Installing E, S, W panels vs all South

[foo1bar]

Poor, or less than most optimal orientations for the sake of getting some required production without considering the (negative) impacts on cost effectiveness that those less than optimal orientations can have, particularly when better or at least more cost effective orientations are available, seems contrary to the goal of most bang for the buck.

The financial aspect definitely has to be evaluated on a case-by-case basis.
I would not make the goal to be "most bang for the buck" - I would make it "best financial result in 5/10/15 year time frame"

And an E/W system can be something that gives that 'best result'.
It is easy to see a scenario where a system that has 5.3kW facing W, and 5.3kW facing E with a 7.6kw inverter is going to be a better financial choice than the other two options of:
A> 7.6kW system facing S with a 7.6kW inverter (fewer kwh, so lower initial cost BUT higher per-month cost for continued POCO power)
or
B> 10kW system with a 10kW facing S. (equal or more kwh - so same POCO per-month charges, but can be significantly more expensive to go up to 10kW inverter instead of 7.6kW)

Or where limited roof space makes E/S/W orientations a better choice than having higher costing (but more efficient panels) on just the S roof.


I think "bang for the buck" implies cheapest cost per kwh produced. Which isn't the right goal IMO.
If I have a choice between a 8000kwh/year system at $.03/kwh and a 12000kwh/year system at $.04/kwh, I'll take the larger less "bang for the buck" one - as the cheaper one means I'll pay $.11/kwh to the POCO for a lot more kwh. The larger one will meet much more of my needs and be much better for me financially over a 5/10/20 year period.

[J.P.M.]

The financial aspect definitely has to be evaluated on a case-by-case basis.
I would not make the goal to be "most bang for the buck" - I would make it "best financial result in 5/10/15 year time frame"

And an E/W system can be something that gives that 'best result'.
It is easy to see a scenario where a system that has 5.3kW facing W, and 5.3kW facing E with a 7.6kw inverter is going to be a better financial choice than the other two options of:
A> 7.6kW system facing S with a 7.6kW inverter (fewer kwh, so lower initial cost BUT higher per-month cost for continued POCO power)
or
B> 10kW system with a 10kW facing S. (equal or more kwh - so same POCO per-month charges, but can be significantly more expensive to go up to 10kW inverter instead of 7.6kW)

Or where limited roof space makes E/S/W orientations a better choice than having higher costing (but more efficient panels) on just the S roof.


I think "bang for the buck" implies cheapest cost per kwh produced. Which isn't the right goal IMO.
If I have a choice between a 8000kwh/year system at $.03/kwh and a 12000kwh/year system at $.04/kwh, I'll take the larger less "bang for the buck" one - as the cheaper one means I'll pay $.11/kwh to the POCO for a lot more kwh. The larger one will meet much more of my needs and be much better for me financially over a 5/10/20 year period.

I agree that every situation is different and stated that in my first post to this thread and more than a few times over my last 10,000 + posts to this forum. I'd respectfully suggest "best financial result..." or "most bang for the buck" being more nomenclature different than definitionally different with respect to what we're each saying.

As you well know, Solar processes, including residential PV, are characterized by relatively high initial cost compared to conventional methods, but have relatively low operating costs - including low or close to zero costs for the electricity produced - which is not so for the alternate POCO power purchased. Because of the difference in the ways the costs are accrued - large initial cost with possible debt service for the PV vs. low or zero initial cost but ongoing and probably increasing payments to a POCO for grid power.

The way I leaned it, the objective of any economic analysis dealing with alternate energy can be viewed as the determination how to achieve the least cost method of meeting the defined/determined design energy need, considering both solar and non solar (usually POCO power) alternatives. For PV (and any solar project) the economic task is to determine the size and other system parameters such as orientation and equipment selection, etc. of both the PV and conventional equipment that gives the lowest life cycle cost combination of solar and conventional energy.

I've always used the definition that most bang for the buck meant most long term cost effectiveness as described above as part of getting the most fit for purpose solution to an identified and defined need using, among other tools, the methods of engineering economics, cost analysis, process economics, life cycle costing and common sense. Such methods take into account the time value of money and allow as detailed considerations as necessary of the complete range of costs to any desired and/or required level of complexity.

