/day of week, i figure out the current rate and add in the cost for the current hour. i think this is why charging the leaf overnight gives a more expensive result - by 3 or 4am i'm already in the middle/higher tiers, and then the rest of the day costs more than it ordinarily would.
or is this wrong, in the sense that any unused lower tiers for a given day must be "banked" to be squared up at the end of the month? in other words if for 30 days the baseline quantity is (say) 300kwh and i (somehow) use 300kwh on one night during off-peak time, is that full 300kwh charged at the lowest tier? my script would allocate the lion's share of that usage to the highest tier, as it's way, way beyond the daily baseline amount of 10kwh.
i think initially i convinced myself that going day by day was equivalent to computing it monthly, but writing the above makes me think that was wrong. reduction to the absurd is often helpful
anyway, anyone have any insights on this? thanks.
i wrote a perl script to take the TED5000 output (hourly) and compute the E1A and E9A costs for my electrical use. using the blink's statistics i hacked the leaf consumption back in and added in a fudge factor to pick up the usage that's not caught by the TED.
the result is a little counter-intuitive. first of all, with my current consumption (around 800kwh/month) the E1 scenario is cheaper than any E9 scenario i can come up with. furthermore charging the leaf at peak times seems to be marginally cheaper than charging at off-peak times. on average i'm using about 4kwh/day on the leaf. E9A does not become cheaper than E1 until i start using about 7kwh per day on the leaf.
i'm thinking that either the script is buggy, or my continuous load of 300-400kw just drives the E9 cost into the stratosphere.
so the question is - is it legitimate to compute the tiered usage on a day-by-day basis? in other words, the script starts at midnight with daily usage 0 and adds up the usage for each hour. using the daily baseline quantity and the time of day/day of week, i figure out the current rate and add in the cost for the current hour. i think this is why charging the leaf overnight gives a more expensive result - by 3 or 4am i'm already in the middle/higher tiers, and then the rest of the day costs more than it ordinarily would.
or is this wrong, in the sense that any unused lower tiers for a given day must be "banked" to be squared up at the end of the month? in other words if for 30 days the baseline quantity is (say) 300kwh and i (somehow) use 300kwh on one night during off-peak time, is that full 300kwh charged at the lowest tier? my script would allocate the lion's share of that usage to the highest tier, as it's way, way beyond the daily baseline amount of 10kwh.
i think initially i convinced myself that going day by day was equivalent to computing it monthly, but writing the above makes me think that was wrong. reduction to the absurd is often helpful
anyway, anyone have any insights on this? thanks.