clouds parting overvoltage transients from MPPT controller algorithm--how to fix?

[Iuval]

Janis

Regarding panels: is the 200 W You mentioned total power of 3 panels You said You have? Be aware, that 75/15 SCC at 12 V system is designed for 220 W nominal PV power.
And what are the open circuit voltage of the panels? 75 V are the hard limit to respect taking the lowest possible temperature into account.
It would be great to look at the datasheet.

No, each panel is 200W, total 600W. But why won't the 75/15SCC work then? I rarely have more than 7A total, 63V (that's about 21V open circuit per panel) from those panels, unless the spiking gets it over, but hopefully, if it's well designed, it should be able to handle it

 

[sunsurfer]

Janis

No, each panel is 200W, total 600W. But why won't the 75/15SCC work then? I rarely have more than 7A total, 63V (that's about 21V open circuit per panel) from those panels, unless the spiking gets it over, but hopefully, if it's well designed, it should be able to handle it

First, look at the numbers on the Victron 75/15. Thats a max input voltage of 75volts from the panels and a max charging amps to the batteries of 15amps. The 15 is battery charging amps, NOT amps from the panels.

Take your charging voltage of say 14 volts and multiply it by the max charging amps 15 and you get 14x15= 210w

You could under panel and use one panel per 75/15 or you could over panel and go with a 100/30 and would be able to run all 3 in series and over panel. I think the victron could handle it but maybe someone else can confirm. You would only see max 420W max charge but with 600w panels, you likely will not see more then that anyway on hot days. Figure your panels at 65% when they are hot.

I would think the 100/30 is best option with your 3 panels in series.

 

[Bop]

So many people seem to think that the current limit is the input rating from the panels- NO- it is what the controllers MAX output is...

For example, my 60A 12/24/48v charge controllers have a 750w panel limit on 12v, 1.5kw on 24v and 3kw on 48v- they can accept 'overpanelling' (as long as you don't exceed the PVmax rating of 150v in my case) but in each case, they will still max out at that 60A OUT....

Your 75/15 on a 12v battery is 15A OUT, 12v x 15A = 180W maximum power into the battery- ever...

(you can hang the 600w off of it- and it will still only put that 180w into the battery- and you will be one of the many extremely disappointed in the performance...)

If that current rating was for the panel current limit in, at 150v max rating and 60A, that would make my controllers capable of handling a ridiculous 9kW!!!! (I wish lol)

Even the little 75/15 max wattage would be 75v x 15A = 1.125kW



(and the PVmax is the HARD limit- do not exceed- go directly to jail, do not pass go, do not collect $200...)

'Most' manufacturers will have a 'little' wriggle room, but not much- it is the breakdown voltage of the components inside that sets that PVmax voltage, exceeding it is pushing them past their maximum breakdown voltage....
Otherwise 'the smoke comes out'...

 

[Bop]

First, look at the numbers on the Victron 75/15. Thats a max input voltage of 75volts from the panels and a max charging amps to the batteries of 15amps. The 15 is battery charging amps, NOT amps from the panels.

Take your charging voltage of say 14 volts and multiply it by the max charging amps 15 and you get 14x15= 210w

You could under panel and use one panel per 75/15 or you could over panel and go with a 100/30 and would be able to run all 3 in series and over panel. I think the victron could handle it but maybe someone else can confirm. You would only see max 420W max charge but with 600w panels, you likely will not see more then that anyway on hot days. Figure your panels at 65% when they are hot.

I would think the 100/30 is best option with your 3 panels in series.

