diy solar

diy solar

Help needed choosing MPPT charge controller

Sorry, missed your reply earlier when I replied to others who’d posted after you.

Thanks for your tips. I’ve already experienced a run of heavily overcast days and was dismayed by how little the panel produced - its what lead me to order a second identical 160W panel to run in parallel, but even so I’m quite sure there will be plenty of days thru the winter when I run out of power and need to switch on a mains powered battery charger.

The purpose of my solar exploits has been something of a moving target. I suppose where I’m at right now is that it will reduce our mains power bills as I transition as many household devices from inefficient “wall wart” PSUs to running direct from solar DC.

The initial aspiration when I first began looking into solar was “how do I run my house ventilation fan on solar power, can I just stick a 20W panel straight on the top of it and have a battery local to the fan”. I decided that wasn’t viable and it would be better to get a larger 160W panel to charge a car battery to run the fan plus whatever other devices the available power would stretch to.

So what started as “small solar panel connected direct to one device” turned into ”bigger panel charging a car battery” and that has now evolved into two solar panels, an MPPT controller and a 100A LiFePO4 battery. The costs seem to go ever upwards once you get sucked into this solar panel lark!
Welcome to the reality of >50° latitude's solar, brother!

You will quickly discover that you will permanently get either too much, or too less energy.
Your 300W panels will just produce 15W for a couple of hours during the winter, not even enough to run a decent internet router.

In the summertime, you will get so much energy that your 100Ah battery will be filled at 11'clock and your SCC will throttle down for the rest of the day...
So optimizing busbars is pretty vain:
-during the wintertime, the panels will have a so much high internal resistance and feed so few amps, that the leads do no really matter,
-during the summertime you will have too much energy anyway, so losses don't matter again.

Of course, you have sunny days in winter as well, and they are a game changer.
Frequently only an hour or two in the morning or in the evening, so it is important to have at least one panel oriented correctly to harvest that opportunity.

I would place the two panels at 60° oriented SSE/SSW wired in parallel and be sure that they have a Schottky diode in SERIES. (most panels are sold with a reverse Schottky diode in parallel, intended to wiring panels in series.
 
Welcome to the reality of >50° latitude's solar, brother!

You will quickly discover that you will permanently get either too much, or too less energy.
Your 300W panels will just produce 15W for a couple of hours during the winter, not even enough to run a decent internet router.

In the summertime, you will get so much energy that your 100Ah battery will be filled at 11'clock and your SCC will throttle down for the rest of the day...
So optimizing busbars is pretty vain:
-during the wintertime, the panels will have a so much high internal resistance and feed so few amps, that the leads do no really matter,
-during the summertime you will have too much energy anyway, so losses don't matter again.

Of course, you have sunny days in winter as well, and they are a game changer.
Frequently only an hour or two in the morning or in the evening, so it is important to have at least one panel oriented correctly to harvest that opportunity.

I would place the two panels at 60° oriented SSE/SSW wired in parallel and be sure that they have a Schottky diode in SERIES. (most panels are sold with a reverse Schottky diode in parallel, intended to wiring panels in series.
Thanks! What’s the purpose of the schottky diode in series with my parallel wired panels?
 
Yes, I think I answered these in my other comment from a short time ago, except the last point. In summer I would expect to have an excess of power to use and will likely have an inverter by then to run some mains devices such as fridge and freezer, outdoor lighting overnight, maybe TV if there’s really plenty of excess. Basically my strategy will be to scale demand to match the solar energy supply, as I can switch plenty of devices in the home between mains or solar.

I’ll only be using an inverter for devices which won’t run DC and will use low power DC/DC buck/boost voltage converters to drop down to 9V, 5V etc as needed.
So the next device to budget is a grid-tie inverter...
Not really cheap, and the ROI will probably be 20 years or more.
Here in Germany where grid power is horribly expensive, i gave up. It was not economical.
 
Ups, I made a mistake: you ought to place your two panels at 90° (maybe even more) because the winter sun is turning quite a lot.

You could also just forget solar during the winter time, it's probably not worth the effort.

