Are your Home Solar Panels Facing in the Wrong Direction

[Jason]

We are not suggesting that you get up and change the direction of the panels on your roof. A new study by the Pecan Street Research Institute does make you think about the conventional wisdom that home solar panels should be placed on the southern facing roof to be most efficient. You may not have heard of Pecan Street before, but their study certainly had a way of garnering the headlines with experts coming down on both sides of the argument-and this from a study of just 50 homes. Some may discount the study because the sample was small and because the output over the course of the year is only a couple of percentage points different, but the major issue is in peak load from 3pm to 7pm west-facing solar panel arrays produced 49 percent more electricity during this peak demand time. Now what all of the commentators are missing of course is that most homeowners are not going to be redirecting their houses to the south or west just to capture a few more rays.

More...

[Ian S]

This is an interesting question for sure but there are other factors to consider. Not all utilities have 3-7 pm as their peak time; here in Phoenix, APS has noon to 7 pm or in a very few grandfathered cases, 9 am to 9 pm. After being on the noon to 7 pm with solar and net metering for a full calendar year cycle, I'm convinced that I would actually be WORSE off with more westerly facing solar panels even on the T.O.U. plan! This is because of a combination of factors including the date of the annual true-up, the amount paid for excess banked kWh's, the fact that weekends and holidays are all off-peak, the fact that on and off peak credits are separate (and can never be combined) and the particular energy usage schedule of a particular homeowner. The reason I know I'd be worse off with a more westerly orientation is because with the true-up in December and net metering, my system offsets all of my on-peak usage with the bank drawn down to its lowest ebb in September after which the credits begin to accumulate until the true-up. With a more western orientation, I'd get even more excess on-peak credits at year's end at the expense of off-peak production during the year. The year end credits are at the $0.029 per kWh, far below the cost of either on or off-peak electricity.

I guess that's an advantage of my flat roof: different orientations are simpler to achieve. But for me, I the 180 deg azimuth is clearly best.

[Volusiano]

I agree that it may work for some but not others.

It works for me because I only have a choice of east or west facing roof top. I've been on the TOU plan which has on-peak from 1-8pm in the summer time (and 5-9am and 5-9pm in the winter time). So I put all my panels on the west facing side so I can maximize the value of my on-peak rate, which is about 3x the cost of the off-peak rate in the summer. Looks like this article validates my decision. Putting the panels on the east facing side would have generated a little bit more energy in my case just because the panels will run cooler in the morning. But my focus is to produce energy during on-peak time for the most part so I can use it and not have to pay for the high 3x on-peak cost.

Also because my on-peak time in the winter is 5-9am & 5-9pm (no solar energy during these times), any excess rolled over from the summer on-peak bucket can be well used for heating (I use heat pump, no gas) in the winter during the on-peak time.

So, different strokes for different folks. But the article does have a point that orientation for max solar generation should not be the only consideration. You need to combine that with other factors to decide on the VALUE of your solar generation, not just the amount.

[J.P.M.]

We are not suggesting that you get up and change the direction of the panels on your roof. A new study by the Pecan Street Research Institute does make you think about the conventional wisdom that home solar panels should be placed on the southern facing roof to be most efficient. You may not have heard of Pecan Street before, but their study certainly had a way of garnering the headlines with experts coming down on both sides of the argument-and this from a study of just 50 homes. Some may discount the study because the sample was small and because the output over the course of the year is only a couple of percentage points different, but the major issue is in peak load from 3pm to 7pm west-facing solar panel arrays produced 49 percent more electricity during this peak demand time. Now what all of the commentators are missing of course is that most homeowners are not going to be redirecting their houses to the south or west just to capture a few more rays.

More...

This is simplistic to my way of looking at it - simplistic and incomplete, borne of ignorance, and therefore caution may need to be exercised. Looks like it was conceived and written by folks who exemplify the old saw that a little knowledge is a dangerous thing.

