Estimates of system revenue using old vs. new T.O.U times and rates.

[J.P.M.]

I've done a rate comparison for how the proposed (likely ??) changes to SDG & E's schedule DR-SES (= T.O.U. schedule for residential customers with a PV system) might impact system revenue.

The system revenue for each version of DR-SES under the assumptions used will be different regardless of actual usage, provided the home's usage exceeds the system's production.

I do not claim this estimate is applicable to any ratepayers other than SDG& E customers who are now on, or may be going to schedule DR-SES. The results are deemed accurate, but ther's no one checking my work so readers are cautioned to confirm. I'll welcome any comments and ask any errors be brought to my attention.

The estimates and methods used here apply only to DR-SES. The methodology used here will not work for other T.O.U. schedules such as those that have the old tiered system laid over them such as DR-TOU.

The revised rates and times (for the "new" DR-SES) I used for these estimates are those that will be in effect for those customers who do not elect this schedule before 07/28/2017, and therefore will miss the cut for the 5 yr. grandfathering on the current (old) DR-SES schedule.

Specifics I used in compiling the revenue estimates:

1.) The new rates and times are those as published by SDG & E. I do not believe they have been officially approved by the CPUC, but approval seems likely. The old DR-SES rates and times used are those in effect as of 03/01/2017. Note that the peak rates for the new schedule are lower than the old rates, and that the off peak rates for the new schedule are higher than the off peak rates for the old schedule. While this may seem unbelievable to some, it seems to follow SDG & E's statements of their intent to follow AB 327 's mandates or desires to do just that. Also, keep in mind that the rates are subject to change, but they seem to be th best available at this time.
2.) Summer schedule season is 5 months long for the new schedule. Winter season is therefore 7 months. The seasons per the old schedule were each 6 months long.
3.) Included in the analysis is use of the 1000 - 1400 hrs. super off peak rate times for weekdays in March and April. For other winter months, those times (1000-1400 hrs.) are off peak.
4.) I did not make any adjustment in the revenues for NBC's. NBC's will not change the revenue differential from old to new schedule numbers, but their impact will reduce the system revenue for both the old and new schedules each by an approximately equal amount, depending on how much of a system's production actually goes back to the grid. Power generated and used on site, that is, not sent to the grid, is not subject to NBC reductions. I've added a comment on estimates of min./max. impact of how usage patterns may impact system revenue for old and new schedules.
5.) The estimated numbers assume no overgeneration. Overgeneration will probably not impact the revenue differential old to new DR-SES schedules much, if at all. My guess is any impact on the differential will be slight.

With those assumptions, I used PVWatts to model a 1 kW system using TMY3 data for Miramar MCAS. The system orientation is due south, 20 deg. tilt., "standard" panels (using "premium" panels increased annual production by ~ 2%). The system was modeled as a roof mount. The system loss parameter used was 10 %.

Results:

System revenue using the current (old) schedule DR-SES : $515.60 per year per installed STC kW. The raw (non NPV) per kWh value is $515.60/1,720kWh = $0.3000/kWh.
System revenue using the new (likely) schedule DR-SES : $401.34 per year per installed STC kW. The raw (non NPV) per kWh value is $401.34/1,720 kWh = $0.2333/kWh.

The difference = $515.60 - $401.34 = $114.26 per year, or a $114.26/$515.60 = 0.2216 = 22.16 % reduction in annual system revenue for the same system operating in the same orientation and the same location.

There will likely be some revenue reduction from NBC's. The greatest impact NBC's will have on the revenue generation under either the old or the new schedule is $30.01. That is, if all the generation is subject to NBC's the revenue under the old schedule will be $485.59, and the revenue under the new schedule will be $371.33.

For a system where the above assumptions are applicable, that is, a roof mounted system where Miramar TMY3data is applicable, facing mostly south at ~ 20 deg. tilt, Multiplying the numbers for a 1 kW system as given above by the actual system size in STC kW will give a rough estimate of that system's revenue.