Even the 30A is going to limit the charge current in good conditions (30A x12vnominal = 360W usable)- with 600w nominal of panels available, a 50A or 60A would be a better choice

(you do get slightly higher 'actual' charge wattages, as the current limit remains the same, but the voltage varies- ie the 30A would likely be about 14v on charge, at 30A it would be putting 30A x 14v = 420w actual- assuming the panels can provide it)

 

[Iuval]

Thanks

Even the 30A is going to limit the charge current in good conditions (30A x12vnominal = 360W usable)- with 600w nominal of panels available, a 50A or 60A would be a better choice

(you do get slightly higher 'actual' charge wattages, as the current limit remains the same, but the voltage varies- ie the 30A would likely be about 14v on charge, at 30A it would be putting 30A x 14v = 420w actual- assuming the panels can provide it)

Thanks, I cancelled the order. Might be cheaper to buy 2 inductors and 2 capacitors to make a Butterworth filter. The 50A victron is $182.

 

[Iuval]

First, look at the numbers on the Victron 75/15. Thats a max input voltage of 75volts from the panels and a max charging amps to the batteries of 15amps. The 15 is battery charging amps, NOT amps from the panels.

Take your charging voltage of say 14 volts and multiply it by the max charging amps 15 and you get 14x15= 210w

You could under panel and use one panel per 75/15 or you could over panel and go with a 100/30 and would be able to run all 3 in series and over panel. I think the victron could handle it but maybe someone else can confirm. You would only see max 420W max charge but with 600w panels, you likely will not see more then that anyway on hot days. Figure your panels at 65% when they are hot.

I would think the 100/30 is best option with your 3 panels in series.

Thanks. I just saw their 30A version. Might get it. Or might buy 2 inductors and 2 capacitors to make a filter

 

[Bop]

Thanks. I just saw their 30A version. Might get it. Or might buy 2 inductors and 2 capacitors to make a filter

Victron is $$$- if you don't want to spend big $$$ then look at less pricey gear- or live with having panels doing nothing towards charging...

 

[Iuval]

Victron is $$$- if you don't want to spend big $$$ then look at less pricey gear- or live with having panels doing nothing towards charging...

They were definitely charging with the epever, just having the spike problem. So why not fix that with a Butterworth filter?

 

[Bop]

They were definitely charging with the epever, just having the spike problem. So why not fix that with a Butterworth filter?

It isn't a 'problem'- it is what is actually happening- 'ye canna break the laws of physics Jim' to quote Scotty from Star Trek lol

The best solution is simply stay away from the PVmax voltage, and buy gear rated appropriately rated for the job...

 

[v_green57]

They were definitely charging with the epever, just having the spike problem. So why not fix that with a Butterworth filter?

See post #4.

A MPPT charger does NOT expect to see a filtered array and may NOT be able to work with it.

 

[ConnerLabs]

Have you priced the inductors you’ll need for this filter? Don’t forget to take account of their DC resistance. Personally I’d try one of those huge capacitors used by car audio people, in parallel with your inverter’s DC input terminals.

In a previous life I tested solar panels in a solar simulator chamber that used a bank of mercury arc lamps running off 50Hz AC mains. The DC output from the panels had a huge 100Hz ripple that freaked out every MPPT we tried. We had to change to incandescent lamps for that work and cope with the excess of red and infrared compared to real sunlight.

I guess the moral of the story is that the MPPT algorithm programmers assumed direct connection to a solar panel exposed to real sunlight and anything else, be it mercury arc lamps, capacitors or Butterworth filters, could confuse it.

 

[Iuval]

It isn't a 'problem'- it is what is actually happening- 'ye canna break the laws of physics Jim' to quote Scotty from Star Trek lol

The best solution is simply stay away from the PVmax voltage, and buy gear rated appropriately rated for the job...

The problem is that it can't filter out that spike. The nominal open voltage of the 3 panels in series is 63, much less than the 138 spec on the epever, and also less than the 100V spec on the victrons. So the only question is does the victron filter out the spike or not?

 

[Iuval]

See post #4.

A MPPT charger does NOT expect to see a filtered array and may NOT be able to work with it.

Was thinking on the output, across batteries and inverter, not input.

