Schneider 60/150 Blacked Out?

[Nobadays]

I thought I was beyond beginner level.... but apparently not!

Background, I had 1500 watts rooftop solar, feeding my Schneider 60/150 SCC, 48 volt lithium system. This is 250 watt panels, 3sx3p. Being on the roof we found that during the winter problems arose when 2"-3" of snow covered them and even if the sun came out later/next day it could take a full day or more if it was really cold for the snow to slide (45* angle) and we commence charging again. The solution was to add a 3rd string attached to the front deck that could be reached to sweep the snow off. Identical panels.

Here are the stats:
Power250Watts
VOC37.6Volts
VMP30.3Volts
ISC8.85Amps
IMP8.27Amps

Nominal Battery Volts 48Volt lithium (Chevy Volt Modules, 12kw)
PV Array Number Of Modules In Series3 Number Of Parallel Strings3 Total Modules9
Rated PV Array Power2250Watts
Anticipated Array Power @ 93 F2160Watts
Rated PV Array Current24.81Amps
Battery Charging Current @ 57.6V39.1Amps
VMP (Maximum Power Point Voltage90.9Volts
VOC (Open Circuit Voltage)112.8Volts
VMP @ -33 F115.8Volts
VOC @ -33 F135.6Volts

The two strings on the roof are combined up there using MC4 wyee connectors and the +/- then come off the roof to a Midnite Solar combiner box with one 30a breaker. Got the new string wired yesterday and brought their +/- leads to the combiner box where I added another 30a breaker on the bus bar. From there the lines run to another breaker box and through another 30a breaker before feeding the SCC. Output from the SCC returns to that breaker box then through a 60a breaker before the connection to the battery bank positive bus bar.

When I kicked on the breaker yesterday to bring the new string online all seemed to go well. It was about mid day and the panels were supplying about 1600 -1700 watts... maybe a bit of shade, plus panel tilt is fixed and not optimum for this time of year. That said, 1600 - 1700 watts is pretty good. The battery level was at 46.5v when I energised the system and within an hour we floated. I noticed then that the input wattage had dropped to 2 watts... should have alerted me as all loads will run off the panels first when Floating, then pull from the battery bank. Our average load during the day is around 40 watts... fridge, freezer, satellite internet.

That evening I looked at the SCP and noticed the SCC was not showing connected! I went into the room the equipment is in and sure enough the SCC screen was blacked out... nothing, no power. I quickly scanned the manual and saw a note about the ground fault fuse. So I kicked both breakers - 60a/30a, to and from the SCC - and popped out the fuse to check continuity... it was good. I put it back in then flipped the breakers back on and the SCC came back to life. It was dark so no input from the panels, just from the battery bank. Unfortunately I had not noticed when I shut the breakers off if one or both were tripped.... I assume both though as the SCC was off...?

Is my math bad? What am I missing.... I don't see how I overload the SCC, breakers? When I wired the new string into the combiner box... did this increase the voltage... I thought it would increase the amps, but not enough to matter? Any thoughts?

Thanks for the help!

 

[Nobadays]

Anybody? Did adding these new panels at the combiner box increase my voltage beyond the 150vdc the SCC is rated for? Or , as I believed, just opted the amperage?

Thanks!

 

[Hedges]

You added a new string - same panel as the rest? How many in series?
What is the temperature coefficient of Voc for the panel? Could be something like -0.3% per degree C, or volts per degree C.

Your charge controller is OK up to 150V max, so with 3 panels in series you're fine but 4 panels would go way over.

Conext MPPT 60 150 Charge Controller

The Conext MPPT 60 150 is a PV charge controller that delivers the max available current to charge batteries while improving battery life.

solar.schneider-electric.com

https://41j5tc3akbrn3uezx5av0jj1bgm-wpengine.netdna-ssl.com/wp-content/uploads/2020/01/DS20200117-Conext-MPPT-60-150.pdf

It is also rated for max 60A PV current, so four strings about 32Asc is fine.

If you have 12 panels now, 3s4p, that's 3000W STC. Could go higher on a cold day.
Into 48V, about 62.5A (less inefficiency). While battery voltage lower, could be higher.

Your charge controller spec is 60A max, might barely exceed it depending on conditions.