That's a lot different than simple first cost/lowest quote per STC Watt. I've always taken the long(er) view approach to cost effectiveness that is specifically NOT lowest first cost or lowest initial price - that first/lowest initial price costing method usually being a formula for not getting much of anything except screwed when acquisitions involve durable goods such as PV systems.

If one big project goal (of perhaps several goals) for a residential PV project is to provide some portion (all ?) of an ongoing/future electrical load in combination with conventional methods to meet a load in a way that results in the lowest overall long term cost, that usually involves an array that is sized and oriented to meet it's portion of the demand safely, reliably, and at a cost that's within project goals.

With respect to the OP's situation as to orientation and array splitting: If an array is oriented in such a way that it's long term cost effectiveness is less than some other equally possible orientation would allow, then that initial orientation will simply be less cost effective. If there is a choice of several orientations of perhaps somewhat limited size, then the most cost effective orientation should usually get the maximum utilization possible, with the next most cost effective orientation after that and so on.

For most, but not all design situations, if one goal is to produce the lowest cost electricity via PV, the idea that spreading out array orientations and so flattening out production throughout the day will be more cost effective is simply not correct. There may be climatic (example, commonly very rainy middays) or site related constraints (south facing blockage by buildings, etc.) that may dictate splitting an array. Or there may simply be no optimally oriented area available. In such (probably few) cases of poor orientation or severe shading, PV may not be an economically viable option at all. But, if solar access is good and several "good" orientations and array sizes are reasonably attainable, one orientation among all orientations that produces the most long term cost effectiveness will be the better (first) choice than several orientations where all but one have lower cost effective potential.

When the goal is to maximize annual bill offset under most current rate tariffs/billing schedules I know of, unless there exists a T.O.U. schedule somewhere that has the pricing periods cycling up and down more than once over a 24 hour period (possible I suppose), and if I believe what models like PVWatts and others tell me about annual output, and taking the unavoidable annual variation in annual system output due to weather into account, the reality is, one orientation among all available orientations will always have the greatest long term cost effectiveness (most long term cost effective mix of conventional and PV supplied power) for the design requirements, and assumptions chosen for the cost analysis, and for the equipment chosen than will the other orientations.

If all that is a reasonably true and correct representation of a design situation, then using more than one orientation for an array will be less cost effective, as a little thought will demonstrate. A larger system will be required for arrays with multiple orientations to produce the same annual output as a smaller, single orientation system that is optimally oriented, or better oriented than the other possible orientations. Well, larger systems usually cost more than smaller systems. More money for the same output == less cost effectiveness. That's just the way it is. That most optimal orientation should be the first one used to the greatest degree possible before other orientations.

Multiple orientations may be necessary for an application due to physical site constraints, personal choice, shading, bureaucratic or other reasons, but if you think about it, if one orientation is more cost effective, that means that all the others are less cost effective. That means using other orientations when it's not necessary (or even if it is necessary for that matter) will result in an array of multiple orientations being less cost effective. If multiple orientations are necessary for whatever reason, that's just a fact of life that may impact the design to the extent cost effectiveness is balanced against the other design goals, criteria and priorities, but to meet the same duty, the system with unneeded multiple orientations will be less cost effective than the system with a single orientation.

[foo1bar]

For most, but not all design situations, if one goal is to produce the lowest cost electricity via PV, the idea that spreading out array orientations and so flattening out production throughout the day will be more cost effective is simply not correct.

We aren't talking about most situations though.
I would agree that in most situations (ie. at least 50% of them) the best choice is a single array facing one direction.
But we aren't talking about most situations. We're talking about a specific situation and similar situations. The fact is that spreading out the production over a wider time period and having a smaller peak production IS a better solution financially for MANY situations.
It can mean that you have to spend more on modules but less on an inverter and less on the A/C wiring (which at certain points it can have a huge step function in cost when increasing the inverter size)
Whether it is a better choice for the OP's situation is something they'll need to spend more time evaluating. But IMO you are being quite dismissive of the idea when it actually makes financial sense for many situations (not all nor even >50% - but for many)

[DanS26]

OP......you are in OK.....do you have a net metering plan or a net billing plan?.....will make a big difference in how you evaluate alternatives.

[J.P.M.]

OP......you are in OK.....do you have a net metering plan or a net billing plan?.....will make a big difference in how you evaluate alternatives.