To the degree my numbers represent some version of reality, the impact of old vs. new DR-SES rates and times look to have the effect, at least to a rough 1st approx., of extending payback times by ~ 20 - 25 %, or reducing ROI's by about the same amount.

This is a bit of a PITA method, but within the limitations cited above, and as those limitations may be applicable, couple of hours of learning how the rates work and a couple of hours with an ~ 8,800 row spreadsheet will allow most anyone to do the same thing.

[sensij]

Sounds like the right approach for determining the NEM Credit side of the equation. Determining the change in NEM Charges will also be important to evaluating the full effect of the shift in TOU period definition. When I get the next iteration of my spreadsheet updated, I'll post it here. Thanks for starting this thread and posting your results!

[J.P.M.]

Sounds like the right approach for determining the NEM Credit side of the equation. Determining the change in NEM Charges will also be important to evaluating the full effect of the shift in TOU period definition. When I get the next iteration of my spreadsheet updated, I'll post it here. Thanks for starting this thread and posting your results!

You're welcome. Look forward to it. The adventure continues. We're probably the only two souls interested in this stuff.

I do believe treating revenue separate from charges for analysis has definite advantages, at least at the early stages. Not mixing the two allows somewhat of a different perspective on some things that's not available when only looking at the net effects. Example: Average cost/kWh for use and average revenue per kWh for generation can be useful tools for economic optimization.

[sensij]

Ok, i've updated my spreadsheet. Here are the inputs:

Consumption, as described in this post, with net annual consumption of 877 kWh.
DWR-BC: 0.00549
UDC: 0.14184 / kWh, year-round
EEEC:
Summer peak: 0.35896
Summer off-peak: 0.10375
Summer super off-peak: 0.07988
Winter peak: 0.10000
Winter off-peak: 0.09000
Winter super off-peak: 0.08000


5 months summer, new hours including the march/april exception.

Obviously, the winter EEEC prices are just a guess, I'm happy to substitute in whatever numbers might have more support there.

Under the current DR-SES tariff and hours, NEM credits were 606.74, NEM charges were 544.53, giving a -62 total annual bill.
Under the new TOU period definitions considered here, NEM credits were 239.76, NEM charges were 504.78, giving a 265 total annual bill..

Again, for reference, this usage laid onto the current DR tariff would result in a $170 annual bill, and without solar at all, on the DR tariff, I had estimated something around $1550 (I accidentally deleted the sheet where I had determined PV production and backed it out, so that needs to be re-created). When I get that sheet re-built, I can more directly compare revenue numbers in the framework used in this thread, but ballpark for this 3.1 kW system was ~$1600 revenue before, agreeing with the $515 / kW estimate, and about $414 / kW in the new TOU hours, probably only different from the $401 estimate because of slightly different rate assumptions.

In other words, my findings basically agree with the idea that the TOU shift means 20% revenue reduction equating to 20% longer payback times, etc. The NBC discussion still has some ground to cover before it is implementable in a sheet for me. For someone who had projected 5 year payback, adding a 6th year might not be a big deal. For someone already at 8 year payback, this change would push it close to 10 years, perhaps a dealbreaker.

[Mike7381]

Ok, i've updated my spreadsheet. Here are the inputs:

Consumption, as described in this post, with net annual consumption of 877 kWh.
DWR-BC: 0.00549
UDC: 0.14184 / kWh, year-round
EEEC:
Summer peak: 0.35896
Summer off-peak: 0.10375
Summer super off-peak: 0.07988
Winter peak: 0.10000
Winter off-peak: 0.09000
Winter super off-peak: 0.08000


5 months summer, new hours including the march/april exception.

Obviously, the winter EEEC prices are just a guess, I'm happy to substitute in whatever numbers might have more support there.

Under the current DR-SES tariff and hours, NEM credits were 606.74, NEM charges were 544.53, giving a -62 total annual bill.
Under the new TOU period definitions considered here, NEM credits were 239.76, NEM charges were 504.78, giving a 265 total annual bill..