1.) IF folks are on T.O.U. the electricity generated MAY be more "valuable". However, not all users are on T.O.U. I'm not sure the piece makes that clear. Every situation is different. Without more definition of what the study's parameters were, it's useless and misleading and just plain sloppy. I guess that the piece may be trying to make the case that azimuth and elevation (not mentioned BTW), may be considered as a way to reduce one's electric bill, depending on billing and rate structure. the big deal however seems to say west facing outperforms south facing, without the courtesy and common sense of explanation.

2.) The piece seems to be implying at least to the uninformed and well meaning but clueless, that west facing panels will outperform south facing panels. This may lead some uninformed people to do something not in their best interest.

3.) It may be that in some cases, sunshine distribution as a function of time of day is skewed so as to make an off south orientation more favorable. In fact, depending on whose #'s you use, how picky you want to be and methodology, slightly off south is probably, at least on paper, better. I don't think that is what the authors are referring to. For example, at my house, non T.O.U., using TMY3 data and Sunpower 327's my optimum az. is 196 deg. at an optimum tilt of 29 deg. But west facing in all cases - I sincerely doubt it - and I wonder if the authors of that drivel would even understand what I just wrote.

4.) Unless everything I've learned in a classroom as well as through hands on and more self study is wrong, this "study" is crap as presented, both in detail and presentation. People who write this stuff are supposed to be informed. I've seen better high school science projects. These people should get a clue, get informed and put out something constructive and helpful. This is not helping.

[J.P.M.]

We are not suggesting that you get up and change the direction of the panels on your roof. A new study by the Pecan Street Research Institute does make you think about the conventional wisdom that home solar panels should be placed on the southern facing roof to be most efficient. You may not have heard of Pecan Street before, but their study certainly had a way of garnering the headlines with experts coming down on both sides of the argument-and this from a study of just 50 homes. Some may discount the study because the sample was small and because the output over the course of the year is only a couple of percentage points different, but the major issue is in peak load from 3pm to 7pm west-facing solar panel arrays produced 49 percent more electricity during this peak demand time. Now what all of the commentators are missing of course is that most homeowners are not going to be redirecting their houses to the south or west just to capture a few more rays.

More...

This is simplistic to my way of looking at it - simplistic and incomplete, borne of ignorance, and therefore caution may need to be exercised. Looks like it was conceived and written by folks who exemplify the old saw that a little knowledge is a dangerous thing.

1.) IF folks are on T.O.U. the electricity generated MAY be more "valuable". However, not all users are on T.O.U. I'm not sure the piece makes that clear. Every situation is different. Without more definition of what the study's parameters were, it's useless and misleading and just plain sloppy. I guess that the piece may be trying to make the case that azimuth and elevation (not mentioned BTW), may be considered as a way to reduce one's electric bill, depending on billing and rate structure. the big deal however seems to say west facing outperforms south facing, without the courtesy and common sense of explanation.

2.) The piece seems to be implying at least to the uninformed and well meaning but clueless, that west facing panels will outperform south facing panels. This may lead some uninformed people to do something not in their best interest.

3.) It may be that in some cases, sunshine distribution as a function of time of day is skewed so as to make an off south orientation more favorable. In fact, depending on whose #'s you use, how picky you want to be and methodology, slightly off south is probably, at least on paper, better. I don't think that is what the authors are referring to. For example, at my house, non T.O.U., using TMY3 data and Sunpower 327's my optimum az. is 196 deg. at an optimum tilt of 29 deg. But west facing in all cases - I sincerely doubt it - and I wonder if the authors of that drivel would even understand what I just wrote.

4.) Unless everything I've learned in a classroom as well as through hands on and more self study is wrong, this "study" is crap as presented, both in detail and presentation. People who write this stuff are supposed to be informed. I've seen better high school science projects. These people should get a clue, get informed and put out something constructive and helpful. This is not helping.

[silversaver]

From what I read, the forks on west coast best optimum az. is toward SW than true 180. It really depends on your location. But hey, if your only option is facing North then you have no choice. My neighbor for example.....

[J.P.M.]

From what I read, the forks on west coast best optimum az. is toward SW than true 180. It really depends on your location. But hey, if your only option is facing North then you have no choice. My neighbor for example.....

