TOU for SCE in Southern California analysis

[sensij]

One more comment... in columns BL and BM, showing the charges in the standard tiered plan once you get into Tiers 2 and 3, you should also show credits for the hours of generation. Right now, you just show zero for those values. Those credits are calculated at the appropriate Tier 2 or 3 rate, in the same way that you put the net generation credits into column BK for Tier 1.

Edit: It looks like fixing that would knock another $30 off the annual bill for that plan, bring the total down to just under $230.

[sensij]

I should have known the answer would already be in the forum. In this post, an example TOU-D-T bill is provided. Since the bill is almost three years old, I verified that the language in the tariff applicable at that time that addressed the baseline and tier calculations is unchanged in today's tariff. From it, just looking at summertime since each season is handled separately when a billing period covers both, I can see the following:

Baseline allocation = 214.2 kWh (21 days of summer at 10.2 kWh / day)
Level 1 allocation (130% of baseline) = 278.4 kWh
Overall net consumption = -118 kWh
Peak: -367 kWh
Off-peak: 249 kWh

To calculate the % of the baseline that gets allocated to each TOU category, the minus sign is dropped.

% allocated to peak = 367 / (367+249) = 59.58%
% allocated to off-peak = 249 / (367+249) = 40.42%

Therefore, the level 1 allocations are (not sure how the rounding works, but this is what is on the bill)
peak: .5958 * 278.4 = 166 kWh
offpeak: .4042 * 278.4 = 112 kWh

With that figured, the bill then breaks down to
peak, level 1 = -166 kWh
peak, level 2 = (-367--166) = -201 kWh
off-peak, level 1 = 112 kWh
off-peak, level 2 = (249-112) = 137 kWh

The same math works to explain the winter time portion of the bill, with the baseline allocation coming from 9 days * 9.2 kWh / day.

Programming this into a spreadsheet will take more work, but at least we can see what the logic needs to be.

Edit:
Also note that the DWR charges and credits in that linked bill are based on the seasonal net kWh within the bill. That suggests that in a spreadsheet, the published bundled rates (inclusive of delivery, generation, DWR bond, DWR energy credit, etc) should be OK to use, even when energy is piecemealed into the different TOU's and tiers.

[insaneoctane]

One more comment... in columns BL and BM, showing the charges in the standard tiered plan once you get into Tiers 2 and 3, you should also show credits for the hours of generation. Right now, you just show zero for those values. Those credits are calculated at the appropriate Tier 2 or 3 rate, in the same way that you put the net generation credits into column BK for Tier 1.

Edit: It looks like fixing that would knock another $30 off the annual bill for that plan, bring the total down to just under $230.

Senji,
Great catch. I certainly appreciate you looking into the spreadsheet in the detail necessary to catch that. I've fixed it now.

[insaneoctane]

Programming this into a spreadsheet will take more work, but at least we can see what the logic needs to be.

Edit:
Also note that the DWR charges and credits in that linked bill are based on the seasonal net kWh within the bill. That suggests that in a spreadsheet, the published bundled rates (inclusive of delivery, generation, DWR bond, DWR energy credit, etc) should be OK to use, even when energy is piecemealed into the different TOU's and tiers.

Wow, my head has been rolling on this one. I followed your link to the Tesla forum and implemented the math the way miimura was suggesting as a starting point, but I had serious reservations about negative baseline percentages. Then when I returned here to status things, I see you've found more clarification and it makes more sense to me. I've adjusted the TOU-T to match the logic you last presented and it does make the TOU-T more appealing, but still not better than the TOU-D-A for me. I'm including a link if you want to see the TOU-D-T changes. The way the TOU-D-T requires month end data to calculate it's values (which are needed to arrive at month end values) seems almost circular, but if I just use the usage in kWh to calculate the month end data necessary to calculate % for baseline, then calculate the $'s I'm okay. I'll warn you that this new math does add a layer of complexity to the math in the spreadsheet to where it has noticeably slowed the calculation time down from near instant to a few seconds (depending on your hardware).

New TOU-D-T calcs:
Solar_TOU_Rev_M14_WIP.xlsm

Thanks again for hunting this understanding down, it's amazing how complicated the TOU-D-T baseline calcs really are comparatively speaking!

[J.P.M.]

Wow, my head has been rolling on this one. I followed your link to the Tesla forum and implemented the math the way miimura was suggesting as a starting point, but I had serious reservations about negative baseline percentages. Then when I returned here to status things, I see you've found more clarification and it makes more sense to me. I've adjusted the TOU-T to match the logic you last presented and it does make the TOU-T more appealing, but still not better than the TOU-D-A for me. I'm including a link if you want to see the TOU-D-T changes. The way the TOU-D-T requires month end data to calculate it's values (which are needed to arrive at month end values) seems almost circular, but if I just use the usage in kWh to calculate the month end data necessary to calculate % for baseline, then calculate the $'s I'm okay. I'll warn you that this new math does add a layer of complexity to the math in the spreadsheet to where it has noticeably slowed the calculation time down from near instant to a few seconds (depending on your hardware).

New TOU-D-T calcs:
Solar_TOU_Rev_M14_WIP.xlsm

Thanks again for hunting this understanding down, it's amazing how complicated the TOU-D-T baseline calcs really are comparatively speaking!

As said, welcome to POCO rate hell - be prepared for many iterations. You're doin' fine.

[sensij]

Thanks again for hunting this understanding down, it's amazing how complicated the TOU-D-T baseline calcs really are comparatively speaking!

Wow, it took you less than 2 hours to roll that in, while most of us Californians were sleeping. Really nice work. It looks good in a spot check, but it will take me longer than that to verify the result by my own calculations.