Again, for reference, this usage laid onto the current DR tariff would result in a $170 annual bill, and without solar at all, on the DR tariff, I had estimated something around $1550 (I accidentally deleted the sheet where I had determined PV production and backed it out, so that needs to be re-created). When I get that sheet re-built, I can more directly compare revenue numbers in the framework used in this thread, but ballpark for this 3.1 kW system was ~$1600 revenue before, agreeing with the $515 / kW estimate, and about $414 / kW in the new TOU hours, probably only different from the $401 estimate because of slightly different rate assumptions.

In other words, my findings basically agree with the idea that the TOU shift means 20% revenue reduction equating to 20% longer payback times, etc. The NBC discussion still has some ground to cover before it is implementable in a sheet for me. For someone who had projected 5 year payback, adding a 6th year might not be a big deal. For someone already at 8 year payback, this change would push it close to 10 years, perhaps a dealbreaker.

Just a side question, when you calculate pay back time, do you include things like roof shingle that need to be replace because of solar and/or 200A panel upgrade? I think 0200A Panel upgrade should be included when calculate it because you don't really need it if not for solar. For Shringle, I don't think it is fair to include it because you need it rather or not solar is installed.

[ButchDeal]

Just a side question, when you calculate pay back time, do you include things like roof shingle that need to be replace because of solar and/or 200A panel upgrade? I think 0200A Panel upgrade should be included when calculate it because you don't really need it if not for solar. For Shringle, I don't think it is fair to include it because you need it rather or not solar is installed.

few installs NEED a 200A MSP upgrade. Ones with smaller inverters don't and others often have other cheaper solutions than MSP upgrade.
But yes either way normally all electrical work required for the install is included in the cost of the system.

The initial cost of the system has little to do with the discussion in this thread though as they are looking at ongoing revenue differences with the different utility plan options.

[J.P.M.]

The initial cost of the system has little to do with the discussion in this thread though as they are looking at ongoing revenue differences with the different utility plan options.

Actually, one of several motivating reasons behind why I started looking at the impact of T.O.U. time and rate changes was to see not only how, and in what ways T.O.U. time and other changes would impact a bill, but equally and almost of central importance, is how those changes would impact the initial price requirements to maintain a required cost effectiveness, however a potential user might calculate such cost effectiveness requirements.

For the same required cost effectiveness, if the system's annual $$ return decreases, and if the same cost effectiveness is to be maintained, the system's initial cost will need to decrease.

For a simple analysis, and maybe a 1st approx., I'd respectfully suggest that to maintain (required ?) cost effectiveness, the same % of system revenue reduction will need to be applied to system initial cost. So, in that sense, since most ways to calculate cost effectiveness usually have the aim of getting most bang for the buck, or at least not spending more up front (or in financing a system) than necessary to achieve the desired effect, savings differentials will impact initial cost considerations.

So, while the immediate discussion is about revenue changes, the larger goal is as a tool to help potential PV users by showing what can happen to PV savings under T.O.U. tariff changes taking place now and in the future. Those changes can and most certainly will impact how much a PV system is worth, which is, I'd guess, one upper limit to initial cost.

Butch, I appreciate where you may be coming from, being a vendor or someone who makes a living off PV and all. FWIW, and IMO only, what you do here is a real help to folks. But I also appreciate that my stirring the pot about how things beyond vendor's control that will probably lower system cost effectiveness and thus prices can be a PITA for you, but that's just how I see it. I'd be interested to read any thoughts you may be willing to share about ways lower system revenue might negatively impact (that is lower) what potential customers might be willing to pay for PV, particularly in a residential situation.

To Mike's question about payback time, for purposes of this thread, and to keep it as simple and applicable to as many users as possible, the $$ discussions refer to direct system costs separate from roof repair or other things not directly related to system costs. I'd suggest some flexibility is necessary however. For example, a new service panel if required ?- probably. Update other house wiring while you're at it ? - probably not.