Either or any way, what's necessary for intelligent choices is to know what choices in rate plans are available, the consequences that usage and usage patterns can have on each one, a way to figure a bill based on usage and usage patterns for each available rate schedule, and a way to an educated SWAG on how rates and tariffs may be changing in the future, both near and long term. Knowledge is power, or at least a path to better choices.

[df0rster]

OP......you are in OK.....do you have a net metering plan or a net billing plan?.....will make a big difference in how you evaluate alternatives.

Yes we have net metering. But it is month to month.

yes In OK. On a small coop called vvec. They seem to discourage net metering but at least allow it. The engineer tells me they have just a few people who participate.

[df0rster]

For some reason when replying to posts on my phone it cuts off after just a few words..

[df0rster]

Either or any way, what's necessary for intelligent choices is to know what choices in rate plans are available, the consequences that usage and usage patterns can have on each one, a way to figure a bill based on usage and usage patterns for each available rate schedule, and a way to an educated SWAG on how rates and tariffs may be changing in the future, both near and long term. Knowledge is power, or at least a path to better choices.

We dont get to choose rate plans.

[J.P.M.]

We dont get to choose rate plans.

Good. Then your analysis task will be easier - you'll only need to look at one tariff schedule. Just be prepared for some involvement to get it all right. Best way I found is to generate a bill using your current usage and the tariff and dig around/correct/iterate until your result matches what the POCO bill says. Do that for a few months until what you calc matches the bill.

[DrLumen]

When I was researching there were situations that southwest facing was actually better than south due to weather. LIke one place was in the mountains that was almost always covered in clouds in the morning but the weather cleared later in the day. This may not apply and may be something that was considered. Just to throw it out there as a reminder

[DanS26]

Since you are on a net metering plan with one rate structure, month to month, with no carry over, then I think you are on the right track to try to maximize production in the lowest production months. Probably Dec thru Feb.

But you have to build a system that stays within the maximum size allowed by your coop and by the constraints of your electrical service on the highest production months. Probably Jun thru Aug.

Good luck....sounds like a fun planning exercise....at least to me.

[J.P.M.]

When I was researching there were situations that southwest facing was actually better than south due to weather. LIke one place was in the mountains that was almost always covered in clouds in the morning but the weather cleared later in the day. This may not apply and may be something that was considered. Just to throw it out there as a reminder

That the weather patterns cause the optimal array orientation to be different for different locations is well known and well documented.

That's one big reason why models of solar systems exist in the first place.

A lot of the time such modeling confirms what common sense, and having one eyeball and one balloon knot and a wet finger in the wind and a little critical thinking has already pointed to. The models help zero in on an orientation, allow estimates of consequences of other orientation choices, and provide confirmation/refutation of the hunches - some of those initial hunches made less than accurate by ignorance.

If the model used is a decent approximation of reality, and the user understands how to use it, its limitations, and how to interpret the results, optimal orientation, whatever that means for a particular application and user/owner's goals, can be estimated with a reasonable degree of confidence.

As an example, around a lot of so. CA that's, say, within 10 - 20 miles of the coast, morning fog is very common which is why, all other things being equal, and if used against the new T.O.U. non grandfathered rates, the optimal azimuth around here for maximizing annual bill offset is ~ 10 - 30 deg. west of south. Best tilt is ~ 25 - 30 deg. or so.

By way of contrast, when I lived back east (Buffalo), in the summer a mid/late afternoon shower was a common event (like much of FL). In the winter, when the sun did make a rare appearance, it was usually until late morning and then lake effect clouds and snow would roll in. Based on those weather patterns and POCO electricity rates, optimal PV orientation at that location would have been ~ 10 deg. east of south with a tilt of ~ 45 deg. However, because while economics are a big consideration, they are not the only considerations. Other realities like the PITA factor of snow removal and the fact that A/C is not much of, if any, consideration in that location, and because only nitwits heat with electricity in such environments (nat. gas is nearly ubiquitous), and other reasons, the best/most practical tilt would probably have been ~ 60 deg. or more, or all those things would conspire to make PV a less than viable option and probably eliminated PV as an option altogether, at least for me.

As it turns out, depending on how picky one wants to get, and also depending on other variables besides the weather - like equipment particulars, shading, aesthetics, etc,, not to mention the practical limitation that most roofs are fixed, as well as user/owner choices/preferences, most of the time the chosen and to a large degree limited orientation choices for an application in any location are not strictly south facing.

