S
Scott Willing
- Jan 1, 1970
- 0
(Dedicated to gigawatt, home power chronicler extraordinaire.)
==============
It's crazy I know.
I've had a shiny new Outback MX60 charge controller sitting on the
floor of my office for months. What kind of solar enthusiast (much
less an off-grid guy like me) could allow this to happen??
I have an excuse. A hard-core techie with a (distant now) background
in the design of related equipment, I had been determined to install
the MX60 alongside my existing Trace C40, with a convenient input
selection switch, so that I could do some A/B comparisons before
removing the Trace completely from the setup, or relegating it to a
low voltage disconnect gig.
I'm also poised to install a new Outback inverter (also holding down
the floor of my office) and will be switching from 12 to 24V primary
as part of that operation. So the plan had been to do some A/B's
between the controllers at 12V array / 12V primary, then again at 24V
array / 24V primary. Then, finally, I'd rewire my array for something
exceeding 24V to give the MX60 some decent overhead to work with, in
terms of maximum power point tracking (MPPT), and to further reduce
the current in the home run (i.e. the wire pair that takes array
current to the power room).
On top of all this was a related task: the need to turn the lengths of
aluminum angle I'd bought into the new roof mounts I'd designed almost
a year ago. They're simple enough, with only seasonal tilt adjustment,
but to help with shedding snow / clearing snow all panels will be
mounted vertically. As anyone who's ever built their own mounts knows,
this complicates things since you can't mount the panels directly to
the vertical legs; you have to add horizontal panel mount frame
members. A pain, but it'll pay off.
Meet the Bizarre Panel Crew
======================
My collection of panels is a total freak show, a real mixed bag that
was assembled piece by piece by the hardworking but minimally-funded
back-to-the-landers who hand-built our house. I knew it would be a
real challenge for the MX60, and for me to decide how to rewire too.
Briefly:
1 set QuadLams (~100W)
2 x Carrizo M51 (~25W each)
2 x Siemens M75 (48W each)
2 x Siemens SM55 (55W each)
2 x Sharp NE80 (80W each)
Hard-core readers can check out complete details of this stuff at the
bottom of the post.
Was it Lennon who said that life is what happens while you're making
other plans? It's been an exceedingly busy year, both at the job and
around the homestead. So every time I passed the power room and
thought "I should chuck that controller in there" I remembered the
issue of mounting the switch, the fact that I'd have to move the
lightning arrestor, that extra wiring would have to be added to the
already crowded Trace DC disconnect, that I hate breaking knockouts
that I know I'm going to have to plug later... and suddenly the quick
changeover wasn't. On to other priorities.
MX60 in service at last
==================
Finally last weekend, with winter knocking, no mounts built yet, and
fall chores still piled high, I said "screw it" and just took the C40
out of service and put the MX60 in its place.
I'VE BEEN HAVING FUN EVER SINCE.
What a great controller! I fully understood the advantages of this
unit before buying it, but as is often the case, you don't really
appreciate the features until you apply them and see them work.
We haven't had a day of solid sunshine since the controller went in.
Mix of sun and cloud every day, which is driving me crazy, teasing me
with glimpses of greatness.
Initial impressions at 12V array / 12V primary
====================================
First of all, though the array was still crudely wired for 12V (and by
"crudely," I mean that the wire gauges are less than ideal, wiring
layout even worse etc.) I was still able to observe occasions wherein
the controller was able to provide an amp or two higher output than
input. Considering the motley collection of stuff and the way it was
hooked together, that alone seemed like a miracle to me.
A few other initial impressions...
1. Setup was a snap.
2. I liked the fact that I didn't have to play with a screwdriver and
two voltmeter probes on a live circuit board to adjust the setpoints.
Add to that the fact that the C40 display has to hang off the unit on
a very short tether if you don't want to unplug it and lose the
accumulated data, and this is one of my least favorite things to do.
3. I liked the fact that (assuming the temp sensor is installed, of
course) you can page through the menus on the controller and see
exactly what the MX60 considers the compensated voltages to be.
