Finding True North for Solar Panel Placement: A Simple Method

[gary350]

I learned this in college 48 years go, problem is I have forgotten more than I ever knew. I know the answer is not very hard like 20 degrees but the internet is pissing me off. I looked this up and I get FAKE stuff like advertisements. Novels with many pages to read that never tell the answer. Formulas to calculate the answer for lots of things but what I want to know. I give up I am tired of looking. I know magnet north moves around but there is a average place where it usually is? I am trying to position solar panels so they get the same amount of sun in the morning as they do in the evening. I have a 1/4" dowel rod stuck straight up in the yard the sun shadow lines up the compus needle at 1:10 PM. This is magnetic north not true north. Setting solar panels at true north can not be that hard.

 

[jerromyjon]

Apparently it isn't that easy... I'm curious to know, too.

 

[rumborak]

The magnetic North is less than 10 degrees from the true North. And yes, while it moves around, the effect is so little that it should be negligible for your purposes. The only ones affected by the difference and movement to my knowledge are airports that are very far North. When the magnetic North moves around, they have to relabel runways, since their label indicates the magnetic North.

EDIT:

That said, historically it used to be much further South, beyond 20 degrees off. But, for the next few decades it will be very close to true North.

 

[OCR]

Apparently it isn't that easy... I'm curious to know, too.

Magnetic declination ...

East is least, and west is best....

 

[Janus]

I learned this in college 48 years go, problem is I have forgotten more than I ever knew. I know the answer is not very hard like 20 degrees but the internet is pissing me off. I looked this up and I get FAKE stuff like advertisements. Novels with many pages to read that never tell the answer. Formulas to calculate the answer for lots of things but what I want to know. I give up I am tired of looking. I know magnet north moves around but there is a average place where it usually is? I am trying to position solar panels so they get the same amount of sun in the morning as they do in the evening. I have a 1/4" dowel rod stuck straight up in the yard the sun shadow lines up the compus needle at 1:10 PM. This is magnetic north not true north. Setting solar panels at true north can not be that hard.

What you need to do is look up the magnetic declination for where you live. This is the difference between true North and the direction a compass needle will point.
For instance, the magnetic declination for where I live in Portland, OR is +15 degrees 20 min.
Here's a site that will give you that info.
http://www.magnetic-declination.com/

 

[Flash Kellam]

I learned this in college 48 years go, problem is I have forgotten more than I ever knew. I know the answer is not very hard like 20 degrees but the internet is pissing me off. I looked this up and I get FAKE stuff like advertisements. Novels with many pages to read that never tell the answer. Formulas to calculate the answer for lots of things but what I want to know. I give up I am tired of looking. I know magnet north moves around but there is a average place where it usually is? I am trying to position solar panels so they get the same amount of sun in the morning as they do in the evening. I have a 1/4" dowel rod stuck straight up in the yard the sun shadow lines up the compus needle at 1:10 PM. This is magnetic north not true north. Setting solar panels at true north can not be that hard.

If you are trying to determine the direction of true north in order to position your solar panels – or, better still, your new home - there is a method for determining that direction that is far simpler than fumbling with a magnetic compass. Here’s how:

First, consult your local newspaper or weather blog for the times of sunrise and sunset in your area. Then, calculate the time that lies midway between sunrise and sunset. That time is called solar noon. Example: If sunrise occurs at 7:14 a.m. and sunset at 6:28 p.m. on a given day, then 12:51 p.m. is the time of solar noon on that day. Suspend a plumb bob or similar weighted object above the ground. At the moment of solar noon, note where the Sun casts the shadow of the plumb bob line. The shadow points towards true north. The direction 180 degrees opposite points towards true south. That is, solar south.And, of course, the mirror case applies for Southern Hemisphere blog visitors who wish to determine the direction of solar north.It is much simpler to use this method - would you not agree? Source: Solar Remodeling, a Sunset book

 

[gary350]

I was thinking that would be the best way but I never watch TV and do not get a New paper. Maybe I can fine sun rise and sun set online for my zip code. I already have a plum bob set up.

 

[jim mcnamara]

try https://www.timeanddate.com/sun/usa/albuquerque
replace the red city name with a nearby name or go to the site and work your way down. I'm near Albuquerque.

 

[Flash Kellam]

I was thinking that would be the best way but I never watch TV and do not get a New paper. Maybe I can fine sun rise and sun set online for my zip code. I already have a plum bob set up.

Get the times of sunrise and sunset for your city by using the following website:
https://www.timeanddate.com/
It is a superb free almanac.