Fortunately, for most applications and equipment choices, the reality is that provided an orientation is not way off optimum, mother nature is fairly forgiving of orientation. Most of the PV game is far from rocket science. Usually, but not always, ~ +/- 20 deg. of optimal azimuth and ~ +/- 5 to 10 degrees of optimal tilt will wind up with ~ 90 % or better of the production and/or the bill offset of an optimally oriented array.

[J.P.M.]

Since you are on a net metering plan with one rate structure, month to month, with no carry over, then I think you are on the right track to try to maximize production in the lowest production months. Probably Dec thru Feb.

But you have to build a system that stays within the maximum size allowed by your coop and by the constraints of your electrical service on the highest production months. Probably Jun thru Aug.

Good luck....sounds like a fun planning exercise....at least to me.

If there is no month/month carryover, then orienting for the most cost effectiveness per STC kW, and subsequently sizing an array by maximizing winter production will have a higher probability of producing an excess in the summer months and over production in an annual basis.

That doesn't sound like the best way to go for a grid tied and net metered system and never a good way to start out if a cost effective system is one of the design goals.

FWIW, to me anyway, system design and sizing based on either winter or summer extremes of production doesn't sound like the best way to go, particularly if T.O.U. billing adds a time function into the considerations.

I agree that it can be a fun exercise, and not all that complicated - just involved.

The involvement starts with getting an estimate of the most bill offset over a total of 12 months of billing periods as f(orientation) per installed STC kW - and if on T.O.U. (OP - ??? - T.O.U. or tiered, or other ??) for possible orientations.

To get the 12 month max. if there is no carryover, get the production and bill offset for each billing period. Then, sum all the results for the billing periods. Orient and size a system that gives the most economic benefit for the sum of the billing periods (12 months) using any economic figures of merit as the OP sees fit.

I'd respectfully suggest looking at the LCOE (Levelized Cost of Electricity) method of life cycle costing for this as using LCOE makes it easier to see the often large negative economic effects of oversizing.

Then - and here's the part that's usually left out by solar peddlers and tree huggers - and maybe drifting off topic a bit - see if that orientation and size, and indeed choice to get PV at all, makes economic sense compared to other alternatives or other ways to invest the money/assets spent on PV acquisition that would produce a return on investment that may be greater than the long term value of bill reduction brought about by the PV acquisition (including BTW, a choice to not do PV at all - always a choice and in spite of the B.S. from peddlers with skin in the game, often offering more economic benefit than getting PV in the first place).

Then, making the choice of where to put the assets based on such a comparison. I suspect that considering economics alone and separate from all other considerations for a moment, most folks would drift toward the alternate investment choice that produces the highest probability of the greatest returns with tolerable risk while hoping their assumptions and guesses about the future aren't too far off reality. Based of assumptions of future conditions, what's seen as that best economic choice may or may not be PV.

I appreciate that economics alone are but one consideration, but it's also and often the biggest and sometimes the only criteria considered, sometimes to the exclusion of the others. Either or any way, better information leads to more informed choices.

Include putting some economic value on the (usually small but nonzero) PITA/risk/benefits of PV in the economic analysis, as well as the possible risks from the alternate investments.

Then, compare all the economic benefits of the economically optimum PV system vs. all the other choices and possible places to put assets to produce a return, and make a choice of where to put the assets - into PV or into some other income producing or bill offsetting investment, or some mix of alternates - smaller PV + some other investment(s) for example.

Include some risk analysis for each possible alternative investment, and don't forget that investing in PV is not devoid of risk either - as most all the people who leased PV in So. CA are finding out now that the new T.O.U. billing tariffs and schedules have made PV about 20-25 % less cost effective than when they signed their lease - but their lease payments which were based on more favorable net metering POCO billing are not only not lower but still going up about 3 % per year.

Anyway, at the end of the analysis, take your best shot knowing that while there are no guarantees in life, informed choices have a greater probability of more favorable outcomes, and that fortune favors the bold and informed, but usually slaughters the rash, the emotional and the ignorant.

As an aside, and perhaps to put some context to all this, the usual goal is to reduce the long term costs of supplying energy, usually electricity, to a home or business - not getting PV or other solar for it's own sake - PV is a means to an end, and usually the most expensive one.