4. I noticed that the MX60 has stayed much longer in the absorption
phase during these days of mixed insolation, and my nighttime battery
voltages are looking a little perkier as a result.
One of my big complaints about the C40 is that the fixed absorption
time of one hour seems most often to be either too long (especially in
summer) or too short (in winter), and that in summer I sometimes
suffer double jeopardy in that passing clouds can result in an
unnecessary second absorption cycle.
As a result I have had to make seasonal setpoint adjustments with the
C40 and sometimes this is not even sufficient. I'm always cognizant of
the fact that home power stuff tends to be too complicated for the
average non-technical homeowner, and things like this don't help.
It's a bit early to be making this assessment, but I have a feeling
the MX60 -- quite apart from maximizing output from my array -- will
do a better job of charging, period.
Design for a higher voltage array
===========================
But on to rewiring the array. I was dying to try some higher voltage
arrangement just to see it work. Especially, I was curious to find out
if it would offer enough of an advantage, by wiring higher-voltage
strings, to offset some of the theoretical lost potential re: rough
matching of panel currents in the series strings.
I played around for a while and came up with a scheme to try. First, I
would cobble together six 12V panels averaging 5A each, as follows:
1. 1 set QuadLams
2. 1 x NE80
3. 1 x NE80
4. 1 x SM55 + 1 x M51 in parallel
5. 1 x SM55 + 1 x M51 in parallel
6. 2 x M75 in parallel
(If this looks utterly insane, I'll just have to refer you to the full
panel specs at the end of the post. If you have better ideas, I'm all
ears.)
Since the MX60 maximum input voltage is 120V, I can't put these 6
panels / panel combos together in a single string of nominally 72V @
5A, much as I'd like to. Open circuit voltage could easily exceed the
limit of 120V, especially when you consider that -40 degrees is not at
all uncommon in the winter here.
So I thought I'd go with two strings of three, i.e. nominal 36V. This
would still give the controller some decent overhead to work with when
I change the primary over to 24V.
After playing with the estimated and/or specified performance figures,
I decided to put the lowest-voltage panels (M75's) in the same string
as the highest-voltage panels (QuadLams) in a crude attempt to balance
off overall maximum power point voltage of each string. So the final
plan would go like this:
String 1 String 2
SM55+M51 SM55+M51
NE80 QuadLams
NE80 2 x M75
A Quick Trial at 36V
================
For convenience of a quick trial and due to the physical arrangement
and the existing wiring, I first decided to try one string temporarily
built out of "the weird sisters":
SM55+M51
SM55+M51
2 x M75
It actually worked, and far better than I expected. What a hoot!
When's Winter Coming?
====================
Before I even wrote down any numbers from the 36V trial, I had a
sobering thought. As I had been working, the day had gone from sunny
and quite pleasant to cold and increasingly cloudy, with the distinct
tinge of winter in the air. You must understand that we've had winter
arrive quite suddenly here, earlier in the year than this, and stay,
freezing fall jobs literally in their tracks.
I'd just disabled half of my available solar input power. What if I
never managed to take the next step before it was too late for this
year? Oh-oh...
A Quick Leap to 60V
=================
But wait, this controller was a veritable Swiss Army knife. Not really
wanting to mess with the QuadLams at the moment anyway, I grafted a
couple of MC plugs into the string on the ground mount, so that I
could lean the NE80's up against a bench temporarily and add them to
the series string. This gives me a nominal 60V string at about 5A,
with a safe Voc of 110V worst case. That's probably the way it's going
to stay for the week at least.
That was Saturday. Today (Sunday) I had to turn to another fall task.
An ugly business to be sure, but if I don't address a leaky sagging
roof in the trailer that serves as a storage shed / rough workshop
before the snow hits it again, I could have a full on collapse before
spring. So, until the light failed altogether, I was out in the
trailer dealing with that nightmare, running back to the house
periodically when there was a break in the clouds to see how the MX60
was doing with my nutball panel collection.
It's looking incredibly good!