 

[jim hardy]

This .gov site doesn't try to sell you something

If you type in your US zipcode and hit "Calculate" it calculates the variation/declination (two names for it) for you and shows a map.
https://www.ngdc.noaa.gov/geomag-web/"Add Westerly Variation "

in other words , if variation is West X degrees, ADD X to your compass reading to get the truth.

Last place i lived i had to add about 6 degrees

right now it's about zero here in Arkansas
fifty years ago it was about zero there.

old jim

 

[alayman]

Here is another method to find true North that should be accurate and easy. Locate the North Star (Polaris) in the night sky. Drive two stakes in the ground several meters apart that line up with the star. Of course, if you are on a rooftop, you wouldn't drive stakes to make a pointer, but you could simply mark the roof with a suitable tool.

 

[russ_watters]

I have a 1/4" dowel rod stuck straight up in the yard the sun shadow lines up the compus needle at 1:10 PM. This is magnetic north not true north.

It's true north, if you do it right. You need to know what time is your "local noon". Time zones are 15 degrees or 1 hour wide/apart, so depending on where in the time zone you live, you might get an answer that is off by +-7.5 degrees. What you do is find your longitude on a map or with a GPS that reads position and subtract from the nearest time zone center (75 degrees, 90 degrees, 105 degrees, etc.). Divide the difference by 15 and multiply by 60 to find the time added or subtracted from what you see on your watch (er...minus the hour for daylight savings time too) to find the time of local noon.

I'd prefer the Polaris method though.

 

[rumborak]

Ok, here's a question based on my own post above:

In 1859, magnetic North was around Cambridge Bay in Canada. Does that mean that a sailor in Iceland had to deal with something like a deviation of 45 degrees from true North?

 

[Anachronist]

I am trying to position solar panels so they get the same amount of sun in the morning as they do in the evening. I have a 1/4" dowel rod stuck straight up in the yard the sun shadow lines up the compus needle at 1:10 PM. This is magnetic north not true north.

You almost got it.

Here's what I learned in an outdoor skills and survival class about navigating without a compass. It was just about as long ago as you were in college, and it stuck with me ever since -- and has proven useful now and then. You don't need to look up sunrise or sunset or anything.

1. Get a stick, a couple of feet tall (taller the better as long as it's stiff enough to hold still) and drive it into the ground so that it stands up. It doesn't have to be straight up. You're interested only in the top of the stick.
2. The stick will cast a shadow. Mark on the ground where the shadow of the top of the stick falls.
3. Wait several minutes, up to an hour, but you can get a decent reading after 10 minutes or so.
4. Put another mark on the ground where the shadow of the top of the stick falls.
5. Draw a line between the two marks.
6. Because the sun moves from east to west, shadows move in the opposite direction; therefore, your first mark will be on the west end of the line, and your second mark will be the east end.
7. Ninety degrees from this line is true north. Voila!

You can also use a board with a hole in it, and see where the sunlight through the hole hits the ground. The important thing is the stick or the board cannot move during the course of this measurement.

 

[FactChecker]

View attachment 208260

The magnetic North is less than 10 degrees from the true North. And yes, while it moves around, the effect is so little that it should be negligible for your purposes. The only ones affected by the difference and movement to my knowledge are airports that are very far North. When the magnetic North moves around, they have to relabel runways, since their label indicates the magnetic North.

EDIT:

View attachment 208261

That said, historically it used to be much further South, beyond 20 degrees off. But, for the next few decades it will be very close to true North.

I never realized how fast magnetic North moved till I saw your post.

 

[FactChecker]

Ok, here's a question based on my own post above:

In 1859, magnetic North was around Cambridge Bay in Canada. Does that mean that a sailor in Iceland had to deal with something like a deviation of 45 degrees from true North?

Stars and a clock would be much better, or at least help them to calibrate their magnetic devices. Ocean navigation was a big motivator in the development of clocks.

 

[Janus]

I never realized how fast magnetic North moved till I saw your post.

I've got at National Geographic map of Glacier National Park. It gives the magnetic deviation at the time of printing (2009) and the approximate change in deviation per year to keep it up to date.

 

[Anachronist]

I've got at National Geographic map of Glacier National Park. It gives the magnetic deviation at the time of printing (2009) and the approximate change in deviation per year to keep it up to date.

Or, you can go to https://skyvector.com/ to look at up-to-date pilot's navigational charts, which show magnetic north deviations as isogonic lines. You have to zoom into find them. They're purplish in color and are labeled with degrees and minutes, like this screenshot showing a region southeast of Dallas, TX:

I still think my suggestion about using the motion of the tip of a shadow to track a west-east line is an easier way of finding true north wherever you are. Or at night, Polaris (the North Star) is extremely close to true north also.