With that in mind, know that as a way to reduce an electric bill, getting turnkey PV installed on a residence is often and usually the least cost effective way to reduce an electric bill compared to all the other ways. Thus, if electric bill reduction in the most cost effective way is the chosen path, use reduction, conservation and finally, and lastly PV, is always done up to the point where the long term cost of the most cost effective measures (starting with simply turning stuff off where cost== $0) of bill reduction just equals the long term cost of providing the last kWh of electricity - which point, at least most of the time, with a good use reduction/conservation plan, is reached long before PV becomes an economically viable option.

Take what you want of the above. Scrap the rest.

[DanS26]

Over producing in the summer for which you do not benefit (but the coop and your neighbors do) is a sacrifice to absorb so that your electric bill is zero or close to zero in the winter. The correct answer from an economic standpoint I suspect is somewhere in between. A zero bill in April and October may be the sweet spots.

A good analysis will tell how much that sacrifice will cost over the life of the system. The further south (ie OK) then the less the sacrifice.

[J.P.M.]

Over producing in the summer for which you do not benefit (but the coop and your neighbors do) is a sacrifice to absorb so that your electric bill is zero or close to zero in the winter. The correct answer from an economic standpoint I suspect is somewhere in between. A zero bill in April and October may be the sweet spots.

A good analysis will tell how much that sacrifice will cost over the life of the system. The further south (ie OK) then the less the sacrifice.

A good analysis will provide a reasonably reliable number for what's going on in terms of actual vs. expected sysytem performance, be that performance good, or bad, or anything in between, including some idea of expected economic return regardless of what that performance or what those goals may be.

A good system design is one that uses a good and sensible design approach that will produce a design that will meet the design goals, one goal of which is probably maximizing economic return as a priority, but also with considerations for and balance with other system design goals and practical limitations such as actual available orientations and areas as the OP seems to have.

Due to the balance of various design priorities, the design with the max. possible economic return is usually not the one chosen. The economic sweet spot may well then need to be and de facto is often redefined as the one that results after the design is finalized considering all the prorities taken together. Life is full of compromises.

Over production at any time when there willl be no or much reduced net metering offset or system produced revenue to offset an electric bill will kill system profitability in very short order and ought to be avoided to the greatest extent possible as a consideration from preliminary design stages unless and until very careful consideration is given to the consequences and impact on cost effectiveness to the extent that that cost effectiveness is a priority.

From a design standpoint, it's pretty common for a well designed system to over and underproduce at some times of the year just like vehicle gas mileage changes going up and down the same hills on the same trip taken every day. It usually cannot be avoided.

FWIW, from my experience, and because cost effectiveness is usually a high priority, but mostly because the usually paltry payout for overproduction with respect to most net metering schemes has to kill cost effectiveness, one way of several I usually look at the economic part of a design is to start with a system size that does not overproduce over any, say, 30 day running consecutive period with only the limitations of orientation and area in mind. Then, I run the system size up and down with the other system goals and constraints in mind. That approach, after a few iterations, tends to mimimise overproduction and also zero in on what's usually close to the max. reasonably acheiveable cost effective system size. The system cost effectiveness in terms of LCOE usually, but not always drops with the LCOE usually increasing as a result of consideration of other design priorities and goals.

More FWIW, and somewhat parocially, and to my experience only, for SDG & E billing and tariffs in place at this time, the best long term cost effectiveness (lowest LCOE) for size seems to be somewhere betrween ~ 60% to ~ 80% or so offset of an existing annual bill, with generally the lower the bill, the lower the most cost effective offset % tends to be. Before the big tariff changes of the last year or so, those percentages were about 10 % higher. So, folks who go to 100 % offset with the idea of "getting even" with the POCO, or having no or a minimum bill often wind up shooting themselves in the foot if maximizing return on investment is a goal.

Around here, and I suspect a lot of other places, a good design may well not include a need to obtain an electric till that is zero or close to it to produce the greatest long term ecomomic benefit.

Bottom line: IMO only, the best economic "answer" and goal of the excercise is usually to pay as little as necessary over the selected term, usually years, to provide a desired level of electrical service to a residence, using the most cost effective mix of use reduction, conservation measures, conventional methods such as purchased POCO power, and alternate energy sources as the owner or the party paying the bills may see it.

All the vendor (and their mostly well meaning but alternate energy ignorant treehugger shill) induced rush to PV as a magic bullet solution to self inflicted high electric bills only sidetracks that economic goal.