"ground fault fuse" - not sure how yours work, but my SMA grid-tie inverters are first commissioned with the fuse out to measure no shorts from PV string to ground. Then with a 1A fuse in the circuit, if a ground-fault occurs the fuse blows and is detected.

Check voltage coming in from PV to the charge controller between + and -, also from + to ground and - to ground.
With PV array isolated from charge controller, check again. Each + and - lead should briefly register a voltage relative to ground, then decay toward zero as parasitic capacitance drains into the meter. If one holds a steady voltage that would indicate a short to ground.

Interesting that it woke up after cycling breakers. Some charge controllers tolerate over-voltage by opening a relay to the PV panels, while others just blow up.

What happened after the sun came back up?

 

[Nobadays]

Hedges... thanks for the reply. Yes the new string are the same panels. The calculations I show is for all the panels including the three new ones. Originally 3s2p, now 3s3p...9 panels total, only 3 in series. Yes.... -0.3% per degree C, I believe. I have the new string turned off so don't know what could have happened when the sun came back up. Definitely afraid of cooking the SCC!

As stated the originals were 3s2p combined on the roof the brought down to a combiner box and breaker. I added another breaker in the combiner box then brought my new string of 3s into the combiner box to add to the other 2 strings.

I guess my confusion... other than why the breakers tripped... is what happens to amps and voltage when I added the 3rd string in to the two that were already combined. In my understanding, I would not have raised the voltage but just the amps. But I don't know if that is correct.

My calculations give me for 3s3p.... 2250 watts, VMP ~91vdc, ~25amps well within the limits of the SCC. I guess I could kick those new panels back in and put my meter on the combiner box output, at least I would know the voltage. If the breakers trip again I can ascertain which one... or both.

Thanks!

 

[Nobadays]

So talked to my solar engineer this is what he wrote back.

With three strings of panels now in parallel, the controller’s input breaker sizing would look like this (8.68A + 8.68A + 8.85A)*1.25 = 32.76A

The 25% is a safety which is used to actually size the breaker. So this is technically over the 30A, and if this were a new system I would be using a minimum 40A breaker for the input. This would explain why the 30A is nuisance tripping.

Now I understand! Ordered 40amp breaker but with 40amp I really should pull 8AWG to replace the 10AWG. That said..... I’m just barely over the 30amps and only with really good Simon the panels.

thoughts?

 

[Hedges]

Three strings - maybe tripping the breaker, although nominally about 27A.

Try orienting in several directions. If one aimed at 10:00 AM sun and another at 4:00 PM sun, instead of 9 + 9 = 18A, it'll be about 0.7 x that or 12.5A
That flattens the peak and spreads it out over the day.
Consider one Noon winter sun, and the other two with a summer tilt (emphasizing more power for refrigeration in the summer)
Or, two with winter tilt and one noontime summer sun (emphasizing more lighting loads in the winter.)

It's not a big deal to have strings of different orientations connected in parallel to one MPPT. Just don't want one string partially shaded.

Of course, on the roof makes arbitrary angles more difficult. Tilted panels on a flat surface need to be separated further to avoid shadowing each other.

 

[Nobadays]

Hedges... thanks! Definitely a possibility except the two strings on the roof are primary and since the roof is at a 12/12 pitch - 45* not going to go up there and make changes. At this time of the year I have a lot if shading on the third string, it only has good direct sun from about 12 noon to 2:30/3:00 PM. They are positioned for winter, angle and location.

I'm going to drop the 40 amp breaker in and give it a shot. I have serious doubts I will cook the 10AWG wire. As you point out, I must be under 30amps most of the time and only spiking occasionally. My solar engineer recommended I put in 50amp breaker and said with the 8AWG and 50amp breaker I would still be protected. This kind of tells me there is a pretty good safety factor built into sizing breakers to wire guage. If 8AWG will be protected by a 50amp breaker, surely 10AWG will be protected with a 40amp breaker.... though I know not code.

Thoughts on wire size/ breakers where I am only getting the occasional spike?

 

[Hedges]

A spike to < 2x rating won't trip a thermal/magnetic breaker, but fast-trip breaker might.
Two hours of noontime sun isn't a spike.
On a cold winters day with direct sun plus extra illumination coming off clouds, PV panels can exceed their label specs.
So, breakers and wire should be significantly oversized compared to Isc label ratings.