 

[phyzguy]

It's true north, if you do it right. You need to know what time is your "local noon". Time zones are 15 degrees or 1 hour wide/apart, so depending on where in the time zone you live, you might get an answer that is off by +-7.5 degrees. What you do is find your longitude on a map or with a GPS that reads position and subtract from the nearest time zone center (75 degrees, 90 degrees, 105 degrees, etc.). Divide the difference by 15 and multiply by 60 to find the time added or subtracted from what you see on your watch (er...minus the hour for daylight savings time too) to find the time of local noon.

This method ignores the equation of time. Depending what time of year you do it, you could be as much as 15 minutes off. Better to take local noon as halfway between local sunrise and local sunset as suggested by Flash Kellam, or use Polaris, as suggested by others.

You almost got it.

Here's what I learned in an outdoor skills and survival class about navigating without a compass. It was just about as long ago as you were in college, and it stuck with me ever since -- and has proven useful now and then. You don't need to look up sunrise or sunset or anything.

1. Get a stick, a couple of feet tall (taller the better as long as it's stiff enough to hold still) and drive it into the ground so that it stands up. It doesn't have to be straight up. You're interested only in the top of the stick.
2. The stick will cast a shadow. Mark on the ground where the shadow of the top of the stick falls.
3. Wait several minutes, up to an hour, but you can get a decent reading after 10 minutes or so.
4. Put another mark on the ground where the shadow of the top of the stick falls.
5. Draw a line between the two marks.
6. Because the sun moves from east to west, shadows move in the opposite direction; therefore, your first mark will be on the west end of the line, and your second mark will be the east end.
7. Ninety degrees from this line is true north. Voila!

You can also use a board with a hole in it, and see where the sunlight through the hole hits the ground. The important thing is the stick or the board cannot move during the course of this measurement.

This method doesn't really work unless the sun is directly overhead, like on the equator at one of the equinoxes, for example. Elsewhere, if you do it near local noon, it will work, but if you know when local noon is, you can just take the direction of the shadow as north, as discussed elsewhere. If you try your method near sunrise or sunset, the line you described will point nowhere near north. Try it and see.

 

[Anachronist]

This method doesn't really work unless the sun is directly overhead, like on the equator at one of the equinoxes, for example. Elsewhere, if you do it near local noon, it will work, but if you know when local noon is, you can just take the direction of the shadow as north, as discussed elsewhere. If you try your method near sunrise or sunset, the line you described will point nowhere near north. Try it and see.

I have tried it many times. You have a rather large window (an hour or so) before and after local noon where the shadow movement method gives very good results. You get perfect results on or near the equinoxes, when shadows move in a straight line from sunrise to sunset.

But even on the worst possible days (the solstices), doing the exercise when the sun is approximately near its zenith still gives good results. You don't need to know when local noon is, you just have to be reasonably close. If you want perfection near the solstice, you can get a good west-east line by marking the shadow any time before noon, marking intervals as you observe the shadow get shorter, and then stop marking when the shadow is the same length as the original mark. The shadow traces a hyperbolic path with the apex pointing at your stick, so you just go for symmetry of the hyperbola to get your west-east line. And the point where the shadow is shortest would be local noon, I suppose.

For the purpose of the OP, to line up solar panels, it doesn't have to be exact.

 

[DaveC426913]

I never realized how fast magnetic North moved till I saw your post.

Something taught to every sailor is how to read both the declination of magnetic North and how to calculate its movement over the period since the map was published.

Every nautical map has a compass rose that shows both Mag North and its rate of change.

 

[Number 42]

Apart from the good advice note that the sun input to your panels is a cosine funktion of the angle and not super critical to get very accurate.

Important is that you clean the panels , more frequently if there is a lot of dust in the air , and you should run a test now and then to determine that the efficiency still is OK. It will deteriorate with time and hopefully you have a guarantee that set the max amount.
Good luck with the project.

 

[SKHanson57]

I just found the true midpoint in the day where I live (halfway between sunrise and sunset). Then drove a stake ahead of where I wanted my solar panels mid point to be then followed the shadow in a northward direction. Lined up a stake with the shadow and first stake at where exactly I wanted the panels and that was the midpoint of the panel or panels. Then pepindicular to the line of first 2 stakes on each side of my midpoint one to the east and on to the west would line up with the 2 ends of my panel or panels.

 

[DaveC426913]

BTW, a study by some junior whizbang a few years back pointed out that all solar panels today are (or at least were) configured wrong.

He pointed out the power usage is not symmetrical across the day. More power is used in the early evening than in the early morning. A set of panels that takes this into account (by being turned more westward) is more efficient than one that assumes an even demand of power over the day.

