Sun hours

[ILFE]

Checking the archives revealed only my last 500 posts are accessible. So at 600+, every one I make drops another early one. I was going to compare some of my early system calculations & expectations with what actually happened, but guess that will have to be from memory. Bruce Roe

They are probably accessible, Bruce. They just may not be showing up on historical searches.

I'm not very familiar with vBulletin forum software. Personally, I use IPB. But, search history can be changed. Members on one of my forums complained about it being too short, like going back only one year. So, we changed it a couple of years ago to extend it to very early posts.


If I may, Bruce, how large is / are your array(s)?

[bcroe]

ONE YEAR of OPERATION

At the anniversary of startup, my installer came to see how things were holding up.
It had been running at max power all day long, and everything was warm to the touch
including the 4 gauge wire carrying 58A AC output. The inverters had logged 26,800
KWH despite a rather cloudy past year, exceeding predictions for the 15KW AC system.
Thats equivalent to 1000 gallons of propane, enough to get me through a fairly cold winter.

Everything looked like the day it was installed; no PV component failures. The
installer commented on running the inverters at maximum so many hours every
day. We shall have to see, if their life span is dominated by the calendar, or hours
a day at maximum, or total number of KWH produced. I have a spare.

Operational problems included a couple wiring issues. Relocating one so that an
array could be fully tilted, I managed to blow a 1A GFI fuse. I concluded that both
DC wires should be opened by the DC disconnect switch, so I added a second switch
allowing breaking both wires on each system half. A 40A circuit breaker had failed.
High line voltage here was causing false trips, though well within inverter capability.

Checking the archives revealed only my last 500 posts are accessible. So at 600+,
every one I make drops another early one. I was going to compare some of my early
system calculations & expectations with what actually happened, but guess that will
have to be from memory. Bruce Roe

[inetdog]

I also reduced start up from 5 minutes to half that. Why wait 300 sec? Bruce Roe

My guess is that the 300 second window is part of the UL listing requirement for grid tied inverters. As long as you do not care about that, you are free to change it.

[bcroe]

Trip out voltage

New experience today. My FRONIUS IG inverters have detectors to shut down if line voltage
gets out of range. However they are designed to handle up to 277VAC, so even a very high
240VAC circuit is no threat.

My line voltage has always been quite high (over 120 or 240), and pushing 58A back through
a 600' loop of 4 gauge adds about another 9V. For a year I occasionally noted a high line trip
code, which soon reset itself.

After the recent storm & outages, my line is higher than ever. An inverter would come on
and voltage went up. The second inverter would come on, and the first would trip out.
After a while the first inverter would come back on, and the second would trip out.

So for the first time I got into the Fronius setup menu. Default high line limit was 264VAC,
which is where it was lately running. Optional was up to 287VAC, I raised it to 284. It still
tripped, because there was also a neutral monitor set at half, 132VAC. That I raised to 142
and now everything seems to be fine.

On reflection I don't see any reason to even have a neutral monitor. There is no current
there, and the inverters have the option (a DELTA option) to run 240 with no neutral. Seems
to me it would be less complicated to run that mode, no neutral wiring even required.

I also reduced start up from 5 minutes to half that. Why wait 300 sec? Bruce Roe

[bcroe]

More simulation efforts

Just got my electric bill for June, with longest sun days of the year. My
inverters logged 3171 KWH after subtracting my (600' loop) line loss of 3%.
That is pretty much on target; maybe 4000 would be possible with no clouds.
THAT will never happen in NW ILL, the longest day of the year was covered
by a week of storms, seen over much of the country. 2684 KWH were exported
leaving 491 KWH used. I did run the A/C some; otherwise that would be
more like 330 KWH. So my numbers are very close to those of the power
co and design intent.

There are still 34 panels not yet deployed, need to work out positions to
boost early & late sun day production. Additional panels of any alignment
will also boost cloudy day production, currently running 30% to 70% depending
how thick the overcast is. Spotty clouds run more like 80% to 90%.

Perhaps I have found a way to use PVwatts, to simulate an hour by hour
output for each string alignment. The idea is to use different panel alignments
to get a flatter summed power curve over the sun day, and over seasons.
Adding together the hour by hour points of the different curves will give an
overall curve for good sun. Under clouds pretty much all panels come into
play, no matter what direction they face. At least that is what my clamp
on ammeter has been telling me.