It's facinating to watch the controller work, e.g. periodically
checking Voc and sweeping for a new MPPT point. To be able to initiate
a sweep manually is very cool as well, especially when you have to get
back to that crappy trailer job and the sun is playing peekaboo.
How Many Watts?
===============
For a few minutes at one point - cut short by yet another cloud - I
saw 32A going into the batteries at 14.7V! Shook my head and
double-checked the readings against the Trimetric. It was no joke, 32A
@ 14.7V = 470W. This is actually more than the estimated, standard
temperature, net power output of this set of panels, despite their
being connected in this wacky arrangement, and at an assortment of
sub-optimal angles to boot. Yes, it *was* close to solar noon, but I
considered this performance to be miraculous.
When the next cloud came along I went outside and yelled something
unrepeatable at the sky. (Didn't work.)
Now, I've freely admitted that the prior wiring of the array was less
than ideal, folks, but the present wiring is not that much sexier, and
I had NEVER seen 32A going into my batteries from this array under any
circumstance. And at that voltage?! HA! Maybe 2/3 of that at most.
YEOW!
And then I remembered that the mondo-funky QuadLams weren't even part
of the picture. Holy crap... I was getting *far* more power into the
batts than ever before, with one panel tied behind my back.
Da Future
========
The next couple days are supposed to be sunny. I'll be running back
and forth checking the MX60 display to be sure. Maybe I should
temporarily wire the Mate remote display / control panel over to my
office.
Some day the MX60 will be fed by a nice consistent array of identical
panels, just like normal folks do it out there in the civilized world.
Rarely, I would think, will it have to deal with such a bizarre
collection of panels, but I have to say that it's a credit to the
folks at Outback that it can, and do so elegantly.
I doubt that I'll ever get to do the back-to-back comparison with the
C40 now, but to say that there's no comparison anyway is, at this
point, abundantly superfluous.
The Swiss Army Knife Thing, Again
============================
The ability to reconfigure my array as needed, alone, is a real treat.
For example, while working on my mounts, I'll be able to borrow panels
out of the setup to double-check measurements etc. as required, wiring
up whatever is left behind without even having to reprogram the
controller.
When I get around to adding a couple of new NE80's to the setup, I can
go to two strings of 48V, with the NE80's in one string, and the
QuadLams and all the weird sisters in the other string. Very cool.
Apologies
========
Sorry about the length of the post, but I'm somewhat excited eh?
I'd better sleep on this, review it tomorrow and see if it makes any
sense before I post it.
-=s
The PS: Complete Details of the Ugly Array
==================================
1 set Carrizo QuadLams.
Voc: 26V
Isc: 5.6A
Vpmax: 19V
Ipmax: 5.2A
The copper QuadLams are a set of four surplus frameless laminates.
Each panel looks like a brown SM55; the 36 square cells are arranged
as three rows of twelve. The laminates have been mounted in a 2x4
wooden frame that has somehow withstood the elements quite nicely for
quite a few years now. I think they were the first panels installed
here.
For the longest time I thought that they were simply four 12V panels;
only last year did I realize that in each panel, the strings of 12
cells are paralleled; that is, each panel is 3V(!!) and the four are
wired in series to create one 12V nominal panel of about 100W.
The meaning of "Quad" suddenly became clear. My heart sank a bit. All
the other panels I have are in pairs, and I had been expecting to be
able to rewire the QuadLam panels into a 24 or 48V arrangement. Drat!
Not without some additional surgery... possible, but not terribly
attractive.
[BTW, there's a Home Power Magazine article (#39, Feb/March 1994) on
my CD collection of back issues. From the Voc and Isc tests I've done,
my set seems to closely match the set tested by Sam Coleman ten years
ago, though my measurements were made at an ambient temp of around 10
degrees C.]
===
2 x Carrizo M51 (aka Arco M51)
35 round cells, 2 x 12 + 1 x 11
Voc: 22-ish V
Isc: 1.8-ish A
Vpmax: NA, 17V?
Ipmax: NA, 1.5A?