It was not insignificant; it was at least 10 degrees.

 

[NTL2009]

BTW, a study by some junior whizbang a few years back pointed out that all solar panels today are (or at least were) configured wrong.

He pointed out the power usage is not symmetrical across the day. More power is used in the early evening than in the early morning. A set of panels that takes this into account (by being turned more westward) is more efficient than one that assumes an even demand of power over the day.

It was not insignificant; it was at least 10 degrees.

I don't think efficient or wrong are the right terms here. The panels are clearly most efficient when they capture and convert the most sunlight over the course of the day. Efficiency is power out over power in.

In most areas, solar is a small percentage of total power, so matching it to demand isn't an issue, so it's not wrong for them to be symmetrical. If solar is a large percentage of demand, it may be more effective for the grid to shift some solar production to later in the afternoon.

 

[DaveC426913]

The panels are clearly most efficient when they capture and convert the most sunlight over the course of the day.

Panels are merely a single component in a system whose ultimate purpose is to get power to where its needed, when it's needed, and do so economically.
One must prioritize the whole system, and its overall efficiency in meeting its goals, over any single component.

For example, panels producing electricity when that electricity is not being used is wasted - unless one has a sufficient storage system (and even then they can have losses.) We don't know what the OP's storage system is, and how much power it can store or for how long. So, the system needs to be prioritized above the actual max output of the panels.

 

[rumborak]

I agree. The right thing to do is to analyze your power usage during the day, and configure your panels to produce maximum efficiency at those times.

 

[FactChecker]

I guess that some suppliers charge more during the time of day when there is higher usage (where I live, there are contract options with free electricity during some hours.) In that case, you would want to turn the panels to work best when you are paying the most.

 

[NTL2009]

Panels are merely a single component in a system whose ultimate purpose is to get power to where its needed, when it's needed, and do so economically.
One must prioritize the whole system, and its overall efficiency in meeting its goals, over any single component. ...

Yes, but the OP mentioned only the panels - no system restrictions were described. And it would be a special case (but not unheard of) to need to optimize the system by reducing the overall efficiency of the panels through time shifting (and thus lowering) their output.

But I think (especially on a technical forum like this), we should be careful to use the term "efficiency" in it's technical sense, power out / power in. If we need to reduce the efficiency of the panels to match other restrictions of the system, I think that should be referred to as"optimizing" the system, or tweaking its "effectiveness".

For example, panels producing electricity when that electricity is not being used is wasted - unless one has a sufficient storage system (and even then they can have losses.) We don't know what the OP's storage system is, and how much power it can store or for how long. So, the system needs to be prioritized above the actual max output of the panels.

Yes, I covered that in other parts of the post you quoted, but you didn't quote that part.

I agree. The right thing to do is to analyze your power usage during the day, and configure your panels to produce maximum efficiency at those times.

Not necessarily. As long as you aren't overproducing (wasting solar), what difference does your demand curve make? You really only care about an "over production" curve, not simply your demand curve. For example, if my peak demand were @ 5 PM, but my demand at mid-day is above my solar power output, shifting my panels to the 5 PM sun would just be reducing their power output with no benefit.

I guess that some suppliers charge more during the time of day when there is higher usage (where I live, there are contract options with free electricity during some hours.) In that case, you would want to turn the panels to work best when you are paying the most.

Yes, but I think it's a little loose to just say you want the panels to "work best" (maximum output?) when your rates are highest. You'd need to look at the entire production and rate curve and optimize them. Your peak output of the panels may still be during the lower rate, it's all about the area under the curves.

Apart from the good advice note that the sun input to your panels is a cosine function of the angle and not super critical to get very accurate.

Important is that you clean the panels , more frequently if there is a lot of dust in the air , and you should run a test now and then to determine that the efficiency still is OK. It will deteriorate with time and hopefully you have a guarantee that set the max amount.
Good luck with the project.

Relative to the OP, I think this post is key. Unless you are really far north (and likely not using solar panels!), the delta between True North and Magnetic North is probably a very tiny effect on total power out.

 

[FactChecker]

Yes, but I think it's a little loose to just say you want the panels to "work best" (maximum output?) when your rates are highest. You'd need to look at the entire production and rate curve and optimize them. Your peak output of the panels may still be during the lower rate, it's all about the area under the curves.

Yes, it's a problem of optimizing an objective function. But I think that the real dangers at this point are:
1) neglecting an important factor or
2) designing a permanent fixture based on factors that may change.

 

[DaveC426913]

Yes, but the OP mentioned only the panels - no system restrictions were described.