The problem with PVwatts, is it includes the hour by hour effect of clouds
for the specified area, perhaps an actual recent measurement. That means
it can't be used as a design tool for good sun. HOWEVER, perhaps it
could be used by specifying an area having very little clouding. I tried
using STOVEPIPE WELLS (near DEATH VALLEY), and suddenly the curves
looked pretty good. My work has been done using curves for 19 June, 19
Sept, and 19 Dec to cover the seasons. DEATH VALLEY is farther south
than here, but I think that will mainly affect the length of sun day and sun
intensity, to a small degree. Since I want to see the RELATIVE shape of
curves (not absolute values), I think this will be pretty close.

There might be a sunny place closer to my latitude (N Cal?), but perhaps
clear skies are much more important than latitude. I'm seeing that vertical
south facing panels in Dec work almost as well as a much lower angle in
June. Maybe even better, with reflection off the snow. Not so well in Sept.
So the snow accumulation avoidance plan will be get them vertical for the
duration of serious snow. My east and west facing panels are already near
vertical, for sun near the horizon. Last year they picked up far less snow,
even though the south facing had their back to most incoming snow.

May & June banked 4308 KWH, that's 160 gallons of propane I won't be
buying next winter. Bruce Roe

[bcroe]

Really good sun day today, not perfect. Set a record of 146 KWH, or 9.73
SUN HOURs. Still trying for 10, maybe I need more east facing panels..... Bruce Roe

Perfect sun, got a record 148 KWH. But that is still only 9.86 SUN HOURS on one of the
longest days of the year. Cutting some trees would get me 10, but some belong to my
neighbors. Bruce Roe

[bcroe]

10 sun hours

If we ever have
a clear day this summer, I do hope to see a day of 150 KWH. Regardless,
it did produce a new record today, 144 KWH squeezed through 15 KW of
inverters, 9.6 SUN HOURS in a day. Bruce Roe

Really good sun day today, not perfect. Set a record of 146 KWH, or 9.73 SUN HOURs.
Still trying for 10, maybe I need more east facing panels..... Bruce Roe

[J.P.M.]

The east facing and west facing are a fixed, DOUBLE SIDED array, near vertical. That
helps minimize the additional cost. The south facing can be tilted; need to get them
vertical before snow. Bruce

Thank you.

[bcroe]

Bruce: Are your panels fixed or moveable?

The east facing and west facing are a fixed, DOUBLE SIDED array, near vertical. That
helps minimize the additional cost. The south facing can be tilted; need to get them
vertical before snow. Bruce

[J.P.M.]

As the longest day gets closer, the next 3 predicted clear days might get me that
10 SUN HOUR day. I suppose that means my inverters will only last 5 years instead
of 10, maybe can learn to fix them. I have a spare.

I don't have an account on line, the numbers appear on a pair of 7.5 KW Fronius
inverters, logged regularly in a notebook. If you click bcroe, view profile, about me,
you can find my PHOTOBUCKET link showing details in sub albums under album
ENERGY CONSERVATION.

With the most straightforward PV system located in sunny southern Cal, it might be
fine to list a system as so many KW. I prefer to list DC and AC limits. That starts
to matter when adding panel strings to compensate for the weather here in cloud
land. So a simple ratio of those numbers isn't the whole story.

I have ENOUGH AC capacity, and no way to expand it without replacing EVERYTHING.
The OBJECTIVE is to keep the inverters loaded as much of the time as possible. So
panels aimed at rising, midday, and setting sun do it, and bring mild overcast output
up to 60% to 80% of max. I have seen 4KW during a rain storm. A year will do at
least 25 Megawatt hour.

BUT if you take my DC capacity (29 KW & counting) against AC output, its a very low
number. That because the sun never hits all panels at the same time, and if it did,
there is no way to invert it all. So PVOutput using DC installation might put me in
the basement, I don't care. I'm not sure how your calculation worked.

I don't see a really good rating system for this sort of design, never heard this type
discussion before construction. But minimum panels initially were only a quarter of
the system cost, and more are a really good investment in cloud land. Bruce Roe

Bruce: Are your panels fixed or moveable ?

[bcroe]

Bruce, that is impressive. That gives the PVOutput.org definition of efficiency (kWh divided by kW DC) of approximately 7.68 kWh/kW to 8.64 kWh/kW (assuming inverter efficiency of between 80-90%) putting you at the top 1% of the efficient pv systems in PVOutput. Do you have a PVOutput account and if so, may I look at that account?

As the longest day gets closer, the next 3 predicted clear days might get me that
10 SUN HOUR day. I suppose that means my inverters will only last 5 years instead
of 10, maybe can learn to fix them. I have a spare.