More surplus stuff. From what the original owner said, confirmed by
one web reference I ran across, it appears that these are refugees
from some concentrator experiments. Originally a 50-ish W panel, they
are a bit shell-shocked, and good for about 1.5A each under load.
(25W?)
===
2 x Siemens M75
33 square cells, 3 x 11
Voc: 20V
Isc: 3.4A
Vpmax: 16V
Ipmax: 3.0A
These were purchased by the original owner as part of an RV solar
power kit. The lower voltage presents a bit of a challenge to
integration with the other panels. I can't tell you how often I've
looked up at those things and tried to imagine 3 more cells on each.
===
2 x Siemens SM55
36 square cells, 3 x 12
Voc: 22V
Isc: 3.5A
Vpmax: 17V
Ipmax: 3.2A
These were the newest panels of the collection when I arrived.
===
2 x Sharp NE80
36 square cells, 4 x 9
Voc: 21V
Isc: 5.3A
Vpmax: 17V
Ipmax: 4.7A
I bought a pair of these last year from a relatively local source.
This is the panel I intend to standardize on. This pair spent most of
the winter leaning against the angled glass of our sunroom, crudely
wired in parallel with the rest.
===
At present, the QuadLams and the M75's are on the roof of our
sunroom/greenhouse. The M51's and SM55's are on a home-brew, manually
pan/tiltable ground mount built by the previous owner's son as a shop
project.
I want to get all of these panels up on the roof. Whatever advantage
the ground mount gives in terms of being able to aim the panels is
somewhat offset by the fact that the panels on the roof get the light
earlier down here in our little valley. And, the challenge here (50
degrees N.) is the winter, where aiming is of little advantage anyway.
Each of the two mounts I'm building are designed to take four NE-80's.
With the addition of a few extra holes, one mount will accept "the
weird sisters" i.e. the M51's, M75's and SM55's, and the other will
take the two NE-80's with space waiting for the next pair. Eventually
the wierd panels will be assigned elsewhere and replaced with more
NE-80's, but for now I can't be choosy. As long as they're producing
power I have to try to use them.
All of this stuff will just barely fit across the roof of the
sunroom/greenhouse when complete.
=======
==============
It's crazy I know.
I've had a shiny new Outback MX60 charge controller sitting on the
floor of my office for months. What kind of solar enthusiast (much
less an off-grid guy like me) could allow this to happen??
I have an excuse. A hard-core techie with a (distant now) background
in the design of related equipment, I had been determined to install
the MX60 alongside my existing Trace C40, with a convenient input
selection switch, so that I could do some A/B comparisons before
removing the Trace completely from the setup, or relegating it to a
low voltage disconnect gig.
I'm also poised to install a new Outback inverter (also holding down
the floor of my office) and will be switching from 12 to 24V primary
as part of that operation. So the plan had been to do some A/B's
between the controllers at 12V array / 12V primary, then again at 24V
array / 24V primary. Then, finally, I'd rewire my array for something
exceeding 24V to give the MX60 some decent overhead to work with, in
terms of maximum power point tracking (MPPT), and to further reduce
the current in the home run (i.e. the wire pair that takes array
current to the power room).
On top of all this was a related task: the need to turn the lengths of
aluminum angle I'd bought into the new roof mounts I'd designed almost
a year ago. They're simple enough, with only seasonal tilt adjustment,
but to help with shedding snow / clearing snow all panels will be
mounted vertically. As anyone who's ever built their own mounts knows,
this complicates things since you can't mount the panels directly to
the vertical legs; you have to add horizontal panel mount frame
members. A pain, but it'll pay off.
Meet the Bizarre Panel Crew
======================
My collection of panels is a total freak show, a real mixed bag that
was assembled piece by piece by the hardworking but minimally-funded
back-to-the-landers who hand-built our house. I knew it would be a
real challenge for the MX60, and for me to decide how to rewire too.
Briefly:
1 set QuadLams (~100W)
2 x Carrizo M51 (~25W each)
2 x Siemens M75 (48W each)
2 x Siemens SM55 (55W each)
2 x Sharp NE80 (80W each)
Hard-core readers can check out complete details of this stuff at the
bottom of the post.
