[RE-wrenches] losses to competing angles in single strings

Sat Sep 12 21:30:37 PDT 2009

Peter and Nick,

I believe the original question related to two different orientations in the
same series string. Peter, your response was related to two parallel strings
of the same length run into the same inverter. These are two EXTREMELY
different scenarios. They have been discussed in detail many times on this
forum, but bad information continues to be circulated-predominantly by a
variety of inverter manufacturers.

Although it may be technically best to put every string, and even every
module, on a different inverter to provide best tracking for PV, the
realities of the IV curve must be understood. Since IV curve tracers are so
expensive, most people have not had the luxury of testing many difference
scenarios of arrays as a few of us have had.

With crystalline silicon products, strings of the same length (number of
modules), that do not experience different shading, will have nearly
identical operating voltages regardless of orientation. There will be small
differences due to the temperature impact of irradiance. The operating
temperature will vary about 3C for every 100W/m^2. This will cause a slight
change in operating voltage. However, higher irradiance, which causes a
slight depression in voltage due to increased temperature, has a very weak
increase in voltage due to irradiance. These are essentially offsetting
results, making the max power voltage nearly identical for two identical
strings in two different orientations-therefore no problem exists.

However, differential shading on two identical strings will always have a
significant impact-usually on the string with the most shading-depending on
whether the inverter can see both maximum power points or not.

Back to the first question from Nick. Many people want to know what happens
if you split a single string between two orientations. The first answer is
don't ever do it, but the second answer is found in the IV curve. The
irradiance will favor one face or the other at different times of the day
and year (as in your example). In all cases, the surface with the lowest
irradiance will govern the current flowing through the string-meaning the
lowest current will always prevail. For two surfaces close together (30 and
45 tilt and same azimuth) the loss will be small-maybe 5-10% annually. If
the difference is dramatic-I have seen strings split between east and west
faces at a 45 degree tilt angles-the loss can be 80-90% annually. Nick,
although you would be breaking the first rule of array design by doing what
you suggest, the result will likely be tolerable for this specific example.
I would only allow such a situation when all other options of mounting and
aesthetics have been exhausted. For example, I would support what you
suggest instead of jacking up the 30-degree section of the array to
45-degrees if this is on a prominent roof face that can be seen by the
world. 

Aesthetics needs to take precedence over performance for highly visible
applications when other options make the array look like an idiot installed
it. If we are going to get PV on as many roofs as possible, it has to look
good. Putting up huge billboards on the top of houses that scream "AN IDIOT
WHO LOVES SOLAR LIVES HERE", will not help us achieve our goal. As much as
my technical side hates to admit it, sometimes performance has to take a
back seat to aesthetics-but only if there is no other way.

Bill.

From: re-wrenches-bounces at lists.re-wrenches.org
[mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Peter
Parrish
Sent: Friday, September 11, 2009 4:53 PM
To: 'RE-wrenches'
Subject: Re: [RE-wrenches] losses to competing angles in single strings

In principle, two strings with different orientations should not be wired in
parallel into the same inverter. However, when the mismatch is minor (15
degrees in pitch or 15 degrees in azimuth) there is practically no
degradation. Fronius circulated a white paper for a while that argued that
their inverters did maximum power point tracking on the string with the
greatest irradiance and therefore did an optimum mppt  job on the string
with the greatest "available power". This part of the argument makes sense.
What I have never been clear about is how close to "optimum" this
arrangement is (optimum being two separate monitors one for each string).

Fronius appears to be backing away from this white paper, as I was told
about six months ago from a member of their tech support staff that they
recommended two separate inverters.

- Peter

Peter T. Parrish, Ph.D., President
California Solar Engineering, Inc.
820 Cynthia Ave., Los Angeles, CA 90065
CA Lic. 854779, NABCEP Cert. 031806-26
peter.parrish at calsolareng.com  
Ph 323-258-8883, Mobile 323-839-6108, Fax 323-258-8885

  _____  

From: re-wrenches-bounces at lists.re-wrenches.org
[mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Nick Vida
Sent: Wednesday, September 09, 2009 6:38 PM
To: wrenches
Subject: [RE-wrenches] losses to competing angles in single strings

hello wrenches.

I was wondering if any of you have any data on what the losses would be if
you had your string split with 2 different angles, but at the same azimuth.
For instance, string of 11 with 8 facing 180 south at 30 degrees, and 3
facing south 180 at 45 degrees. I have heard that the voltages and currents
would differ and the IV curve would be inefficient as a string, but I don't
know how much of a loss would be expected.

thanks in advance, and hopefully I didn't miss anything in the archives.

thanks,

nick

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20090912/9e98b0b3/attachment-0004.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 4274 bytes
Desc: not available
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20090912/9e98b0b3/attachment-0007.gif>