View Full Version : Fix solar wired with a Tracking Solar system
January 21st, 2009, 21:53 PST
Currently I am running ten PhotoWatt 95watt panels wired in five series pairs providing on average 62 to 68 vots at about 11 amps. The Vmp is 34volts per panel and Imp is 2.9 amps. When I turn the breakers off at the combiner panel I read about 72 volts on each pair of panels. When I turn the breakers on and they connect to the MX60 it drops to about 64 volts. We have installed a Zome Tracker unit that will have eight BP585 panels at 85 watts each the Vmp is 18 volts and Imp if 4.7 amps per panel. I am thinking about wiring them in two pairs of four serie wired panels giving me about 72 volts Vmp per string at 4,7 amps. I would like to connect the two series strings in parallel with the PhotoWatt fixed panels. Not sure how this arrangment will behave, with the fix panels working best between 10AM and 2PM and the tracking panels working best between 9AM and 3PM. Anyone have any suggestions? I was thinking I could add one more panel and wire three series BP585 and get 54 Vmp but I think the 72 Vmp would better match the PhotoWatt panels. arcandspark
January 22nd, 2009, 6:08 PST
While the collection of panels fits (total wattage-wise) with a single MX60 at 24V, the differing panel voltages (and track .vs. nontrack) strongly suggest a second controller.
However, the mppt will manage what it can - it's just that you might be throwing away enough power that a second charge controller will be cost effective. I would agree that that the 4 stack will probably work with the other panels better than the 3 stack. But if you're willing to consider spending more for another panel, perhaps you'd be willing to spend more for another CC, so you'll get the most out of each set of panels.
January 22nd, 2009, 7:46 PST
As soon as money allows we plan to add another controller. But for now we need some extra wattage and I figured this set up might work and provide a few extra watts. Thanks for the advise. arcandspark
January 22nd, 2009, 7:50 PST
In that case, just muddle along with it - the MPPT will get the best that can be got out of the combination while it's a combination, and when you have funds to split them off to their own controllers, you'll likely get more out of them. In point of fact, I'd be happy to see you collect data before and after and post it here when/if you get to that point.
January 22nd, 2009, 8:08 PST
I will post as much data as I can, currently I have a chart recorder hooked up to the solar panels to collect some voltage output data from the panels over a week of two time span. I am also going to put it on the output side of the MX60 to record current output over a week or two. arcandspark
January 22nd, 2009, 8:08 PST
I merged the two threads--easier to keep all responses in one location.
January 22nd, 2009, 15:46 PST
Was not said nor implied, you should consider blocking diodes on each array. This is to help for in the eary morning and late evening when the tracker array is outputting power and the fixed is not.
Maybe a non-issue, but hate to lose the added power if it is getting sucked up the the fixed array.
January 22nd, 2009, 16:47 PST
Ken, excellent suggestion, I completely forgot about this, but you are right. That is a great way to separate the two systems and keep from causing any kind of problems. I will dig into my junk box and find some diodes that will handle the current and provide low voltage drop. Thanks again, that is why these fourms are so great, cant always remember everything. arcandspark
January 22nd, 2009, 17:12 PST
I am not sure that diodes are really needed... Once there is any "sky shine"--the Vmp/Voc should be very near operational range (although, with very little available current).
Diodes themselves have losses (0.2-1.0+ volt drop. DiodePowerLoss=Vdiode*Idiode)... Not the end of the world--but may not gain you anything useful either (other than more connections and components that can fail--diodes may need insulated heat sinks too).
January 22nd, 2009, 18:05 PST
I will use a SCHOTTKY diode to provide low voltage drop. I have hundreds of heatsinks in my garge. Been working in electronics feild for over twenty years. For me this little addition in wiring will be a peice of cake. By adding the diode to the Fixed solar panels combiner box connection, it will solve all my problems. arcandspark
January 22nd, 2009, 18:27 PST
May want to do a little test. Using a amp meter in-line on fix array, cover the fix array with something and see if or how much current is feed into the fix array.
The Unisolar panels I have, did not need a diode on either of my two arrays.
January 22nd, 2009, 20:03 PST
If the heat sink is electrically connected to the diode--You might want to install the diode on the negative lead (assuming negative ground) to lessen any "surprises".
February 6th, 2009, 10:46 PST
I found four Shotkey (mis spelled) diodes 100 volt 100 amps dropping about .3 to .4 volts for turn on loss on ebay for under $25.00.
Ken I will do some testing to see if there is any feedback into the fixed panels before installing the diode. Hope to pass my findings on to the group.
Bill, I know the heat sink will need to be isolated from anything that is (-) grounded. I have isolation mounts for that and will add sheild guards. I am figuring ten 90 watt panels thats 900 watts of heat to size the heat sink, ma I correct with that figure?