There are always choices. I'd suggest the best option is the one that makes the most sense for each situation as defined by the end user - regardless or azimuth and elevation (tilt). Adding solar electric is an option, not a mandate. Not adding solar because it does not fit the economic requirements is also a sensible option and a choice. If any option is not economically viable due to layout, circumstances, etc. then I'd take it off the table. From an economic standpoint (only), the best orientation is the one that gives the best economic return, regardless of azimuth and elevation. The "least best" orientation is the one that just meets the customer defined economic criteria. Some situations may not have even a least best option. Usually compromises with and among other, non economic competing priorities - aesthetics, HOA's, shading, etc. etc.- are necessary. Sometimes one's resource availability is not enough to make economic sense, or as much economic sense as specified/defined by the potential user. Defining the economic parameters and understanding what types of energy resource information are necessary to make an informed decision is somewhat involved but not impossible. The rub, as I've come to observe is that many, maybe even most people don't understand what's involved in this type of thing and don't want to know or even take the time to become aware of what may be involved for a more informed choice, making them easy marks for solar peddlers of the ilk that will latch onto sloppy crap like this article and use it, for example, to justify a way off south orientation to a west azimuth limited and solar ignorant customer. Caveat Emptor.

[GridGrants]

I'd suggest the best option is the one that makes the most sense for each situation as defined by the end user - regardless or azimuth and elevation (tilt).

I was in disbelief when I first read the original article, myself. It was poorly written and misinformed on a subject that isn't that difficult to explain. To say you will generate more power with a western-facing array is a yes and no, but highly incomplete statement. Besides... we are more concerned with how much ENERGY we are producing and not how much POWER. That said... using the terms interchangeably like most lay people, it still just did not give enough information.

Can you produce more energy in the early morning with an east-facing array instead of a south-facing one? Sure. That doesn't mean you should face your array east. It takes someone knowledgeable to do proper siting. Choosing the correct configuration includes all the physical parameters PLUS everything else. In other words... Site specific data + user specific data = best configuration.

What we need is a PVWatts that includes a questionnaire where you can input time of use data specific to your situation and even work schedules and a breakdown on when some is home vs. property is vacant. Then you scan in a shading diagram and it spits out a couple numbers for a fixed installation or a program for an automated tracking system. Now that would be cool.

[J.P.M.]

I was in disbelief when I first read the original article, myself. It was poorly written and misinformed on a subject that isn't that difficult to explain. To say you will generate more power with a western-facing array is a yes and no, but highly incomplete statement. Besides... we are more concerned with how much ENERGY we are producing and not how much POWER. That said... using the terms interchangeably like most lay people, it still just did not give enough information.

Can you produce more energy in the early morning with an east-facing array instead of a south-facing one? Sure. That doesn't mean you should face your array east. It takes someone knowledgeable to do proper siting. Choosing the correct configuration includes all the physical parameters PLUS everything else. In other words... Site specific data + user specific data = best configuration.

What we need is a PVWatts that includes a questionnaire where you can input time of use data specific to your situation and even work schedules and a breakdown on when some is home vs. property is vacant. Then you scan in a shading diagram and it spits out a couple numbers for a fixed installation or a program for an automated tracking system. Now that would be cool.

Something similar to what you describe may already exist. Check out something called the "System Advisor Model", SAM, on the NREL website. It will allow and use hourly usage input for 8760 hrs. I think the new PV Watts (beta) is using some of the same features. Just beware of the electric rate structure it gives you. The one for SDG & E (mine) is a mess. I don't know about the other rate structures listed, but there are a lot of them. I put in my own rate structure which SAM will accommodate. I'm quite sure mine is accurate and up to date and SAM gives mostly, but not completely correct bills. SAM also works w/ T.O.U. for times once you get the rate thing straight. The rest of SAM is like PV Watts on steroids. It really is a bear. System pricing and fin. analysis are flexible and useful, but a little like giving a loaded pistol to a 2 year old if you don't know what you're doing. I've found it gives system output results comparable to stuff I've written, which may be like damning w/ faint praise depending on who you talk to.