Was it Lennon who said that life is what happens while you're making
other plans? It's been an exceedingly busy year, both at the job and
around the homestead. So every time I passed the power room and
thought "I should chuck that controller in there" I remembered the
issue of mounting the switch, the fact that I'd have to move the
lightning arrestor, that extra wiring would have to be added to the
already crowded Trace DC disconnect, that I hate breaking knockouts
that I know I'm going to have to plug later... and suddenly the quick
changeover wasn't. On to other priorities.
MX60 in service at last
==================
Finally last weekend, with winter knocking, no mounts built yet, and
fall chores still piled high, I said "screw it" and just took the C40
out of service and put the MX60 in its place.
I'VE BEEN HAVING FUN EVER SINCE.
What a great controller! I fully understood the advantages of this
unit before buying it, but as is often the case, you don't really
appreciate the features until you apply them and see them work.
We haven't had a day of solid sunshine since the controller went in.
Mix of sun and cloud every day, which is driving me crazy, teasing me
with glimpses of greatness.
Initial impressions at 12V array / 12V primary
====================================
First of all, though the array was still crudely wired for 12V (and by
"crudely," I mean that the wire gauges are less than ideal, wiring
layout even worse etc.) I was still able to observe occasions wherein
the controller was able to provide an amp or two higher output than
input. Considering the motley collection of stuff and the way it was
hooked together, that alone seemed like a miracle to me.
A few other initial impressions...
1. Setup was a snap.
2. I liked the fact that I didn't have to play with a screwdriver and
two voltmeter probes on a live circuit board to adjust the setpoints.
Add to that the fact that the C40 display has to hang off the unit on
a very short tether if you don't want to unplug it and lose the
accumulated data, and this is one of my least favorite things to do.
3. I liked the fact that (assuming the temp sensor is installed, of
course) you can page through the menus on the controller and see
exactly what the MX60 considers the compensated voltages to be.
4. I noticed that the MX60 has stayed much longer in the absorption
phase during these days of mixed insolation, and my nighttime battery
voltages are looking a little perkier as a result.
One of my big complaints about the C40 is that the fixed absorption
time of one hour seems most often to be either too long (especially in
summer) or too short (in winter), and that in summer I sometimes
suffer double jeopardy in that passing clouds can result in an
unnecessary second absorption cycle.
As a result I have had to make seasonal setpoint adjustments with the
C40 and sometimes this is not even sufficient. I'm always cognizant of
the fact that home power stuff tends to be too complicated for the
average non-technical homeowner, and things like this don't help.
It's a bit early to be making this assessment, but I have a feeling
the MX60 -- quite apart from maximizing output from my array -- will
do a better job of charging, period.
Design for a higher voltage array
===========================
But on to rewiring the array. I was dying to try some higher voltage
arrangement just to see it work. Especially, I was curious to find out
if it would offer enough of an advantage, by wiring higher-voltage
strings, to offset some of the theoretical lost potential re: rough
matching of panel currents in the series strings.
I played around for a while and came up with a scheme to try. First, I
would cobble together six 12V panels averaging 5A each, as follows:
1. 1 set QuadLams
2. 1 x NE80
3. 1 x NE80
4. 1 x SM55 + 1 x M51 in parallel
5. 1 x SM55 + 1 x M51 in parallel
6. 2 x M75 in parallel
(If this looks utterly insane, I'll just have to refer you to the full
panel specs at the end of the post. If you have better ideas, I'm all
ears.)
Since the MX60 maximum input voltage is 120V, I can't put these 6
panels / panel combos together in a single string of nominally 72V @
5A, much as I'd like to. Open circuit voltage could easily exceed the
limit of 120V, especially when you consider that -40 degrees is not at
all uncommon in the winter here.
So I thought I'd go with two strings of three, i.e. nominal 36V. This
would still give the controller some decent overhead to work with when
I change the primary over to 24V.