I've tested a 20A breaker at room temperature and 30A load. It tripped after 10 or 15 minutes.
The wire is dissipating 1.5^2 or 2.25 times as much power, 2.25 times the temperature rise. That would bake out the insulation eventually, maybe after 100 trips of the breaker, if it exceeded ampacity of the wire. Since 12 AWG actually has ampacity of 30A but NEC calls for 20A breaker, household wiring that consumers can easily overload is protected. Larger gauges like for range or furnace don't have that extra safety margin.

Breakers work by heat, so ambient temperature (and how well the breaker is cooled by air circulation), self-heating from the current it measures, and also added heat from a poor connection can trip it. I think I had a 100A breaker tripping under 70A due to bad screw connection.

Since your third array is shaded much of the day, how about tilting it back for summer sun? That would reduce this winter peak and improve summer production if it gets sun more hours. Or orient it for 2:00 PM rather than noon, shift some of your production later in the day when the other two aren't peak.

The third string doesn't have its own MPPT, is wired in parallel with the other two. When partially shaded, depending on how your MPPT works it might do little for you or might actually reduce power production by tricking the MPPT to find a local-maxima peak.

Is the 10 AWG after strings combine and before charge controller? Maybe run one more 12 AWG just for the third string, to its own fuse/breaker.

 

[Nobadays]

Good information! My breakers are all Midnite Solar, I assume they are thermal/magnetic. They are rated to carry 100% of the rated capacity. This time of year the thunderstorms roll through pretty much every day so we have "full sun breaks" and I think it is in one of these that I have tripped the breaker. I would guess with clear blue sky days the amperage is going to hover right at 30 when the sun hits all three strings at a direct angle... no shade. I really have no where to go with the third string and tilting it is not an option I want to pursue, it is on a wooden mount so the only option would be to rebuild it.

All of my wire runs in conduit, #10 coming off the roof carrying both those strings parallel/combined. Then from the third string also #10 to a combiner box with two breakers.... one 30amp for strings 1&2 and one for string 3. It is after this, now that all three are combined the issue of wire size comes into play. This combined feed then goes inside - a distance of about 25' in 3/4" conduit - to another DC breaker box with the 30amp breaker that is nuisance tripping, this line then goes to the SCC, (also a 60amp breaker in there between the SCC and the busbar for the lithium modules. ) Here is a post with pictures of my setup: https://diysolarforum.com/threads/colorado-off-grid-cabin-system.2293/#post-21480

Seems I really only have 3 options,
Put the 40amp breaker in, turn it on and take a chance on cooking the #10 wire between the combiner and SCC.
Pull #8 wire in to replace the #10, or
Leave the breaker off on the third string except in the winter when the first 2 strings are snow covered and/or bright cloudy days when a little extra production would help but leave little chance of exceeding the 30amp capacity of the #10 wire. This is what I am doing now, I kick it on when it is bright but cloudy and do see a nice rise in production.

One other question... I downloaded a section of the NEC on amperage capacity of wire.... I admit it was over my head, but it seemed that #10 has a capacity of 40amps.... is that correct?

Thank you!

 

[Hedges]

Doesn't meet code, but I don't think your PV can actually put enough current in any of the wires to overheat it.

Ampacity ratings are really about the temperature insulation can withstand; the copper can carry far more, just runs hotter.
NEC tables show three insulation ratings, 90 degree C insulation is good for typical THHN & THWN house wiring, also MC cable used for PV panels.

The usual table is for up to 3 current carrying wires in a cable or conduit.

For 120/240VAC split phase and for 120/208Y three phase, the two or three hot wires count. Ground doesn't count, and neutral doesn't count because any current it carries is offset by reduction in a hot. The red and black for split phase, or three colors for 3-phase, have to be of different polarity (180 degrees or 120 degrees apart) so their current subtracts rather than adding in the neutral.

If you put two circuits, e.g. a red and a white plus a black and white in one conduit, either two 120V outlets or two PV circuits, that is four current carrying conductors, and ampacity has to be derated. If hotter than normal environment, ampacity has to be derated.