One interesting use of this is to see how small your array could have been and resulted in the same net bill, keeping the household consumption constant. Under the TOU-D-A, plan, it looks like it could be about 15% smaller before you have to start paying. Another way of looking at is it that if your usage stays the same, and the TOU structure and rate proportions stay the same (yeah, right), you could tolerate panel degradation of about 15% before you have an electric bill again. Of course, the difference between actuals and estimates will prevent that degree of sizing accuracy in a real system.

Actually, that sort of brings up another point... I think the fixed charges associated with each plan need to be paid, there is no way to credit them out by TOU accounting, so even though you are showing a negative number for your TOU-D-A result, there is still likely to be a bill to be paid at the end of the net metering year.

[insaneoctane]

Actually, that sort of brings up another point... I think the fixed charges associated with each plan need to be paid, there is no way to credit them out by TOU accounting, so even though you are showing a negative number for your TOU-D-A result, there is still likely to be a bill to be paid at the end of the net metering year.

Yes, I do have to pay SCE monthly for the fixed charges and I think you're right that they are a different color of money and can't be erased with production.

[insaneoctane]

Q: Anyone know if pvwatts data accounts for daylight savings in the hourly export? Since I'm using it as the basis for my solar output predicts, I'm noticing that the recent daylight savings change would be beneficial to TOU-D-A/B rates because the high rates don't start until 2pm and with DST my 2pm+ just got a bit fatter....I looked at the data and I don't see any change in the data to support the output moving at all during March, so I'm thinking not. If that's the case, I've got some additional margin on my numbers.

[insaneoctane]

I got to thinking (uh oh) and wanted to exercise spreadsheet to see that TOU-D-A would eventually become more expensive, mostly as a validation of all the rate plans. Probably because I'm still surprised that the TOU-D-A seems like such an obvious choice for me and I always question obvious (particularly when it seems to be different than most others and contrary to standard "recommendations"). So, I simply scaled my usage from 90% to 150% of actual and using the data table function in Excel (makes this easy), I created a sensitivity chart of the 4 rate plans vs USAGE. Pretty interesting, you'll see that once I exceed 12,230 kWh/yr I'm better off on the RATE-D plan. I labeled my "baseline" usage from 2014 (9060 kWh). It doesn't look like the other plans are ever better for me.

tou_analysis_sensitivity_2015.03.11.JPG

[sensij]

PVWatts does not account for savings time. If it did, you would find a 25 hour day and a 23 hour day in there when the time changed. The west coast is modeled as UTC-8, and right now, we are actually UTC-7. Further complicating it is the fact that the rate tariffs may have DST adjustment included in the hours that do not align with when DST is actually enacted (I'm looking at you, PG&E).

Although your intended use of this spreadsheet may be, in part, to see how much energy you can use before incurring a bill, I hope that other forum members find this earlier in their shopping process. It could become a very powerful tool to help people understand how to offset their bill cost-effectively, and shows that an array sized for something less than 100% of expected energy consumption can still fully offset their electric bill. (rates will change, of course, but TOU is likely to continue in one form or another for the foreseeable future.)

You keep using "better", but that isn't really true. Right now, TOU-D-A and TOU-D-T are equally good... you will pay the same amount at the end of the year under either. If you were to add more consumption, it is unlikely to preserve your existing hourly distribution, so it is hard to say which plan may be best at that time until you model it (which you can easily do!).

One more minor catch... the minimum monthly charge for -D-A and -D-T will be about $1.79 when you are net negative, not $0.93. The $0.93 would apply if, when combined with your other charges, you exceeded that minimum. I am not suggesting that distinction is worth trying to program in.

[silversaver]

The calculation is never easy, but I'm still suggesting people to go little larger. The worse is by maxing TOU plan and trying to settle with smaller system. The solar price has drop and cost of production is less than $0.10 per kWh for most base on 20yrs output. You know the electricity rate will definitely goes up in next 20yrs and so is the power demand, why start small? You will spend more money in long run by saving a few bucks. It is a simple math without any calculation.

[insaneoctane]

You keep using "better", but that isn't really true. Right now, TOU-D-A and TOU-D-T are equally good... you will pay the same amount at the end of the year under either. If you were to add more consumption, it is unlikely to preserve your existing hourly distribution, so it is hard to say which plan may be best at that time until you model it (which you can easily do!).

Maybe semantics, but I do believe A to be better. I will admit if everything stays at current predictions, then the plans would be the same result. But, I feel there is uncertainty in my predictions, mostly on the production side. For example, I've only had my solar array online for 4 full months, so I have an estimated my generation to the 85% of a 10k PVWatts model. I may get more output, I make it less output. The more margin that I can get in my cost predictions, the better off I am. But, I feel plan A has more margin and therefore is better based off my current assumptions and predictions. I realize that my spreadsheet and predictions are based off my last years usage profile, I have tried to objectify the decision as to which rate plan to be on. I also realize that I might see more cross over between the rate plans if I varied my usage profile, but with so much hourly data I am not up for that!

[russ]

It will change as politics and public opinion changes - not at all permanent. As it is all made up "facts" it is highly likely to be modified.

[insaneoctane]

FWIW, I just switched my rate plan to Tou-d-a and should be effective in my next billing cycle starting on the 18th. Time to put my wallet where my spreadsheet is.

[J.P.M.]

FWIW, I just switched my rate plan to Tou-d-a and should be effective in my next billing cycle starting on the 18th. Time to put my wallet where my spreadsheet is.

Risk nothing, be nothing. Under the Ready, Aim, Fire category - Take the shot. Looks like you've done a lot of homework. You are now one of the go to people for SCE tariffs.