Understanding tiered TOU rates with PG&E solar

[devsolar]

The last paragraph is good guidance - conservation first. That is the cheapest power around.

I've been paying attention in class...

[wwu123]

The way that monthly baselines are calculated and used can be a bit tricky. To see if you went over your baseline, PG&E first sums up all the power you used and subtracts from it all the power you generated. If your total baseline for the month was 240 kWh, and you exported 150 kWh of peak, 50 kWh of partial-peak, and used 300 kWh of off-peak, then your total consumption for the month would be 100 kWh (300 - 150 - 50 = 100). Well under the baseline. All credits are charges would be calculated using Tier 1 rates. If, on the other hand, you only exported 50 kWh peak, 25 kWh partial peak, and consumed 400 kWh off peak, you would be at 325 kWh consumed for the month, which would be 85 kWh over baseline, and then you'd be subject to Tiers. The baseline is proportionally distributed amongst all the time periods. To calculate how much baseline is allocated to each tier, you divide the baseline, 240, by the total amount of power you consumed, 325 in this case, which equals 0.738446154. Then multiple that number by the total amount of power you consumed or produced for the time periods:

Peak: 0.738446154 * 50 = 36.923 kWh baseline
Partial-peak: 0.738446154 * 25 = 18.462 kWh baseline
Off-peak: 0.738446154 * 400 = 295.385 kWh baseline (yes, this is more than the total baseline, but that's how PG&E calculates it)

Ah, thanks devsolar, for spending the time to completely break it down! I think that fully demystifies it.

It does seem with those calculations that it is hard to bank net generation in signifcant amounts at peak Tier 3 rates, as one would have to be a large off-peak user on those summer months as well. On the plus side, you've shown that the monthly off-peak baseline can actually end up somewhat higher than the normal baseline, so more of the off-peak usage actually stays in Tier 1 and Tier 2 - the October bill was a great example. So while those 20 kwh of net generation were mostly credited at Tier 1 and 2 rates for about $3.70 total, they also effectively created 20 more kwh of off-peak baseline and Tier 2 - each kwh staying out of off-peak Tier 3 has a net savings of about $0.15, so it's another $3.00 benefit.

Odd how it works, but it makes some intrinsic sense. Put another way, it is in essence crediting peak/PP net generation at peak Tier 3 rates closer to, during those shoulder months like October, when offsetting use in the same month. Whereas in those prime solar months of summer, pretty much the net generation to be banked for later months only credits at Tier 1! Doesn't it somewhat mean then that a kwh generated in the shoulder months ($0.28+$0.15) is worth about 50% more than in summer ($0.28) - interesting implications for whether panel tilt should be optimized for summer (maximize annual generation) or not (maximize for savings)?

[wwu123]

The installer came out yesterday to do the site survey. Biggest thing was looking at the impact of afternoon shading by a couple of big trees, which happens at a different time of day for each roof plane - two south facing and one east-facing. The rough assessment is the east-facing gets shadowed first, so the panels would be split (due to space) between the best south roof plane and the other south-facing roof (maybe an hour less production in the mid-afternoon).

He took a bunch of shading measurementns using a Solar Pathfinder and will run it through a detailed calculator to generate the shading derate factors for each roof for each month. He said he'll also assess whether to optimize generation or optimize $$ savings. The $$ calculations won't have my exact daily TOU usage however, he just has my average monthly electric costs, so it must make some average assessments of Tiered and TOU effects. If I can get the detailed shading data from him, I'm going to try to run the detailed numbers myself for a year based on exact hourly consumption (downloaded from my Smartmeter data) and hourly insolation (from PVWatts).

[inetdog]

The installer came out yesterday to do the site survey. Biggest thing was looking at the impact of afternoon shading by a couple of big trees, which happens at a different time of day for each roof plane - two south facing and one east-facing. The rough assessment is the east-facing gets shadowed first, so the panels would be split (due to space) between the best south roof plane and the other south-facing roof (maybe an hour less production in the mid-afternoon).