After playing with the estimated and/or specified performance figures,
I decided to put the lowest-voltage panels (M75's) in the same string
as the highest-voltage panels (QuadLams) in a crude attempt to balance
off overall maximum power point voltage of each string. So the final
plan would go like this:
String 1 String 2
SM55+M51 SM55+M51
NE80 QuadLams
NE80 2 x M75
A Quick Trial at 36V
================
For convenience of a quick trial and due to the physical arrangement
and the existing wiring, I first decided to try one string temporarily
built out of "the weird sisters":
SM55+M51
SM55+M51
2 x M75
It actually worked, and far better than I expected. What a hoot!
When's Winter Coming?
====================
Before I even wrote down any numbers from the 36V trial, I had a
sobering thought. As I had been working, the day had gone from sunny
and quite pleasant to cold and increasingly cloudy, with the distinct
tinge of winter in the air. You must understand that we've had winter
arrive quite suddenly here, earlier in the year than this, and stay,
freezing fall jobs literally in their tracks.
I'd just disabled half of my available solar input power. What if I
never managed to take the next step before it was too late for this
year? Oh-oh...
A Quick Leap to 60V
=================
But wait, this controller was a veritable Swiss Army knife. Not really
wanting to mess with the QuadLams at the moment anyway, I grafted a
couple of MC plugs into the string on the ground mount, so that I
could lean the NE80's up against a bench temporarily and add them to
the series string. This gives me a nominal 60V string at about 5A,
with a safe Voc of 110V worst case. That's probably the way it's going
to stay for the week at least.
That was Saturday. Today (Sunday) I had to turn to another fall task.
An ugly business to be sure, but if I don't address a leaky sagging
roof in the trailer that serves as a storage shed / rough workshop
before the snow hits it again, I could have a full on collapse before
spring. So, until the light failed altogether, I was out in the
trailer dealing with that nightmare, running back to the house
periodically when there was a break in the clouds to see how the MX60
was doing with my nutball panel collection.
It's looking incredibly good!
It's facinating to watch the controller work, e.g. periodically
checking Voc and sweeping for a new MPPT point. To be able to initiate
a sweep manually is very cool as well, especially when you have to get
back to that crappy trailer job and the sun is playing peekaboo.
How Many Watts?
===============
For a few minutes at one point - cut short by yet another cloud - I
saw 32A going into the batteries at 14.7V! Shook my head and
double-checked the readings against the Trimetric. It was no joke, 32A
@ 14.7V = 470W. This is actually more than the estimated, standard
temperature, net power output of this set of panels, despite their
being connected in this wacky arrangement, and at an assortment of
sub-optimal angles to boot. Yes, it *was* close to solar noon, but I
considered this performance to be miraculous.
When the next cloud came along I went outside and yelled something
unrepeatable at the sky. (Didn't work.)
Now, I've freely admitted that the prior wiring of the array was less
than ideal, folks, but the present wiring is not that much sexier, and
I had NEVER seen 32A going into my batteries from this array under any
circumstance. And at that voltage?! HA! Maybe 2/3 of that at most.
YEOW!
And then I remembered that the mondo-funky QuadLams weren't even part
of the picture. Holy crap... I was getting *far* more power into the
batts than ever before, with one panel tied behind my back.
Da Future
========
The next couple days are supposed to be sunny. I'll be running back
and forth checking the MX60 display to be sure. Maybe I should
temporarily wire the Mate remote display / control panel over to my
office.
Some day the MX60 will be fed by a nice consistent array of identical
panels, just like normal folks do it out there in the civilized world.
Rarely, I would think, will it have to deal with such a bizarre
collection of panels, but I have to say that it's a credit to the
folks at Outback that it can, and do so elegantly.
I doubt that I'll ever get to do the back-to-back comparison with the
C40 now, but to say that there's no comparison anyway is, at this
point, abundantly superfluous.
The Swiss Army Knife Thing, Again
============================
The ability to reconfigure my array as needed, alone, is a real treat.
For example, while working on my mounts, I'll be able to borrow panels
out of the setup to double-check measurements etc. as required, wiring
up whatever is left behind without even having to reprogram the
controller.