I don't have an account on line, the numbers appear on a pair of 7.5 KW Fronius
inverters, logged regularly in a notebook. If you click bcroe, view profile, about me,
you can find my PHOTOBUCKET link showing details in sub albums under album
ENERGY CONSERVATION.

With the most straightforward PV system located in sunny southern Cal, it might be
fine to list a system as so many KW. I prefer to list DC and AC limits. That starts
to matter when adding panel strings to compensate for the weather here in cloud
land. So a simple ratio of those numbers isn't the whole story.

I have ENOUGH AC capacity, and no way to expand it without replacing EVERYTHING.
The OBJECTIVE is to keep the inverters loaded as much of the time as possible. So
panels aimed at rising, midday, and setting sun do it, and bring mild overcast output
up to 60% to 80% of max. I have seen 4KW during a rain storm. A year will do at
least 25 Megawatt hour.

BUT if you take my DC capacity (29 KW & counting) against AC output, its a very low
number. That because the sun never hits all panels at the same time, and if it did,
there is no way to invert it all. So PVOutput using DC installation might put me in
the basement, I don't care. I'm not sure how your calculation worked.

I don't see a really good rating system for this sort of design, never heard this type
discussion before construction. But minimum panels initially were only a quarter of
the system cost, and more are a really good investment in cloud land. Bruce Roe

[SunEagle]

Bruce, that is impressive. That gives the PVOutput.org definition of efficiency (kWh divided by kW DC) of approximately 7.68 kWh/kW to 8.64 kWh/kW (assuming inverter efficiency of between 80-90%) putting you at the top 1% of the efficient pv systems in PVOutput. Do you have a PVOutput account and if so, may I look at that account?

Of course the reason the Bruce gets some many hours of "producing sunlight" in one day is due the way he has set up his panels. One group is facing East, another South and a third West. So for most of the day at least one group is producing.

It is a very impressive system. Great job Bruce.

[gregvet]

Last year the sun was beaming down, but the power co didn't put in my net
meter till Aug. This year everything is running, but clouds just won't go away.
On Monday I saw SNOW falling????

Despite that, had mostly sunny today, spotty dark clouds. If we ever have
a clear day this summer, I do hope to see a day of 150 KWH. Regardless,
it did produce a new record today, 144 KWH squeezed through 15 KW of
inverters, 9.6 SUN HOURS in a day. Bruce Roe

Bruce, that is impressive. That gives the PVOutput.org definition of efficiency (kWh divided by kW DC) of approximately 7.68 kWh/kW to 8.64 kWh/kW (assuming inverter efficiency of between 80-90%) putting you at the top 1% of the efficient pv systems in PVOutput. Do you have a PVOutput account and if so, may I look at that account?

[bcroe]

Last year the sun was beaming down, but the power co didn't put in my net
meter till Aug. This year everything is running, but clouds just won't go away.
On Monday I saw SNOW falling????

Despite that, had mostly sunny today, spotty dark clouds. If we ever have
a clear day this summer, I do hope to see a day of 150 KWH. Regardless,
it did produce a new record today, 144 KWH squeezed through 15 KW of
inverters, 9.6 SUN HOURS in a day. Bruce Roe

[bcroe]

Yes it only took me 5 months to increase panels that way. Here in northern IL, with snow and E-W panels, Monday production calculated out to 7.8 equivalent sun hours (117 KWH). Solving the snow issues may take years, but getting all panels near vertical should help.
Bruce Roe

Decided to graph the level of my AC energy output over a sunny + cool day.
Clipping is minimal, probably will stop with warmer weather & snow melted.

This doesn't look much like the usual "bell curve". Power rises very fast as
sun hits the east facing panels. Toward mid day the main power comes from
south facing, and then transfers to west facing approaching sunset. It would
be even closer to "rectangular" if I didn't have any shading issues at
extremes of the sun day. Best day so far is 125 KWH.

The selection of non south panel orientations was chosen by guess & by golly,
no proper simulation. It was based on the panels and space I had, and what
one guy might be able to throw up in a month. For all that, it is working very
well. In addition, output in cloudy weather (light dispersed) is way up; a
typical overcast day produces 60 to 90 KWH.

There is probably room for improvement here. Need to understand some
serious simulation tools. Generally goals are to set panels to: minimize snow
accumulation; minimize clipping in good sun all year; and bring up power for
sun day beginning & end times. Using additional panels "inefficiently" to do
this, will bring up power during clouds, etc, rather common here. Bruce Roe