He took a bunch of shading measurementns using a Solar Pathfinder and will run it through a detailed calculator to generate the shading derate factors for each roof for each month. He said he'll also assess whether to optimize generation or optimize $$ savings. The $$ calculations won't have my exact daily TOU usage however, he just has my average monthly electric costs, so it must make some average assessments of Tiered and TOU effects. If I can get the detailed shading data from him, I'm going to try to run the detailed numbers myself for a year based on exact hourly consumption (downloaded from my Smartmeter data) and hourly insolation (from PVWatts).

Is he proposing two inverters (or one inverter with two MPPT inputs) for the two roof areas?
That would maximize your output as the shade worked its way across the roof.

[wwu123]

Is he proposing two inverters (or one inverter with two MPPT inputs) for the two roof areas?
That would maximize your output as the shade worked its way across the roof.

Enphase microinverters all around as the afternoon shading makes its way across the roof over a couple of hours

[inetdog]

Enphase microinverters all around as the afternoon shading makes its way across the roof over a couple of hours

Good choice. I would say in that area at least, your installer's proposal seems competent.
A range of facing directions to deliver power more uniformly over the course of the day is more important to the off-grid user than to a grid-tied user who is mostly interested in total output per square foot of panel.

[reworld]

what if total consumption for the month is negative? how do you calculate tou/tier?

Dear Devsolar,

Thanks a lot for this calcualtions. I was wondering how does PG&E calculate in the eventual case that the consumption net consumption from Grid for the full month is negative? Which Tier rate applies to each tou period?
thanks for this clarification!
example
August 2012:
Baseline: 240 kWh
Peak: -151 kWh
Partial-peak: -120 kWh
Off-peak: 20 kWh
Total consumed for the month: -251 kWh
All consumption was in Tier 1?? or Went over baseline by xx, split into 2 or 3 tiers?

Peak:
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??
Partial Peak
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??
Off-Peak:
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??

THANKS A LOT!

The way that monthly baselines are calculated and used can be a bit tricky. To see if you went over your baseline, PG&E first sums up all the power you used and subtracts from it all the power you generated. If your total baseline for the month was 240 kWh, and you exported 150 kWh of peak, 50 kWh of partial-peak, and used 300 kWh of off-peak, then your total consumption for the month would be 100 kWh (300 - 150 - 50 = 100). Well under the baseline. All credits are charges would be calculated using Tier 1 rates. If, on the other hand, you only exported 50 kWh peak, 25 kWh partial peak, and consumed 400 kWh off peak, you would be at 325 kWh consumed for the month, which would be 85 kWh over baseline, and then you'd be subject to Tiers. The baseline is proportionally distributed amongst all the time periods. To calculate how much baseline is allocated to each tier, you divide the baseline, 240, by the total amount of power you consumed, 325 in this case, which equals 0.738446154. Then multiple that number by the total amount of power you consumed or produced for the time periods:

Peak: 0.738446154 * 50 = 36.923 kWh baseline
Partial-peak: 0.738446154 * 25 = 18.462 kWh baseline
Off-peak: 0.738446154 * 400 = 295.385 kWh baseline (yes, this is more than the total baseline, but that's how PG&E calculates it)

Lemme give you some examples of how our first few bills were:

August 2012:
Baseline: 240 kWh
Peak: -151 kWh
Partial-peak: -98 kWh
Off-peak: 293 kWh
Total consumed for the month: 44 kWh
All consumption was in Tier 1. Credits and charges were:
Peak: $-42.10
Partial Peak: $-16.68
Off-Peak: $28.66
Total: $-30.12

September 2012:
Baseline: 247.5 kWh
Peak: -91 kWh
Partial-peak: -51 kWh
Off-peak: 209 kWh
Total consumed for the month: 67 kWh
All consumption was in Tier 1. Credits and charges were:
Peak: $-20.37
Partial Peak: $-8.68
Off-Peak: $20.55
Total: $-13.60