I am not that lucky: I am conceiving an complete off grid system, so I MUST find a way to pass a tough winter...
 
Avoiding to discharge the insolated panel into the other one that is less well oriented. That makes only sense if your panels are not side by side.

ok, I need to look into that then, as I thought I could just parallel connect my panels without extra diodes. The panels are these https://www.amazon.de/dp/B07C3ZXGYY

So the next device to budget is a grid-tie inverter...
Not really cheap, and the ROI will probably be 20 years or more.
Here in Germany where grid power is horribly expensive, i gave up. It was not economical.

Why grid tie inverter instead of a standalone unit?
 
With the grid-tie inverter, you just feed power back into your standard installation.
No hassle, plug and forget...
Depending on your meter, you may even feed back into the grid and decrease your KWhs.

A stand-alone inverter will require a separate power installation and you must balance your consumption with the available harvesting.
That can be a nice hobby, or hit your nerves...
The stand-alone inverter has however the advantage of providing a limited back-up in case of outages, which a grid-tie cannot provide, excepted some even more expensive models designed to be both-in one.
 
The initial aspiration when I first began looking into solar was “how do I run my house ventilation fan on solar power, can I just stick a 20W panel straight on the top of it and have a battery local to the fan”. I decided that wasn’t viable and it would be better to get a larger 160W panel to charge a car battery to run the fan plus whatever other devices the available power would stretch to.
I found a 48V fan (from electronics rack) which would start and run as voltage slowly ramped up. Also some 5" box fans.
I put the large fan in a gable vent and the small ones on heatsinks of my inverters. With a couple panels, as the sun shines the fans run, faster on sunnier days. No battery.
 
I found a 48V fan (from electronics rack) which would start and run as voltage slowly ramped up. Also some 5" box fans.
I put the large fan in a gable vent and the small ones on heatsinks of my inverters. With a couple panels, as the sun shines the fans run, faster on sunnier days. No battery.
Thanks. I have a suitable Ebm papst DC brushless fan already installed, which has run from mains using a laptop power supply. It’s been running on solar for over a month now - works well.
 
Not sure if you got your answer to how much does the voltage go up in the cold.
Thanks @SolarQueen, that’s ace info. Does this rise in voltage come with a commensurate reduction in current, so the power supplied from the panel stays the same, or is the panel making more power in colder conditions by having a higher Vmp and same current as at 25C?

I suppose even on a day when the ambient is -10C, the panel temp will be higher due to the solar heating.
 
That's the
Thanks @SolarQueen, that’s ace info. Does this rise in voltage come with a commensurate reduction in current, so the power supplied from the panel stays the same, or is the panel making more power in colder conditions by having a higher Vmp and same current as at 25C?

I suppose even on a day when the ambient is -10C, the panel temp will be higher due to the solar heating.
That's the cool thing about solar. When it is cold, the current pretty much stays the same, but the voltage goes up. The standard test condition is a 25C, so anything colder than that can get you higher voltage x same amps = more watts!
 
That's the

That's the cool thing about solar. When it is cold, the current pretty much stays the same, but the voltage goes up. The standard test condition is a 25C, so anything colder than that can get you higher voltage x same amps = more watts!
That’s something to feel positive about when I’m shivering my socks off this winter! ?

Thanks for the video on blocking diodes also. Most interesting. So is it the case that a blocking diode is only needed in parallel connected panel arrays when one panel is fully shaded when the other takes full sun? Or are they also justified when the shading is partial, like a leafy tree branch casting a shadow which moves across first one panel, then the other?
 
That's the cool thing about solar. When it is cold, the current pretty much stays the same, but the voltage goes up. The standard test condition is a 25C, so anything colder than that can get you higher voltage x same amps = more watts!
Yes, but unfortunately the cold weather comes together with a LOT less lux than the reference, so you may get +12% from -90%... :(
 
Yes, but unfortunately the cold weather comes together with a LOT less lux than the reference, so you may get +12% from -90%... :(
But on sunny days, add in a little reflection from the snow, ooh la la! Albeit for a lot fewer hours than summer. But the few hours it's going, yah baby!
 
That’s something to feel positive about when I’m shivering my socks off this winter! ?

