Strings in different orientations

[Bendesa1962]

Hi all,

I have the following question. I'm curious what you guys think about that. Please let me know. I always appreciate your replies.

Facts:

1 Inverter 100 KW

MPPT number/Max. input strings number 10/20

20 Strings of 18 panels eacht.

I have the following possibilities.

Roof facing North
Roof facing South
Location Indonesia

I can easily install all panels facing South no problem
I can also choose to install all panels facing North.

Or I can decide to install 10 Strings facing North
and 10 Strings facing South

What would be your choice

Thanks everybody in advance

Kind regards

B

[SunEagle]

Maybe a simple answer is to place all of the panels in the same orientation unless you live right on the equator. North of the E the panels do better facing South. And South of the E the panels do better facing North. Since you mention your location is Indonesia I presume you are either South or North of the E.

How you wire the panels will depend on the Max input levels for your inverter. But I am confused of your roof size and if it can actually handle 360 panels (20 strings of 18).

Also the size of your system (100kw) seems to not to be for a home but some other co-generator.

[Bendesa1962]

Hi Suneagle
thx for your replay. This projects is just South of the E. I ask this question because there seemed to be different opinions on this.

And you're right it's not for a residential installation. It's an industrial project of total 1 MW.

Thx for your contribution.

[SunEagle]

Hi Suneagle
thx for your replay. This projects is just South of the E. I ask this question because there seemed to be different opinions on this.

And you're right it's not for a residential installation. It's an industrial project of total 1 MW.

Thx for your contribution.

Depending on exactly what the latitude of the array is should help you decide which way to point the panels. If you are close to the E then pointing the N or S should work but if you are truly S of the E then point them all to the North unless they get shaded.

Up North some of the larger arrays tilt the panels in an Easterly and Westerly direction to get as much of the useful sunlight all day long. But they usually have a single axis control that moves the panels. If you have fixed panels S of the E then my guess is to point all of them to the North and that will get you the best production.

[J.P.M.]

If I had to ask that question and I thought I needed a 100 STC kW array, my choice would be to hire a consultant who knows more about the subject of PV system design than I do.

Beyond that, and just a few other considerations besides orientations, unless there are some intervening factors, your low latitude, essentially on the equator will usually indicate a low panel slope, maybe even horizontal. That means array compass orientation won't matter too much. What are the roof slope(s) ?

However, north or south orientation and divvying up which roof gets how much of the total array may well hinge on which roof orientation gets the most sun as f(rainy season).

Orientation, or how much of the total array goes on which side of the roof, north or south, may also matter with respect to when and how the power will be needed. For example, if more power is needed during the rainy season, you may want to put more panels on the side of the roof that gets the most sun during that time of year even if it results in less total annual array output.

Also, know that horizontal orientations usually generate problems with keeping the panels clean. Reason: The panels will turn into mud pans or evaporation pans in short order, or, maybe not dry off completely during any rainy season. Staying wet for extended periods may well cause other problems as panels may be submitted to moisture conditions they were not designed to handle.

Will this be tied to some utility or other energy source ?

If grid tied, is any form of net metering available ? Other requirements or considerations from the power supplier ?

Is any energy storage planned or required for the application ?

I'd start by doing a load study as to what the loads are, how much they are and when the loads are needed.

And, if it was me and the roof slope is much less than ~ 10 deg. off horizontal, I'd plan on easy roof access and leaving room between panels for cleaning and other service access. And/Or, putting the array on a sawtooth pattern, but that (tilted racking) option means more time/toil/hassle/cost. In any case, plan on about 10 deg. off horizontal slope to have a chance of keeping the array reasonably clean and so have reasonable output reduction from array fouling.

BTW, is this the same application you started with when you first posted here on 11/16/2016 ? Are you an installer ? A 100 kw system is pretty big, even for a big home.

[littleharbor2]

Being near the E this style of single axis tracker may be the best method for production. You probably get enough rain to keep panels clean but this would also ensure you get good drainage and no mud puddling on the edges of the panels.

[J.P.M.]

Being near the E this style of single axis tracker may be the best method for production. You probably get enough rain to keep panels clean but this would also ensure you get good drainage and no mud puddling on the edges of the panels.

Tracking systems may increase specific output (kWh/yr. per installed STC W), but will require more maintenance. And, being more complicated, will have lower reliability. All that is paid for in greater costs both initial and operating.

Having engineered a couple of energy projects in that part of the world I learned to appreciate that depending on location, designing for (and usually around) availability of services can be a greater consideration than at first thought. KISS applies in spades.

[littleharbor2]

It's a good thing that some are willing to go beyond the basic, no moving parts design. We might still be getting around in horses and buggies if it weren't for these visionaries.

[J.P.M.]

It's a good thing that some are willing to go beyond the basic, no moving parts design. We might still be getting around in horses and buggies if it weren't for these visionaries.