October 2012:
Baseline: 210 kWh
Peak: -2 kWh
Partial-peak: -18 kWh
Off-peak: 298 kWh
Total consumed for the month: 278 kWh
Went over baseline by 68, split into 3 tiers:
Peak:
Tier 1: -1.511 kWh * 0.27883 = $-0.42
Tier 2: -0.453 kWH * 0.2964 = $-0.13
Tier 3: -0.036 kWh * 0.44653 = $-0.02
Total credits: $0.57
Partial Peak
Tier 1: -13.597 kWh * 0.17017 = $-2.31
Tier 2: -4.079 kWh * 0.18775 = $-0.77
Tier 3: -0.324 kWh * 0.33788 = $-0.11
Total credits: $3.19
Off-Peak:
Tier 1: 225.108 kWh * 0.09781 = $22.02 (see, off-peak baseline is higher than the total monthly baseline)
Tier 2: 67.535 kWh * 0.11538 = $7.79
Tier 3: 5.360 kWh * 0.26551 = $1.42
Total Charges: $31.23
Total: $27.57

Now compare things to December 2012 when production sucked and consumption was up:

December 2012:
Baseline: 273 kWh
Partial-peak: 65kWh
Off-peak: 388 kWh
Total consumed for the month: 453 kWh
Went over baseline by 180, split into 3 tiers:
Partial Peak
Tier 1: 39.172 kWh * 0.11776 = $4.61
Tier 2: 11.752 kWh * 0.13533 = $1.59
Tier 3: 14.076 kWh * 0.28546 = $4.02
Total charges: $10.22
Off-Peak:
Tier 1: 233.828 kWh * 0.10189 = $23.82
Tier 2: 70.148 kWh * 0.11947 = $8.38
Tier 3: 84.204 kWh * 0.26959 = $22.65
Total Charges: $54.97
Total: $65.07

So while I did get into Tier 3, it wasn't that bad, and it was still cheaper than the previous year's December bill by over $200.

[wwu123]

Dear Devsolar,

Thanks a lot for this calcualtions. I was wondering how does PG&E calculate in the eventual case that the consumption net consumption from Grid for the full month is negative? Which Tier rate applies to each tou period?
thanks for this clarification!
example
August 2012:
Baseline: 240 kWh
Peak: -151 kWh
Partial-peak: -120 kWh
Off-peak: 20 kWh
Total consumed for the month: -251 kWh
All consumption was in Tier 1?? or Went over baseline by xx, split into 2 or 3 tiers?

Peak:
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??
Partial Peak
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??
Off-Peak:
Tier 1: ?
Tier 2: ?
Tier 3: ?
Total credits: $??

THANKS A LOT!

Heh, heh, not devsolar, but since I asked the original questions starting this thread, about once a year I do try to wrap my head around tiered TOU. I think you may be asking a two-part question, first what if the net consumption was negative for the month, say -40 kwh, and then second, what if the net consumption was negative but surpassed the baseline in the opposite direction, per your example -251 kwh is below -240 kwh. For the second part, there's no reason necessarily that PG&E implements baselines in the negative direction - i.e. progressively penalizing you for excess consumption in a month does not mean progressively rewarding you for additional excess generation in a month. So most likely the answer for both parts is that everything is in Tier 1, so whether -40 kwh or -251 kwh for the month, both are less than Tier 1 baseline of 240 kwh, so everything is calculated a Tier 1 rates in each TOU period.