When I get around to adding a couple of new NE80's to the setup, I can
go to two strings of 48V, with the NE80's in one string, and the
QuadLams and all the weird sisters in the other string. Very cool.
Apologies
========
Sorry about the length of the post, but I'm somewhat excited eh?
I'd better sleep on this, review it tomorrow and see if it makes any
sense before I post it.
-=s
The PS: Complete Details of the Ugly Array
==================================
1 set Carrizo QuadLams.
Voc: 26V
Isc: 5.6A
Vpmax: 19V
Ipmax: 5.2A
The copper QuadLams are a set of four surplus frameless laminates.
Each panel looks like a brown SM55; the 36 square cells are arranged
as three rows of twelve. The laminates have been mounted in a 2x4
wooden frame that has somehow withstood the elements quite nicely for
quite a few years now. I think they were the first panels installed
here.
For the longest time I thought that they were simply four 12V panels;
only last year did I realize that in each panel, the strings of 12
cells are paralleled; that is, each panel is 3V(!!) and the four are
wired in series to create one 12V nominal panel of about 100W.
The meaning of "Quad" suddenly became clear. My heart sank a bit. All
the other panels I have are in pairs, and I had been expecting to be
able to rewire the QuadLam panels into a 24 or 48V arrangement. Drat!
Not without some additional surgery... possible, but not terribly
attractive.
[BTW, there's a Home Power Magazine article (#39, Feb/March 1994) on
my CD collection of back issues. From the Voc and Isc tests I've done,
my set seems to closely match the set tested by Sam Coleman ten years
ago, though my measurements were made at an ambient temp of around 10
degrees C.]
===
2 x Carrizo M51 (aka Arco M51)
35 round cells, 2 x 12 + 1 x 11
Voc: 22-ish V
Isc: 1.8-ish A
Vpmax: NA, 17V?
Ipmax: NA, 1.5A?
More surplus stuff. From what the original owner said, confirmed by
one web reference I ran across, it appears that these are refugees
from some concentrator experiments. Originally a 50-ish W panel, they
are a bit shell-shocked, and good for about 1.5A each under load.
(25W?)
===
2 x Siemens M75
33 square cells, 3 x 11
Voc: 20V
Isc: 3.4A
Vpmax: 16V
Ipmax: 3.0A
These were purchased by the original owner as part of an RV solar
power kit. The lower voltage presents a bit of a challenge to
integration with the other panels. I can't tell you how often I've
looked up at those things and tried to imagine 3 more cells on each.
===
2 x Siemens SM55
36 square cells, 3 x 12
Voc: 22V
Isc: 3.5A
Vpmax: 17V
Ipmax: 3.2A
These were the newest panels of the collection when I arrived.
===
2 x Sharp NE80
36 square cells, 4 x 9
Voc: 21V
Isc: 5.3A
Vpmax: 17V
Ipmax: 4.7A
I bought a pair of these last year from a relatively local source.
This is the panel I intend to standardize on. This pair spent most of
the winter leaning against the angled glass of our sunroom, crudely
wired in parallel with the rest.
===
At present, the QuadLams and the M75's are on the roof of our
sunroom/greenhouse. The M51's and SM55's are on a home-brew, manually
pan/tiltable ground mount built by the previous owner's son as a shop
project.
I want to get all of these panels up on the roof. Whatever advantage
the ground mount gives in terms of being able to aim the panels is
somewhat offset by the fact that the panels on the roof get the light
earlier down here in our little valley. And, the challenge here (50
degrees N.) is the winter, where aiming is of little advantage anyway.
Each of the two mounts I'm building are designed to take four NE-80's.
With the addition of a few extra holes, one mount will accept "the
weird sisters" i.e. the M51's, M75's and SM55's, and the other will
take the two NE-80's with space waiting for the next pair. Eventually
the wierd panels will be assigned elsewhere and replaced with more
NE-80's, but for now I can't be choosy. As long as they're producing
power I have to try to use them.
All of this stuff will just barely fit across the roof of the
sunroom/greenhouse when complete.
=======