Thanks for the video on blocking diodes also. Most interesting. So is it the case that a blocking diode is only needed in parallel connected panel arrays when one panel is fully shaded when the other takes full sun? Or are they also justified when the shading is partial, like a leafy tree branch casting a shadow which moves across first one panel, then the other?
Not really worth it for tiny shading, but when you know they are going to be majorly different from each other most of the day, it can certainly help.
 
Thanks for the video on blocking diodes also. Most interesting. So is it the case that a blocking diode is only needed in parallel connected panel arrays when one panel is fully shaded when the other takes full sun? Or are they also justified when the shading is partial, like a leafy tree branch casting a shadow which moves across first one panel, then the other?

I don't think blocking diodes are necessary or beneficial so long as nothing malfunctions.

Multiple PV strings of same/similar Voc & Vmp work well in parallel, even if some are partially shaded or some have different orientations.

Different orientations and current production from the one with less light is reduced, but it's Vmp doesn't drop much. It consumes a small amount of current from the other string and Vmp of the two combined ends up between what either would have been individually. Power production is reduced a couple percent vs. having each on its own MPPT. SMA performed a test and reported those results (users had already figured that out, knew they could take the former advice to orient all strings the same with a grain of salt.)

Two identical long strings oriented the same and one slightly shaded, Vmp reduced to about Vmp of the string with fewer fully illuminated panels. Again impact on power is slight. I tried shading one panel where I had 9S2P, and power was only reduced about 1/18th as reported on the inverter. This is not the same has having a short string paralleled, because the non-illuminated cells in the shaded panel still over their voltage drop to avoid being backfed by the other panel. Performance issue comes if so many cells are shaded that the combined IV curve causes the power/voltage curve to have two bumps, a local maxima which isn't the highest power point. A simple MPPT algorithm starts at open circuit and increases current draw so long as power increases, stops at the point where power decreases, and tracks that point. Only some inverters or charge controllers explore a wide current/voltage range to find the highest peak.


As for blocking diodes - they just prevent backfeed into a string of lower voltage. That won't ever improve power production for similar voltage strings in parallel, except possibly for the case of one heavily shaded string and an MPPT that can't find the lower voltage higher peak. But voltage drop through a diode (a few 1/10ths of a volt, up to a couple volts) consumes power all the time current is flowing, probably exceeding losses even during that shaded condition.

Where blocking diodes will help is if one string is lower Voc, a configuration you shouldn't assemble. If a bypass diode fails (which is reportedly fairly common), the diode will prevent backfeed. If one diode in a string fails (due to shading on that panel while others have full sun), the same will likely happen to other panels as well unless the shade is localized. Having a blocking diode in this case is like having them for redundant power supplies in a computer system, or for redundant batteries. It prevents a failed component from pulling the system down.

Fuses, recommended when more than two strings are paralleled, prevent catastrophic problems in the event of such a failed string. But the fuse won't blow and clear the fault unless several strings are paralleled, dumping enough current to blow the fuse (typically above fuse rating.) Instead system will operate at reduced power so long as it still has enough voltage headroom.

What I don't know is which panels have a problem with diodes failing. The diodes likely are selected for data sheet rating to carry full current, but the panel designer failed to engineer thermal dissipation to get rid of the power produced and the diode overheats & fails if there is shading during full sun. It only serves its purpose under partial sun, letting a shaded panel be bypassed with current to avoid over voltage of reverse-biased cells.
 
I don't think blocking diodes are necessary or beneficial so long as nothing malfunctions.

Multiple PV strings of same/similar Voc & Vmp work well in parallel, even if some are partially shaded or some have different orientations.

Different orientations and current production from the one with less light is reduced, but it's Vmp doesn't drop much. It consumes a small amount of current from the other string and Vmp of the two combined ends up between what either would have been individually.
You mileage may vary.
On higher latitudes, during the winter, when sun is scarce and you have pain to harvest a few Watt, you can't afford to lose mA into a shaded panel.
The sun is also heavily turning so you really need the SE and SW orientations to catch every short sunshine that could pierce the clouds.
 
Back
Top