But IF there are negative baselines, and they use the same formulas in the negative direction, then I believe it would be roughly:

Peak: 0.956 * -151 = -144 kWh neg baseline
Partial-peak: 0.956 * -120 = -114 kWh neg baseline
Off-peak: 0.956 * 20 = 19 kWh neg baseline

Went below negative baseline by 11, split into 3 tiers:
Peak:
Tier 1: -144 kWh * 0.27883
Tier 2: -7 kWH * 0.2964
Tier 3: 0 kWh * 0.44653

Partial Peak
Tier 1: -114 kWh * 0.17017
Tier 2: -6 kWh * 0.18775
Tier 3: 0 kWh * 0.33788

Off-Peak:
Tier 1: 19 kWh * 0.09781
Tier 2: 1 kWh * 0.11538
Tier 3: 0 kWh * 0.26551

So basically about 11-12 kwh would get an additional Tier2 credit, about $0.017 per kwh. If you generated even more to creep into Tier 3, each kwh past the Tier 2 baseline would get additional Tier3 credit of $0.167 per kwh compared to Tier 1 credits. Again, only IF PG&E has made baselines symmetric in the negative direction in their computations (there are no negative baselines articulated in their rate sheets).

So that's the theory, anyone actually have experience with reality? Seems like would have to be someone who's drastically oversized their system for summer months, e.g. doesn't need A/C in summer but sized array larger to run heat pump in low sun winter months....

[sensij]

For the second part, there's no reason necessarily that PG&E implements baselines in the negative direction - i.e. progressively penalizing you for excess consumption in a month does not mean progressively rewarding you for additional excess generation in a month. So most likely the answer for both parts is that everything is in Tier 1, so whether -40 kwh or -251 kwh for the month, both are less than Tier 1 baseline of 240 kwh, so everything is calculated a Tier 1 rates in each TOU period.

But IF there are negative baselines, and they use the same formulas in the negative direction, then I believe it would be roughly:

Both SDG&E and SCE use negative baselines, so there is some justification for thinking that PG&E might, as well.

[reworld]

Thanks a lot wwu123. I think your calculations make a lot of sense!

Heh, heh, not devsolar, but since I asked the original questions starting this thread, about once a year I do try to wrap my head around tiered TOU. I think you may be asking a two-part question, first what if the net consumption was negative for the month, say -40 kwh, and then second, what if the net consumption was negative but surpassed the baseline in the opposite direction, per your example -251 kwh is below -240 kwh. For the second part, there's no reason necessarily that PG&E implements baselines in the negative direction - i.e. progressively penalizing you for excess consumption in a month does not mean progressively rewarding you for additional excess generation in a month. So most likely the answer for both parts is that everything is in Tier 1, so whether -40 kwh or -251 kwh for the month, both are less than Tier 1 baseline of 240 kwh, so everything is calculated a Tier 1 rates in each TOU period.

But IF there are negative baselines, and they use the same formulas in the negative direction, then I believe it would be roughly:

Peak: 0.956 * -151 = -144 kWh neg baseline
Partial-peak: 0.956 * -120 = -114 kWh neg baseline
Off-peak: 0.956 * 20 = 19 kWh neg baseline

Went below negative baseline by 11, split into 3 tiers:
Peak:
Tier 1: -144 kWh * 0.27883
Tier 2: -7 kWH * 0.2964
Tier 3: 0 kWh * 0.44653

Partial Peak
Tier 1: -114 kWh * 0.17017
Tier 2: -6 kWh * 0.18775
Tier 3: 0 kWh * 0.33788

Off-Peak:
Tier 1: 19 kWh * 0.09781
Tier 2: 1 kWh * 0.11538
Tier 3: 0 kWh * 0.26551

So basically about 11-12 kwh would get an additional Tier2 credit, about $0.017 per kwh. If you generated even more to creep into Tier 3, each kwh past the Tier 2 baseline would get additional Tier3 credit of $0.167 per kwh compared to Tier 1 credits. Again, only IF PG&E has made baselines symmetric in the negative direction in their computations (there are no negative baselines articulated in their rate sheets).

So that's the theory, anyone actually have experience with reality? Seems like would have to be someone who's drastically oversized their system for summer months, e.g. doesn't need A/C in summer but sized array larger to run heat pump in low sun winter months....