It behooves us to point out that this may not be the only factor to consider when configuring his panels.
It is entirely up to him to decide whether it applies to his system, but we would be doing him a disservice to omit mentioning such a salient factor.

But I agree with your terminology - the better term is optimization.

 

[DaveC426913]

Relative to the OP, I think this post is key. Unless you are really far north (and likely not using solar panels!), the delta between True North and Magnetic North is probably a very tiny effect on total power out.

It's generally on the order of 10 degrees. Roughly, the COS of that is 0.97.
If the OP didn't care about that 3%, he probably wouldn't be going to the lengths he is - both researching the calcs himself and bringing it to PF for verification.

 

[OmCheeto]

Apart from the good advice note that the sun input to your panels is a cosine funktion of the angle and not super critical to get very accurate.

Important is that you clean the panels , more frequently if there is a lot of dust in the air , and you should run a test now and then to determine that the efficiency still is OK. It will deteriorate with time and hopefully you have a guarantee that set the max amount.
Good luck with the project.

Good points! I did the maths on PV output vs misalignment a couple of years ago. [ref]

misalignment   percent
angle °        of rated
               output
30               86.6%
25               90.6%
20               94.0%
15               96.6%
10               98.5%
5                99.6%
0               100.0%

 

[NTL2009]

It behooves us to point out that this may not be the only factor to consider when configuring his panels.
It is entirely up to him to decide whether it applies to his system, but we would be doing him a disservice to omit mentioning such a salient factor.

But I agree with your terminology - the better term is optimization.

I agree that it it is worth bringing up as a question to the OP, and hopefully they will respond. I disagree with any broad-brush statements that the output needs to be matched to demand - that is system dependent and goal dependent (maximize total output or maximize output at certain times of the day).

It's generally on the order of 10 degrees. Roughly, the COS of that is 0.97.
If the OP didn't care about that 3%, he probably wouldn't be going to the lengths he is - both researching the calcs himself and bringing it to PF for verification.

But that is totally dependent upon location (which I don't think the OP provided). This link: https://upload.wikimedia.org/wikipedia/commons/2/28/Mv-world.jpg

shows a line of "No Variation" roughly from Duluth, MN to New Orleans, LA, and down through part of Central and South America. Another one does a wild curve through Europe, Africa, swooping up through India and beyond.

 

[NTL2009]

Just as another possibly interesting side note - I was reading about someone who had a little off-grid cabin in the woods that he'd visit on some weekends. He had his solar panels arranged in an extremely inefficient manner, but it was an effective way to achieve his goals.

They were near vertical, pointing straight East and straight West. He had some small trickle draw that he wanted to keep active while away. His panels provided far more peak power than he needed for this trickle draw, so this arrangement provided that power from sunrise to sunset. That minimized the draw down (and recharging cycles) on his battery storage. I think that he had them set up, so when he was there, he could easily tilt them out flat later in the day, and have more power available for his needs mid-day, or recharge the batteries from draw down from the previous evening.

The other advantage for him was that a near vertical panel will shed snow. I've seen the panels here in Northern IL have their output drop to near zero for weeks at a time. Panels here are mounted about 20° from horizontal to maximize the annual production (longer summer days), and that leads to the snow clinging to them. So for him, it was more about convenience than getting the most kW-hr out of a panel. Over-sizing the panels was a cheap price to pay for convenience.

 

[Baluncore]

Three other ways.
1. Mark your exact location on a map. Look for a distant visible feature that is also marked on the map. Draw a line on the map between the two. Rotate the map so the line points at the distant feature. The map is then correctly orientated, so the map grid will be N–S at your location.

2. Use a GPS to read Lat–Long on site and somewhere that can be seen from site. Travel back and forth to check the readings are reliable. The bearing of the line between the two sites can be found from the equations for a great circle passing through those two points.

3. If you read GPS Longitude, then find a place North or South with exactly the same longitude, the line between the two will be true North–South. The further apart the points, and the more times you check the numbers, the more accurate the N–S line.

 

[Tom.G]

If you want perfection near the solstice, you can get a good west-east line by marking the shadow any time before noon, marking intervals as you observe the shadow get shorter, and then stop marking when the shadow is the same length as the original mark. The shadow traces a hyperbolic path with the apex pointing at your stick, so you just go for symmetry of the hyperbola to get your west-east line.

That's almost the simplest way. As @Anachronist stated, the apex of the hyperbola will point at the stick at local Noon. This is the same as saying that the shadow will be shortest at Noon. The shortest shadow of stick is your North-South line, and it occurs at local Noon. No need to spend the time creating the other half of the hyperbola.