Renewable Energy Questions/Discussion > Solar (heating or electric)
Intro and questions, if I may
ChrisOlson:
--- Quote from: ksouers on February 01, 2012, 06:14:14 pm ---My main concern was the mix of different sized panels. I've never run panels in parallel before, on the boat each battery had it's own panel charging it while it was off line. I'm concerned about the wide disparity between these panels causing some damage.
--- End quote ---
Kevin, the different size panels don't matter a bit in parallel. If you are using a MPPT controller with panels hooked in series, then you will want/need matched panels.
--
Chris
rossw:
--- Quote from: ksouers on February 01, 2012, 06:14:14 pm ---I don't have diodes on the individual panels. I've read where people recommend they be used, but I've also read reliable sources that say they are not necessary. I just don't know how much, if any, a lighted panel will back feed into a shaded panel. And yes, I do have that situation at my site with trees casting shadows in the morning.
--- End quote ---
I would have agreed, once upon a time.
I have a moderate PV install. 6 arrays each of 600 watts. 4 of them on trackers.
A horrible little tale for you. When I installed the last array, it was about mid-day that I was ready to go. Just like the other arrays, as I got them done, I plugged each array in in parallel with the others. Each other time this was usually late afternoon. This time it was peak production. Well, I plugged in the cable and..... there was crackling and snapping and popping all around me.
Every panel (6) on this last array had to come off. Inside their J-boxes looked like this:
And a handfull of little rattly bits fell out, like these. Do you recognise them?
At the time, I was gutted. I was lucky, the company I purchased the panels from (in china) couldn't help me enough. Concerned and wanted to send me replacement panels, was delighted that I thought it was just the diodes and sent me replacements by overnight international courier.
All the panels came back working once I'd replaced the diodes (thankfully). It is my belief that it was the instant connection of around 120V DC to the array that caused a cascade/avalanche in the schottky diodes, allowing them to pass very substantial current. (The rest of the arrays were producing around 2.8kW at the time).
Since then, I added (as I should have in the first place) - series schottky diodes for each array.
I bolted each of them to a copper busbar for both electrical interconnect, and thermal dissipation.
The thin wire on the leads is very short. It is thin so it doesn't have a tendancy to break the leads off the diodes, and it will make a fusible link in the event of some catastrophic failure.
I now get zero current flowing back into any array, regardless of which ones are in shaddow (clouds moving past, arrays shading each other, etc), and I will never get another incident like this and blow up a panel.
Watt:
--- Quote from: ChrisOlson on February 01, 2012, 06:20:46 pm ---
--- Quote from: ksouers on February 01, 2012, 06:14:14 pm ---My main concern was the mix of different sized panels. I've never run panels in parallel before, on the boat each battery had it's own panel charging it while it was off line. I'm concerned about the wide disparity between these panels causing some damage.
--- End quote ---
Kevin, the different size panels don't matter a bit in parallel. If you are using a MPPT controller with panels hooked in series, then you will want/need matched panels.
--
Chris
--- End quote ---
I would have to be careful with this myself.
The following numbers are just examples not of actual panel output.
If that mppt controller is holding the mppt voltage at say 52 volts due to the Vmp and VmI of one panel which is a better producer and the second panel in parallel is hooked to this same controller with Vmp and VmI 4 volts less than 52 volts, ( Say bulk charge for 48v battery group) that one panel will not be producing as much power into the system ( if any ). Will that work, yes but with limitations. The two in series, if the VmI is very close between the two, series may be a better option. If the current is way different and the Vmp is too far off for parallel, two mppt controllers may be required or three or four or just direct connect.
Watt:
--- Quote from: rossw on February 01, 2012, 06:50:46 pm ---
--- Quote from: ksouers on February 01, 2012, 06:14:14 pm ---I don't have diodes on the individual panels. I've read where people recommend they be used, but I've also read reliable sources that say they are not necessary. I just don't know how much, if any, a lighted panel will back feed into a shaded panel. And yes, I do have that situation at my site with trees casting shadows in the morning.
--- End quote ---
I would have agreed, once upon a time.
I have a moderate PV install. 6 arrays each of 600 watts. 4 of them on trackers.
A horrible little tale for you. When I installed the last array, it was about mid-day that I was ready to go. Just like the other arrays, as I got them done, I plugged each array in in parallel with the others. Each other time this was usually late afternoon. This time it was peak production. Well, I plugged in the cable and..... there was crackling and snapping and popping all around me.
Every panel (6) on this last array had to come off. Inside their J-boxes looked like this:
And a handfull of little rattly bits fell out, like these. Do you recognise them?
At the time, I was gutted. I was lucky, the company I purchased the panels from (in china) couldn't help me enough. Concerned and wanted to send me replacement panels, was delighted that I thought it was just the diodes and sent me replacements by overnight international courier.
All the panels came back working once I'd replaced the diodes (thankfully). It is my belief that it was the instant connection of around 120V DC to the array that caused a cascade/avalanche in the schottky diodes, allowing them to pass very substantial current. (The rest of the arrays were producing around 2.8kW at the time).
Since then, I added (as I should have in the first place) - series schottky diodes for each array.
I bolted each of them to a copper busbar for both electrical interconnect, and thermal dissipation.
The thin wire on the leads is very short. It is thin so it doesn't have a tendancy to break the leads off the diodes, and it will make a fusible link in the event of some catastrophic failure.
I now get zero current flowing back into any array, regardless of which ones are in shaddow (clouds moving past, arrays shading each other, etc), and I will never get another incident like this and blow up a panel.
--- End quote ---
Wow, thanks for sharing this post with us Ross. That's an eye opener for sure. So, did you get to keep the replaced set on top of the replacements?
rossw:
--- Quote from: Watt on February 01, 2012, 07:02:45 pm ---Wow, thanks for sharing this post with us Ross. That's an eye opener for sure. So, did you get to keep the replaced set on top of the replacements?
--- End quote ---
No, I only asked for (and received) the diodes.
There was no damage (that I could ascertain) to the rest of the modules.
Since the man in China was so eager to help out, and sent me 15 diodes (I needed 12), without any hassles, and wouldn't take any money, AND sent them international overnight courier, there's no way I was going to try to screw him for more panels.
I fixed them with his parts and my labour, and was happy to come to such a resolution so quickly and with so little out-of-pocket expense.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version