[reworld]

Dear sensij,
thanks for your insight on negative baselines. Is it possible you would post or refer to an example where SDG&E or SCE use them? Or a reference where they state to use them?
That would be great! thanks again! =)

Both SDG&E and SCE use negative baselines, so there is some justification for thinking that PG&E might, as well.

[reworld]

btw do you know if SCE and SDG&E calculate the same way as described by wwu123 and devsolar? meaning using a monthly allowance, and assigning proportional baselines for each tou period?
thanks !

Dear sensij,
thanks for your insight on negative baselines. Is it possible you would post or refer to an example where SDG&E or SCE use them? Or a reference where they state to use them?
That would be great! thanks again! =)

[sensij]

btw do you know if SCE and SDG&E calculate the same way as described by wwu123 and devsolar? meaning using a monthly allowance, and assigning proportional baselines for each tou period?
thanks !

SCE and SDGE each have a scheme to divide up the baseline allocation among the TOU periods, but none of the three does it the same way.

With regard to SCE, TOU-D-T is the time of use plan that has a baseline allocation, and a long discussion of the rates can be found in this thread, and in other threads linked within it. This post in particular shows how calculation is handled in the event 2nd tier generation is hit, and was verified against an actual SCE bill.

SDGE doesn't really have a baseline allocation with multiple tiers in any of their TOU plans at the moment. TOU-DR is the closest, with a baseline credit, and you can see their allocation scheme with respect to the baseline in this post.

The recent decision by CPUC on the rate reform proceeding also laid a framework for future rollout of TOU, including more plans with baseline allocations. I don't think there is anything that will force convergence between the IOU's with respect to how baseline allocations / tiered TOU's will be handled, but I am anticipating some new structures between now and 2019 when the default plan for all customers switches from lumped volumetric tiers to TOU.

[mikenmar]

I just had my system installed, and I'm waiting to get the sign-off from PGE before net metering starts.

Unlike the OP, I'm on a E-9 TOU schedule because I have an EV. Under E-9, there's no baseline. There are tiers of course, and there are three time periods:

Peak: 2:00 p.m. to 9:00 p.m. Monday through Friday. 3:00 p.m. to 7:00 p.m. Saturday, Sunday and Holidays.

Partial-Peak: 7:00 a.m. to 2:00 p.m. and 9:00 p.m. to 11:00 p.m. Monday through Friday, except holidays.

Off-Peak: All other hours.

For any given tier, rates are obviously highest during peak, less during partial-peak, and lowest during off-peak.

According to the NEM agreement, For an OAS with Time of Use (TOU):

If the eligible customer-generator is a net consumer during any discrete TOU period, the net kWh consumed shall be billed in accordance with that same TOU period in the eligible customer-generator’s OAS.

If the eligible customer-generator is a net generator during any discrete TOU period, the net kWh produced shall be valued at the same price per kWh at the same TOU period in the eligible customer generator’s OAS.

In the event that at the end of the monthly billing cycle, an eligible customergenerator’s net usage for all TOU periods totals zero (i.e. net generation in one or more periods exactly offsets the net usage in all other periods), then the value of usage and/or generation will be calculated using Tier 1 rates (as set forth in the OAS).

My system should generate about 95% of my 600 kwh monthly usage. I will be exporting a substantial amount of power during peak and partial-peak hours, because I'll be at work during the day M-F, and my usage is low in that time period.

If I understand properly, that will work to my advantage. According to the above, it appears net usage/production is calculated for each time period individually, so I should get credited the highest rates for the excess production during peak and partial-peak hours -- is that correct?

[mikenmar]

In case anyone else finds it useful, here's a chart I made showing the summer rates for the E-9A TOU schedule (the kwh prices for the four tiers are at bottom):



I put a copy of this on the fridge for the wife's convenience.