Thanks for the feedback, arcandspark
Check some of my pics at: http://alternativeenergy.meetup.com/153/photos/126202/#5674437
February 6th, 2009, 11:03 PST
Watch the Vrev of the Diodes vs the Voc of the array... The maximum for a MX60 is 150 VDC (IIRC)--over the 100 volt rating of the diodes.
In this case, the diodes would probably fail shorted or open--so there is no danger if they fail. Just something to watch for.
Regarding your connections?
Amps = Watts / Volts
900 watt / (5 strings * 64 volts) = 2.8 amps per string max
Power = I * V (for a single diode per string heat generation)
2.8 amps * 0.4 volts = 1.12 watts drop per string/single diode.
Power loss for all strings through diodes:
1.12 watts * 5 strings = 5.6 watts drop in full sun (900 watts / 60 VDC battery bank as described).
PS: It would be interesting to see two results/charts from your experiment:
One is the amount of reverse current vs voltage you see in the "off-sun" panels (morning evening).
And, what is the diode drop (forward or reverse voltage recorded) on the "off sun" strings.
February 6th, 2009, 14:58 PST
OK I might need some clarificatioin. The fix panel system has five pairs of series panels each panel is 43 Voc 3 amps. I have seen 75 volts at the combiner box main terinal, and I have seen no higher than 68 volts 11 amps at the input reading of the MX60. I was going to put the diode on the main output terminal to the MX60, not a diode on each strings. There is already diodes on each of the panels in the junction box. The BP585's that will be on the Zome tracker have 22 Voc @5amps. I was going to wire them in two strings of four seies wired panels. Should be close to 72 volts at almost 10 amps. I thought it would be better to install the diode on the fixed panels since the tracker will see direct sun light from say 9:30 to 3:30 and the fixed will see direct sunlight from 10:30 to 2:30, just guessing until I colect more chart recorder info. Help me out if I am not on track, I would appreciate the lesson. Not sure about the Vres of the diode but I will look it up, I figure the BP tracking panels could have a max of 88 Voc are my calculations correct? arcandspark
February 6th, 2009, 15:06 PST
Sorry, I made a typo--I intended to type Vreverse of the diode (maximum safe "backwards" voltage before the diode breaks down).
I am going out for a bit--I will have to read your post in detail when I return.
February 6th, 2009, 15:48 PST
You are thinking correctly by putting the (1) diode on the fixed array after the combiner box.
February 6th, 2009, 19:12 PST
Ok found the Diode Data Specs for the IR 203CNQ100 diode.
Voltage DC reverse (Max) 100Volts
I could also maybe go with five separate shottky diodes one on each series pair of panels. They would be smaller easier to install in the combiner box and they could all connect to a common heat sink that would be hard wired to the Positive power terminal block. The heat sink would be small enough to handle roughly 8 to 10 watts of hear. Just a thought. Right now I have four of these large dual diode common cathode units. Tomarrow I go to work on my friends system, pulling 2/0 cable from the zome tracker unit to the fixed combiner box. Running several under ground copper grounding strings with ground bar placed every ten to twelve feet apart. I plan to install those porqupine type needle sphere static disapators on the solar frame assemblies. I grew up in Florida, the lightning capital of the USA. Florida Power and Light is using the new style Porqupine Static Disapators on most all their power systems. I fould a way to make them cheap using heavy stainless steel welding rods and some 4/0 crimp conectors. I hope to get some pictures up soon. Again thenks for the input. arcandspark
February 6th, 2009, 20:09 PST
The Vrev is going to be the difference between the two array's. This will only be close to max at the very early morning and late evenings. Then at that time, I dought the fix array will be above 50-60 VDC.
February 7th, 2009, 11:52 PST
I could also maybe go with five separate shottky diodes one on each series pair of panels. They would be smaller easier to install in the combiner box and they could all connect to a common heat sink that would be hard wired to the Positive power terminal block. The heat sink would be small enough to handle roughly 8 to 10 watts of hear.
1) Schottky diodes have lower Vf at lower amps, so that's a good idea, 1 for each string.
2) Adding them all to a common heatsink, not bad, but you WILL HAVE to isolate them so you do not have a common connection. UNLESS you are sure they can electrically share a common point. I don't have a schematic of your setup. If they can be connected, DO NOT rely on the connection to the heatsink to carry the power, use a terminal ring on them. Also use a spring washer to keep even tension on the nut/bolts. After several years of hot/cold cycles, you will know if you did not.
3) 10W of heat sink inside a combiner box. You are putting a heater inside a oven. BAD idea, unless you have a way to keep it cool. The diodes, when powered, need to be kept below 70C on the hottest day, and a rooftop combiner box likely gets that hot just siting there. The cooler the diodes, the longer their lifetime.
I'm looking at this version of them:
(WOW - they claim up to 175C operating !!) so @ 10A, you are looking at .4V drop, not bad.
Powered by vBulletin® Version 4.1.10 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.