[Volusiano]

Something similar to what you describe may already exist. Check out something called the "System Advisor Model", SAM, on the NREL website. It will allow and use hourly usage input for 8760 hrs. I think the new PV Watts (beta) is using some of the same features. Just beware of the electric rate structure it gives you. The one for SDG & E (mine) is a mess. I don't know about the other rate structures listed, but there are a lot of them. I put in my own rate structure which SAM will accommodate. I'm quite sure mine is accurate and up to date and SAM gives mostly, but not completely correct bills. SAM also works w/ T.O.U. for times once you get the rate thing straight. The rest of SAM is like PV Watts on steroids. It really is a bear. System pricing and fin. analysis are flexible and useful, but a little like giving a loaded pistol to a 2 year old if you don't know what you're doing. I've found it gives system output results comparable to stuff I've written, which may be like damning w/ faint praise depending on who you talk to.

I tried SAM out but it only allows me to enter only 1 inverter/configuration, but I have 2 inverters in my system. Am I missing something?

[GridGrants]

I tried SAM out but it only allows me to enter only 1 inverter/configuration, but I have 2 inverters in my system. Am I missing something?

Just enter them as two different systems and then add the output. Same thing as different orientations. Add each orientation as a different system and add the results.

I have also noticed when helping people that they don't use the right number for their cost per kWh when seeing what they will save. If your utility charges 12 cents per kWh and you see this on your bill... it isn't your true cost. You need to factor in the base charge and taxes, etc. This isn't even getting into a TOU charge. Take the total due and divide by the total KWh used and you will often get a number 20% or more higher than what you see for the "rate". I think it will become a bigger issue in the future as utilities will try to raise the base flat-charge and lower the per kWh charge slightly. We can debate whether that is fair or not. I see the APS point in Arizona where they are going to be tacking on $5 per month if you have solar. They are still maintaining the lines even if you only use it occasionally. I can see other utilities raising their base fees before it becomes an issue with increased rollout of solar. Some increase is justified, but not the $150 per month APS wanted.

[Volusiano]

Just enter them as two different systems and then add the output. Same thing as different orientations. Add each orientation as a different system and add the results.

Yeah, I thought about doing that but it's not worth my time and effort to enter as multiple systems because it looks like it's a lot of effort to set up my TOU pricing structure in there, and I don't care to do that more than once. I also don't like to have to add results together at the end. A robust enough system should have thought of that since it's not that uncommon for people to have multiple inverters.

[J.P.M.]

Yeah, I thought about doing that but it's not worth my time and effort to enter as multiple systems because it looks like it's a lot of effort to set up my TOU pricing structure in there, and I don't care to do that more than once. I also don't like to have adding results together at the end. A robust enough system should have thought of that since it's not that uncommon for people to have multiple inverters.

Agreed. Check the SAM feedback section. If multiple inverters are not uncommon, it's likely someone has already inquired. I'd send them a note. I've found them to be fairly responsive.

[SunEagle]

Interesting information. One reason I have hesitated installing a PV system was due to the small area of Southerly facing roof which would not allow a system bigger than 5kw.

I have more roof facing West that doesn't get much sun until about 10 AM but has no shade until the sun goes down. Now with this little bit of info I may think of at least getting some quotes on a system incorporating both South and West facing roofs. A bigger system may be able to get the $/watt cost down than a small system.

[J.P.M.]

Interesting information. One reason I have hesitated installing a PV system was due to the small area of Southerly facing roof which would not allow a system bigger than 5kw.

I have more roof facing West that doesn't get much sun until about 10 AM but has no shade until the sun goes down. Now with this little bit of info I may think of at least getting some quotes on a system incorporating both South and West facing roofs. A bigger system may be able to get the $/watt cost down than a small system.

The bigger software programs like SAM do take more input and more care that you know/learn what you're doing. Among the rewards is the flexibility to see the effect, sometimes very quickly, that big or small changes to the input makes in the (estimated) output and economics with more confidence (not certainty to be sure) than obtained from dumbed down software that assumes a lot that may not fit your situation. Using the detailed stuff can also be a good learning experience. The downside for me was (is) the time and frustration involved. However, if time spent learning is analogous to money then I guess the adage applies: you want nice, you pay nice. The output is nice. I also try to keep in mind that all the programs/software have their own quirks, none are error free and none are gospel.