[ButchDeal]

OK yes I see where you are coming from J.P.M. but Mike7381 was asking about specifics which would increase the initial cost independent of the T.O.U. changes like MSP upgrade to 200A. I posit that an MSP upgrade is not needed for solar. We tend to do them infrequently and do line side taps much more often. The MSP upgrade has many other benefits though and these other benefits (more expansion for home use, often safer than the old MSP, etc) are the main reasons that we do the MSP upgrades that we do sell. Thus an MSP upgrade is not 100% for solar.
Regardless though to meet your point about the cost, if a line side tap or MSP upgrade is done there is added cost. For us we would consider this an adder and we do them at cost. We make our money in the base price not on adders. But along your thinking of reducing cost the line side tap method is cheaper (at least our rates have it average about $1k cheaper).

As for keeping costs down, as I often state on this forum our company also advises the lowest cost modules that we offer if it is sufficient to provide the power required. Unless customers specifically request some type of module like BoB or US made etc. We currently have consumption meters as add ons but we have been considering adding them to the basic systems. The internal debate is mostly around keeping costs down.
We work with installers to negotiate lower prices all regularly but we are right about where most other national companies are with the rates for installers. Were we come a bit high is in states with little solar installs that the big players don't touch and very remote locations (installers don't like driving for 5 hours each way into the desert for some reason).

Short of using lower grade equipment, inverters and racking, it is hard to keep costs down. Since we back the installer and manufacturers warranties and include a production warranty cutting the equipment seems like a poor idea. We had a nearly 50% failure rate with another well known inverter brand that has bit us in the a$$ to the point that we even changed one of them out to SolarEdge.

[J.P.M.]

Ok, i've updated my spreadsheet. Here are the inputs:

Consumption, as described in this post, with net annual consumption of 877 kWh.
DWR-BC: 0.00549
UDC: 0.14184 / kWh, year-round
EEEC:
Summer peak: 0.35896
Summer off-peak: 0.10375
Summer super off-peak: 0.07988
Winter peak: 0.10000
Winter off-peak: 0.09000
Winter super off-peak: 0.08000


5 months summer, new hours including the march/april exception.

Obviously, the winter EEEC prices are just a guess, I'm happy to substitute in whatever numbers might have more support there.

Under the current DR-SES tariff and hours, NEM credits were 606.74, NEM charges were 544.53, giving a -62 total annual bill.
Under the new TOU period definitions considered here, NEM credits were 239.76, NEM charges were 504.78, giving a 265 total annual bill..

Again, for reference, this usage laid onto the current DR tariff would result in a $170 annual bill, and without solar at all, on the DR tariff, I had estimated something around $1550 (I accidentally deleted the sheet where I had determined PV production and backed it out, so that needs to be re-created). When I get that sheet re-built, I can more directly compare revenue numbers in the framework used in this thread, but ballpark for this 3.1 kW system was ~$1600 revenue before, agreeing with the $515 / kW estimate, and about $414 / kW in the new TOU hours, probably only different from the $401 estimate because of slightly different rate assumptions.

In other words, my findings basically agree with the idea that the TOU shift means 20% revenue reduction equating to 20% longer payback times, etc. The NBC discussion still has some ground to cover before it is implementable in a sheet for me. For someone who had projected 5 year payback, adding a 6th year might not be a big deal. For someone already at 8 year payback, this change would push it close to 10 years, perhaps a dealbreaker.

Thank you.

Lots to think/talk/comment about.

On NBC: the CALSEIA blurb relating to NBC has a method, but no clear chap./verse example of how NBC works to increase a bill under NEM 2.0. Their bottom line: Multiply the energy sent to the grid per time each interval (15 minutes for most purposes) by the NBC/kWh $$ charge, and sum that product for all the time intervals in a billing period. The CALSEIA method also has some explanation as to why this NBC business is a bit obscure and confusing. They do it for commercial users but state that the method will also work for residential users. I'm still on the hunt for more info/background. Film at 11 or when bulletins break.

On how to handle the impact that T.O.U. tariff changes will bring: In a more global sense, I bet this part will get more complicated. To me tat seems to translate to different methods, or different ways to look at financial impact with a different method for T.O.U. tariffs that do not have tiered rates laid over them (such as SDG & E's DR-SES) from those that do.