Yes, 10 AWG ampacity is 40A (up to 3 current-carrying wires in conduit, normal temperatures.) Code calls for 30A max breaker, however. Due to the higher wire ampacity, derating doesn't kick in until you have a larger number of wires or higher temperatures. That lets you get away with two strings wired in one conduit and MC wires out in the sun, still using 30A breaker. The MC wires in free air not conduit also gives them a higher ampacity.

The 30A breaker for 2 strings and 30A breaker for 1 string sound fine for those 10 AWG wires. After they combine, 8 AWG and 40A or 50A ought to be used. They are not technically correct for protecting strings, however.

Because you have 2 not 3 current carrying conductors, heating is 2/3 as much. The wires would have to carry more, square root of (3/2) to cause the same heating, but NEC doesn't allow higher breaker ratings for those of us using just 2 not 3 current carrying conductors.

NEC PV rules let you have wires without fusing, but require that wire ampacity have more headroom above Isc than if fused.

The PV panels will have a "maximum fuse" rating, and it is less than the 30A you are using. If the charge controller is capable of backfeeding from battery into the panels due to a fault, fuses would be needed to protect the PV wiring. When you had just two strings 3s2p, fusing of the panels was not required to keep the second string from putting excess current into the first string if it shorted. Now that you have three strings 3s3p, each string is supposed to have its own fuse. (panel rating is probably 20A fuse, possibly 15A fuse.) If an MC wire to one string gets shorted, e.g. pinched in mounting hardware, the other two strings will each dump 8.85A, total 17.9A, into the first. Cold sunny day and that may exceed 20A. As you can see, with 3 strings, such a fault won't actually exceed ampacity of the MC cable enough to cause a fire. four or more strings could. Code call for fused strings when you have 3 or more in parallel.

Over current in the wires or PV panels isn't the only way to start a fire. Poor connection (like a plug in the wall that gets hot) can do it too. Arc-fault it what's necessary to stop it. You have 2 kW of panels, which would dump up to 2 kW of power into a single point. That's a small arc welder, can melt steel. Our normal house wiring we don't worry about that, although latest codes does and requires AFCI in 120V convenience outlet circuits. Also PV panels on the roof. Maybe your charge controller already has it.

I don't think the sun can produce enough current from PV panels to overheat any of the wires. NEC, however, requires fusing at current limits below what the sun can produce so it isn't possible for you to comply with NEC and avoid tripping so long as all panels are oriented the same.

Your drawing shows MNPV3 combiner. You ought to put three, 20A (or 15A, whatever PV label says) in it and bring each 3s1p string to it. Then run 8 AWG back from combiner to charge controller and protect that with 40A or 50A. To meet code I think you need to double the size of your combiner box or add a second one, fusing both ends of each string. (or get MC4 fuse holders to do the same thing.)

These days, I think code requires separate fusing of + and - PV wires (for more than 2 strings). (Back when I started mine in 2003, code misinterpreted itself and required - terminal to not be switched or fused.) There are places where a short from one string's - wire could cause an over-current condition, dumping all current from other parallel strings into the short, so the rule makes good sense. My original system had each inverter wired to PV either 24s1p or 9s2p, no fuses required. I now have larger inverters with multiple strings, all fused at both ends in a combiner box. The 9s2p array has grown to 12s2p, will be expanded to 12s3p and will have 6 fuses at the array when it gets merged into the other large inverters with additional strings.

 

[Hedges]

Good information! My breakers are all Midnite Solar, I assume they are thermal/magnetic. They are rated to carry 100% of the rated capacity.

What exact breaker model?

http://www.midnitesolar.com/pdfs/Q-Series(1).pdf

http://www.midnitesolar.com/product...e=Breakers&productCat_ID=16&sortOrder=7&act=p ?

Ratings of 80V, 125V, 250V. Some are polarity sensitive (maybe due to how they did magnetic, or how it is designed to pull an arc with magnetism?)
Some require two poles in series to double the breaking ability, in which case my suggestion of two separate combiners wouldn't be good, each string should have its two ends go to a 2-pole breaker. But, if original system was designed with single breaker it should be good for your approximately 125VDC Voc

"Hydraulic-magnetic"

I'm not sure what is hydraulic about it. Eaton website:

Hydraulic-magnetic circuit breakers

Eaton breaker products meet American, European and Asian agency circuit breaker standards, providing equipment, branch and main circuit protection worldwide.

www.eaton.com

• Instantaneous, short, medium or long tripping curve
• Standard InRush level of 8x, 15x, 22x nominal current at 50Hz and 8x, 18x, 25x at 60Hz

I would think the Midnight Solar are fast trip. Their "100% rated" means OK to repeatedly open them at full current, not that it is OK to use them at their 30A rating and expect them to not trip.

