pianotuna
Regina, SK, Canada
Senior Member
Joined: 12/18/2004
View Profile
 Offline
|
Hi N8,
Thanks for taking time to do some testing.
The key is the bypass diodes. If the panels have them, great. If they don't, then parallel may be the better route to follow.
Because my panels are nominally 16.5 volts, have by pass diodes between each cell and there are four I chose series/parallel. I'm getting over 13 hours a day of charging this time of year (April 15, 2012 @ 49.462713,-103.007569)
N8 is farther north than I am.
Regards, Don
Kustom Koach Class C 28'5" 256 watts Unisolar, 875 amp hours in two battery banks 12 volt batteries, 2500 MSW watt inverter.
|
mena661
Southern California
Senior Member
Joined: 03/21/2009
View Profile
Offline
|
X3, thanks for the testing.
2009 Newmar Canyon Star 3205, Ford F53 V10
Trojan L16 6V's 740 Amp-hours
|
KJINTF
Montana
Senior Member
Joined: 09/04/2010
View Profile
|
N8GS
I was wondering if and when you would post the data
I'm sure it will create a lot of discussion
Thanks for the daily updates
The pictures of your very nice setup might be of some help for others
Ken
PT
I agree the integrated diodes are the key
My preference has been series connections and MPPT controllers on systems of 200 watts and greater. My reasoning was to both reduce the cost of the wire as well as to simplify the installation, hiding 4awg wire is a bit more difficult than hiding 8/10awg wire. I too use the Kyocera panels so now have another reason to go series on systems of at least 200 watts.
Always more projects than time
|
N8GS
West Michigan
Senior Member
Joined: 04/24/2006
View Profile
Good Sam RV Club Member
Offline
|
When I get home I will post the photos of the test setup.
PT- You are further north, I am at 43.1 north 85.7 west.
rvreloader- The test that I ran are all labeled on the graphs. Photos will follow. 60 cells/panel. Used foam core board taped to the panel for shade.
One test was with a 1.5" wide board at 30 degree angle across panel.
To get good consistand test results I think you should send me your panels and conroller for testing.
Salvo- As stated above Earlier tests show little difference between series and parallel. What little difference there was I could attribute to wire losses. The change over from S to P took less than 3 minutes.
CA Traveler- I don't understand why series works so well either and that is why I posted these graphs. I hope we can find some solid scientific answers. Where are the electrical engineers when we need them?
Ham radio 73's from Gale N8GS
2009 GMC 2500HD Ext cab, LB 4x4 Duramax
2006 Cougar 290EFS (RK) Axle wt 8200#, Pin wt 1760#
B&W companion
2 Honda EU2000i IOTA DLS90 converter
4-Trojan T1275's(600Ah)+470W of solar w/ Outback FM-60 CC
Samlex 2000w PSW inverter
|
pianotuna
Regina, SK, Canada
Senior Member
Joined: 12/18/2004
View Profile
Offline
|
Hi N8,
Darn, I thought you were in Edmonton--my mistake!
|
|
|
pianotuna
Regina, SK, Canada
Senior Member
Joined: 12/18/2004
View Profile
Offline
|
Hi N8,
Any chance of continuing the testing with say 3 cells totally covered, first "across" and then "lengthwise"? And what about covering an entire panel? (I'll donate a blanket *grin*).
|
renoman69
Edmonton Alberta
Senior Member
Joined: 02/09/2009
View Profile
Offline
|
Very interesting test results. I did a shade test on my panels last year and got very different results. I did not write them down but just made an observation. I cut a piece of white paper exactly the size of one cell and covered one cell while the panels were wired parallel and it cut the output almost exactly in half. When wired in series with one cell covered on one panel it basically cut the output to next to nothing. I have 2 of THESE 135w Kyocera panels with bypass diodes.
2009 Jayco Eagle Superlite 25.5RKS
2008 Silverado 2500HD Z71 4x4 Duramax/Allison
Reese 15K slider
Honda EU2000I,
270 watts of Kyocera solar
Blue Sky 3024i MPPT controller
450 AHs of Trojan power
Iota DLS-75/IQ4 converter
|
pianotuna
Regina, SK, Canada
Senior Member
Joined: 12/18/2004
View Profile
Offline
|
Hi omatty,
Pulls what? I don't understand at all what you are trying to say.
omatty wrote:in series, one panel pulls from the other.
|
CA Traveler
The Western States
Senior Member
Joined: 01/03/2004
View Profile
Offline
|
This is a diagram from http://www.pveducation.org/pvcdrom/modules/bypass-diodes.
For discussion let’s assume we have 60 cells at 30V and 6A and a bypass diode for each group of 10 cells. The shaded cell would eliminate the use of 10 cells or 5V. Plus the 0.6V drop for the green diode or 24.4V (and up to 6A) from the panel.
Nice and simple but since the 235W Kyocera has no output for a single 1 1/2 cell shaded panel they have some other design. I have no doubt that they are targeting large arrays with higher voltage serial panels and pricing of these panels reflect the growing market. ie They would have little interest in parallel panels with their higher voltage panels.
There is other interesting information on the website including the shaded cell http://www.pveducation.org/pvcdrom/modules/shading. Be sure to drag the shade rectangle over the cell to see the effect. Partial shading gets interesting as that cell can still produce some power.
2009 Holiday Rambler 42 Scepter with ISL 400 Cummins
2004 Honda CR-V
Bob
|
3 tons
CA.
Senior Member
Joined: 03/13/2009
View Profile
Offline
|
renoman69 wrote:Very interesting test results. I did a shade test on my panels last year and got very different results. I did not write them down but just made an observation. I cut a piece of white paper exactly the size of one cell and covered one cell while the panels were wired parallel and it cut the output almost exactly in half. When wired in series with one cell covered on one panel it basically cut the output to next to nothing. I have 2 of THESE 135w Kyocera panels with bypass diodes.
X2
3 tons
|
|
|