Because T.O.U. schedules that do not have a tier structure embedded in them are not impacted by changes in usage except for cases of overgeneration (which, if overgeneration reimbursement rates will be < billing rates - which it always is - will negatively impact cost effectiveness), or for considerations of NBC impacts which, while probably of a more minor nature, will still (at least as it appears at this time anyway) need consumption vs. generation as f(billing time increment) to figure out. Also, The $120/yr. annual min. also needs to be accounted for.

But, those considerations are still less complicated than what happens when a tiered rate structure is laid over a T.O.U. rate structure by adding some credit scheme to a T.O.U. schedule for usage below 130 % of monthly tier 1 usage for some T.O.U. pricing times. IMO, that'll take some spreadsheet shenanigans I'm not looking forward to. I've got tiered rate calc methods pretty much dialed in. Putting the two together without bizarre things like imaginary income may be a chore. We'll see. For now, I'm happy DR - SES is a straight T.O.U. affair, making the revenue method possible to get a figure of merit for DR-SES.

[J.P.M.]

OK yes I see where you are coming from J.P.M. but Mike7381 was asking about specifics which would increase the initial cost independent of the T.O.U. changes like MSP upgrade to 200A. I posit that an MSP upgrade is not needed for solar. We tend to do them infrequently and do line side taps much more often. The MSP upgrade has many other benefits though and these other benefits (more expansion for home use, often safer than the old MSP, etc) are the main reasons that we do the MSP upgrades that we do sell. Thus an MSP upgrade is not 100% for solar.
Regardless though to meet your point about the cost, if a line side tap or MSP upgrade is done there is added cost. For us we would consider this an adder and we do them at cost. We make our money in the base price not on adders. But along your thinking of reducing cost the line side tap method is cheaper (at least our rates have it average about $1k cheaper).

As for keeping costs down, as I often state on this forum our company also advises the lowest cost modules that we offer if it is sufficient to provide the power required. Unless customers specifically request some type of module like BoB or US made etc. We currently have consumption meters as add ons but we have been considering adding them to the basic systems. The internal debate is mostly around keeping costs down.
We work with installers to negotiate lower prices all regularly but we are right about where most other national companies are with the rates for installers. Were we come a bit high is in states with little solar installs that the big players don't touch and very remote locations (installers don't like driving for 5 hours each way into the desert for some reason).

Short of using lower grade equipment, inverters and racking, it is hard to keep costs down. Since we back the installer and manufacturers warranties and include a production warranty cutting the equipment seems like a poor idea. We had a nearly 50% failure rate with another well known inverter brand that has bit us in the a$$ to the point that we even changed one of them out to SolarEdge.

Thank you.

Comment: If PV becomes less cost effective due to NEM (and with politics, opinions and persuasions aside here) becoming less lucrative for homeowners, and it sure looks like it's going that way, and you and others are already doing all possible to hold down costs to the point they are as low as possible now, how will you folks stay in business if/when you cannot drop prices any more, but customers see their payback times/ROI's or any other ways of measuring cost effectiveness dropping because of less bill offset due to changing NEM reimbursement rates and look for price relief from you ?

[ccdengr]

We're probably the only two souls interested in this stuff.

Well, I'm a third person and I appreciate all of your effort. I'm a NEM 1.0 customer who snuck just under the wire (started operating in June 2016), have been on DR and overproducing ($-74 bill at my first true-up), and really wasn't prepared to deal with this TOU stuff so soon. I will most likely switch to DR-SES before tomorrow's deadline but I still can't tell if this is a smart move or if I've fallen into a trap set by SDGE.

[sensij]

Well, I'm a third person and I appreciate all of your effort. I'm a NEM 1.0 customer who snuck just under the wire (started operating in June 2016), have been on DR and overproducing ($-74 bill at my first true-up), and really wasn't prepared to deal with this TOU stuff so soon. I will most likely switch to DR-SES before tomorrow's deadline but I still can't tell if this is a smart move or if I've fallen into a trap set by SDGE.

If you are already getting a credit on DR, and don't have any firm plans to change your consumption, I'd stick with what you've got.

[ccdengr]

If you are already getting a credit on DR, and don't have any firm plans to change your consumption, I'd stick with what you've got.