"tripping curves OP, U2, 1, 9", so several to choose from, not sure which the breaker I linked above is.
Tripping Curves refer to how much time at how much overcurrent.

Your breakers ought to be 25% greater than the maximum Isc you can get from PV on a cold day, with extra sunlight reflecting off clouds. (and, not more than NEC code.)

 

[Nobadays]

This is the 40amp breaker I just ordered... the 30 is the same: https://www.amazon.com/gp/product/B00BSYST96/ref=ppx_yo_dt_b_asin_title_o01_s01?ie=UTF8&psc=1

The system was designed by a solar engineer at NAWS, he basically told me the same thing.... probably would be ok just changing the breaker but would not be to code so he could only recommend using all three strings if I change to #8 wire from the combiner in.

The only thing in the conduit from the combiner to the breaker box inside are the +/- PV wires and a #10 ground wire. We are at 9500' in the Colorado Rockies, the highest temp we have seen in two years was 83* F, usually hovers around 70-75 all summer.... which is short! Winters can hit -30 but rarely, usually +10 to -10/-15 is our lows.... mid day usually always in the + range. But we did try to design for the record of -33F to account for over production from the panels.

I will change the breaker to the 40amp and will continue to manually turn the third string on and off as needed.... bright cloudy days like today, and will run the third string when the first two are snow covered. On bright sunny days I will turn the third string off, I don't need it anyway. Next year when our other projects are completed I will replace the #10 with #8 so we will be back to code.

It is good to know that the likelihood of overheating the wires is negligible.

Thank you for all the information!!

 

[Hedges]

Look into a combiner box that takes 6 breakers for both ends of 3 strings, or larger if you might add more strings later. That appears to be the missing part to be completely code compliant.

Looks like NEC ampacity of 8 AWG is 55A, and I think you're allowed to round up to next standard breaker, 60A. 50A or 60A breaker should allow 4 or 5 parallel PV strings without nuisance tripping (more if not oriented the same), so whatever changes you make might as well be good enough for the next expansion. PV panels are so inexpensive these days.

Your charger is rated for 60A output and 48A (STC) of PV, 5 strings in your case.

However, I do NOT see anything about AFCI or arc-fault protection. That may only be available as a separate box. Code requirements have evolved over the years in different locations.

 

[Nobadays]

Thank you!

 

[Nobadays]

Well.... I found the problem... I did go ahead and install the 40amp breaker just before the charger controller, but the combined input from my 3 strings was not what was making the previous 30amp breaker trip. It was a rookie mistake.

When I popped the cover off the breaker box for the MPPT SCC, holding the 30amp input breaker and the 60amp output breaker, I realized I had switched the breakers during the installation. The 60amp was taking the input from the PV strings and the 30amp was handling the output to the batteries from the SCC. Apparently.... without doing the math... before I added the third string, my original 2 strings did not put out enough voltage that even with the MPPT SCC increasing the output current, it was still under that 30amp threshold. Once I added the third string my output increased substantially thus tripping the 30amp.... that should have been a 60amp breaker. They are now reinstalled correctly!

I did not, yet, change the #10 inpuwire to #8, that may get done next spring. Theoretically my three strings shouldn't exceed 27amps.... maybe on a very cold day if.... a big IF... they were oriented and tilted to collect the optimum sunlight on that day. My panels are all oriented directly south ,but are tilted for fixed placement. Meaning, their tilt would only be optimum mid spring and mid fall, not during the winter when the perfect storm could happen. I have been watching the input pretty closely, early afternoon when there is full sun on all three strings, no shading and have yet to see an input current above 21amps. Mostly it fluctuates averaging about 18 - 20 amps on clear bright days, or during a full sun break in the clouds.

Thank you again for your help! I will continue to monitor my input amperage but I am pretty sure I can safely say I will not likely exceed the 30amps and if I do that is what the breaker is for.