 

[Iuval]

Have you priced the inductors you’ll need for this filter? Don’t forget to take account of their DC resistance. Personally I’d try one of those huge capacitors used by car audio people, in parallel with your inverter’s DC input terminals.

In a previous life I tested solar panels in a solar simulator chamber that used a bank of mercury arc lamps running off 50Hz AC mains. The DC output from the panels had a huge 100Hz ripple that freaked out every MPPT we tried. We had to change to incandescent lamps for that work and cope with the excess of red and infrared compared to real sunlight.

I guess the moral of the story is that the MPPT algorithm programmers assumed direct connection to a solar panel exposed to real sunlight and anything else, be it mercury arc lamps, capacitors or Butterworth filters, could confuse it.

Was thinking about filtering output, not input of controller. I have not priced out inductors, could they possibly be more than $182 for the 50A victron?

 

[sunsurfer]

They were definitely charging with the epever, just having the spike problem. So why not fix that with a Butterworth filter?

You know...you could still have a spike issue even with the Victron but I would say you have to try it. Likely, it will be gone if you ever switch over to lithium battery. They have such low resistance, they are good at absorbing spikes. I would NOT trust epever on a lithium battery.

I would still get Victron regardless. A friend of mines epever 40a COOKED his lead acid batts once when it got stuck on a high voltage. Melted them down and made a giant mess of acid everywhere. Luckily no explosion or fire.

All the cheap SCCs are crap. Total waste of money. Sorry but they just are. I wasted to much money on "good priced" SCCs. Victrons are good. They are fast and can capture the small bits of light passing through the clouds. They have a good inductor/cap setup inside and top notch software and will spike less.

 

[Bop]

You are abusing the charge controller, and wanting to bodge up it- rather than just simply fixing the initial problem....
'you do you'....

 

[Iuval]

You are abusing the charge controller, and wanting to bodge up it- rather than just simply fixing the initial problem....
'you do you'....

I don't see why. Epever says it should not hurt it, I asked. If I was guaranteed that the victron could fix the initial problem (lack of filtering, NOT spiking, which as you point out is what IS, not THE PROBLEM), I would get it.

 

[sunsurfer]

You are abusing the charge controller, and wanting to bodge up it- rather than just simply fixing the initial problem....
'you do you'....

Hey could pull some panels out and retest it. That would be a quick check but from my experience with epever, he could still see spikes. I had the same problem and I was NOT over paneling. I also had a lead acid bank at the time. Lead acid just has allot of resistance and those epever controllers are crap. Very slow response to change in conditions and not much output buffer to smooth things out.

 

[Iuval]

Have you priced the inductors you’ll need for this filter? Don’t forget to take account of their DC resistance. Personally I’d try one of those huge capacitors used by car audio people, in parallel with your inverter’s DC input terminals.

In a previous life I tested solar panels in a solar simulator chamber that used a bank of mercury arc lamps running off 50Hz AC mains. The DC output from the panels had a huge 100Hz ripple that freaked out every MPPT we tried. We had to change to incandescent lamps for that work and cope with the excess of red and infrared compared to real sunlight.

I guess the moral of the story is that the MPPT algorithm programmers assumed direct connection to a solar panel exposed to real sunlight and anything else, be it mercury arc lamps, capacitors or Butterworth filters, could confuse it.

Might try just the capacitor, but it probably won't filter this low frequency spike (1/(2 Pi RC) is probably too large a frequency since R is so small). I AM worried about the DC resistance of any inductor. Will have to check.

 

[sunsurfer]

Might try just the capacitor, but it probably won't filter this low frequency spike (1/(2 Pi RC) is probably too large a frequency since R is so small). I AM worried about the DC resistance of any inductor. Will have to check.

If you do, please report back on how it works out. I have one and have wondered if it would work for such a situation.

 

[Iuval]

If you do, please report back on how it works out. I have one and have wondered if it would work for such a situation.

OK, will do. Thanks for everyone's help. You guys rock