Really? My thinking was that on DR I am getting a credit, and the SDGE website tool says that it's a wash being on DR or the current DR-SES, so switching to DR-SES now would lock me into that until 2021, rather than being forced onto what TOU options would be available to me in 2018-19 or whenever that will really happen. I am usually a net producer in the summer 11-6 time period so DR-SES doesn't look crazy. Although I have a year's worth of consumption data, I'm too confused about how TOU works to game out what the real cost difference would be.

[J.P.M.]

Well, I'm a third person and I appreciate all of your effort. I'm a NEM 1.0 customer who snuck just under the wire (started operating in June 2016), have been on DR and overproducing ($-74 bill at my first true-up), and really wasn't prepared to deal with this TOU stuff so soon. I will most likely switch to DR-SES before tomorrow's deadline but I still can't tell if this is a smart move or if I've fallen into a trap set by SDGE.

Well, that makes 3 of us. As you wish, and it may be a wise move I'll regret not making, depending on your use pattern and your flexibility on that use. For NEM 1.0 users (like me), tiered rates ought to (??) be around for awhile, but after conversations with several folks at SDG &amp; E, including an informal one w/ a neighbor who's a C.S. rep. w/them, they or the POCO doesn't know, or won't say how much longer tiered rates will last.

BE CAREFUL !. My understanding is that 5 yr. grandfathering for NEM 1.0, and also 2.0 is calculated as 5 yrs. from P.T.O., not 5 yrs. from when I choose to go to T.O.U. (and that being the case only if I make the move by 07/28/2017). As part of my making myself a PITA to the folks at SDG & E lately, one example I was given was to use my PTO date (10/17/2013) as the date of the start of the 5 yr. clock for me if I were to change over to T.O.U. by 07/28/2017. That would mean my end date for more favorable T.O.U. times would be 10/17/2018, which is one of the reasons I said screw it - I'm staying on tiered until they prey me off it. If you got PTO 06/2016, you ought to be good til 06/2021 if what I was told is correct and understand what I read correctly.

As for how smart a move going to T.O.U. now may or may not be, it may be for you, if you can avoid or reduce power use for the 35 hrs. per week, 26 weeks/yr. (~ = 900 hrs./yr. that comprise peak summer rate times. Reason: The rest of the year outside of those rate times (8760 - 900 ~ = 7860 hrs. yr.) is currently billed at close to tier one rates, maybe a bit more. Additionally, for some of those peak summer T.O.U. times - maybe half of them or more - a system will be generating and accumulating at the peak time rate. For such situations - if/when use can be reduced while generation is credited at peak rates, current T.O.U. rate times can make a T.O.U tariff a better deal than tiered rates for a lot, but not all users.

Two of the big, but not the only PITAs in trying to figure all this out are first, that the POCO is much less than helpful and forthcoming with clear, complete and straightforward information that's necessary to help make informed decisions and second, even with such good in formation, because of every situation being different, there is no easy way to say "do this - it works" and be right. Sometimes a plan will work, and other times it will make things worse. So, until we all, probably on our own, figure out what we think is best for us, we're stuck in a boat with one oar.

[sensij]

Really? My thinking was that on DR I am getting a credit, and the SDGE website tool says that it's a wash being on DR or the current DR-SES, so switching to DR-SES now would lock me into that until 2021, rather than being forced onto what TOU options would be available to me in 2018-19 or whenever that will really happen. I am usually a net producer in the summer 11-6 time period so DR-SES doesn't look crazy. Although I have a year's worth of consumption data, I'm too confused about how TOU works to game out what the real cost difference would be.

Yeah, I understand your logic, I'm just not so sure that the DR plan will be gone in the next 5 years. There is supposed to be an opt out even when everyone starts defaulting to TOU. I would guess that when DR is truly closing, we will see a formal 5 year sunset on it.

But, these are just guesses. If DR-SES looks good for you, go for it. You don't have to convince me of the potential benefits.. I've got one system in the EV-TOU-2 plan, and just switched another to DR-SES to lock it in because it is better than DR for the usage there.