The idea and purpose of design engineering is to come up with equipment and benefits that are safe and most fit for the required service and application.

One of the criteria and methods for doing that is to not make things more complicated than they need to be.

Sometimes/Often, tracking systems with their added complexity and cost - because they often violate the KISS principle - are not the best choice for a given application.

IMO only, based on what I've read from the OP, a tracking system for the described application has a higher probability of being more problematic and costly than a stationary system, and will require more service than the application can likely handle in a practical and cost effective manner.

I therefore don't believe it's a viable solution when compared to a stationary array. Also, that's not meant to imply that I think PV is a good solution for this application. More information would be necessary to make that judgement.

[bcroe]

In some areas (like here), a fixed multi orientation array can have
another advantage. It can help compensate for clouds, a tracking
array can do nothing about them. Bruce Roe

[SunEagle]

The idea and purpose of design engineering is to come up with equipment and benefits that are safe and most fit for the required service and application.

One of the criteria and methods for doing that is to not make things more complicated than they need to be.

Sometimes/Often, tracking systems with their added complexity and cost - because they often violate the KISS principle - are not the best choice for a given application.

IMO only, based on what I've read from the OP, a tracking system for the described application has a higher probability of being more problematic and costly than a stationary system, and will require more service than the application can likely handle in a practical and cost effective manner.

I therefore don't believe it's a viable solution when compared to a stationary array. Also, that's not meant to imply that I think PV is a good solution for this application. More information would be necessary to make that judgement.

I agree with you. Having any type of panel axis moving device increases maintenance costs and possibility of equipment failure. Which is why most home installs are fixed yet a lot of POCO installs (which have a maintenance crew) are single or multi axis. Remember they can pass on the cost of maintenance to the customer. The cost and choice are those for the OP to make.

[J.P.M.]

In some areas (like here), a fixed multi orientation array can have
another advantage. It can help compensate for clouds, a tracking
array can do nothing about them. Bruce Roe

One of the disadvantages of multiorientation of an array is usually and commonly that specific output as expressed in kWh/yr. per installed STC kW will be less than for one optimally oriented tilt and azimuth. That is, multiple orientations will need a larger array to produce the same annual output as one optimal orientation. The multiple orientations will also possibly cost more in terms of $$/installed STC W.

So, less output/STC W for more $$/STC W. For those interested in most bang for the buck out of a PV array investment, that doesn't seem the best way to go.

If the application requires splitting arrays, so be it. Design for it. But to split an array because it's thought that more output per installed STC W will be obtained is not correct. Think about it: If that were the case, commercial and utility or just plain large arrays would have multiple orientations. So, why don't they ?

Also, and for several reasons too involved to describe here, and in spite of what you may think, multiple array orientations will not enhance output per installed STC kW under cloudy conditions.

As a matter of fact, since most irradiance that reaches the ground under cloudy conditions is diffuse, and because a portion of the diffuse irradiance below ~15,000 ft. elevation is composed of forward scattered irradiance (meaning it tends to come from the direction of the sun), that means a portion of the diffuse irradiance will usually come from the direction of the sun. It's usually not noticed since the irradiance under clouds is not much to begin with in a relative sense and for most purposes isn't worth spending a significant amount of time designing for. The point is, more of the diffuse irradiance under cloudy conditions comes from the same direction of the sky as the beam irradiance under clear skies, meaning an array that spends a lot of its daylight hours under clouds will probably gather the most diffuse irradiance when it has a single optimal orientation that is the orientation designed for optimal total output per installed STC W.

There are a lot of applications that require multiple array orientations for reasons unrelated to maximizing output, but it's important to understand that designing for multiple orientations with the sole idea of improving the energy output per installed STC W of an array because multiple orientations will result in more production per installed STC W is a mistake.

Think about this: Two PVWatts runs for a 2,500 STC W array, both at a tilt = to local latitude. One run done at optimal azimuth. The other at, say 45 degrees off optimal azimuth. Which one has greater output ? Now, add the outputs together. Now, do a 3d run. Do this run for a 5 kW array at the optimal orientation, same as the first array. Now, compare the resulting total outputs of the two smaller outputs with that of the single larger output. Which one is greater ?

Back to the cloudy output question, the optimal orientation for energy harvest under cloudy skies is about the same as the optimal array orientation under clear skies.

And, in the same location and application, an array with one fixed orientation that's optimized for maximum annual output will have greater output per installed STC W than any other combination of array orientations for the same total area.

BTW, the same idea holds true for applications that involve most common T.O.U. tariffs. There, the idea is to find one orientation that produces the greatest total annual "revenue" per installed STC kW - probably and not necessarily the greatest energy harvest per installed STC W - that can be used to offset the cost of the total amount of electricity used.