Which Controller

[GTom]

Originally Posted by Steve O

...he said a 100/30 might be a better choice. He also suggested a Smart Controller so i could monitor the output via Bluetooth.

Depends how you grow your array:
In general, using multiple controllers are better on a boat than a single large one. (My only reason for my own 100/30 is that I bought the controller first and few days later got an irresistible offer on a package of smaller panels instead of the a single large one I planned). If you buy more panels, you just add another controller - 2x100w flex for the bimini? = another 75/15. You decide for a gantry with a huge 360W panel? = a new 100/30. Smart controllers: make sense if you have only 1 controller. In case you have more, than just use the basic version and buy one bluetooth dongle to tune all controllers.

[CatNewBee]

makes somehow sense compared to the serial aproach, and is also profitable for victron. I still think a parallel setup will be even better. serial arrays are built to reduce current and save on cabling in large arrays. in a 200w setup not relevant.

[CatNewBee]

why add a separate solar system with pwm just for winterizing. why dont you use what is already installed?

[newhaul]

Originally Posted by SailsWithFists

A fuse is blown by over amperage, not voltage. If the controller fails closed, then a solar panel bank wired in series is sending damaging amounts of voltage into the system. A fuse helps not in the least bit.

obviously you have never seen what happens when someone misswires a circuit to the wrong voltage.

[DotDun]

Originally Posted by newhaul

obviously you have never seen what happens when someone misswires a circuit to the wrong voltage.

Do explain how the current limited solar cell(s) will put out enough current to blow the fuse between the failed controller and batteries.

[GTom]

Originally Posted by CatNewBee

makes somehow sense compared to the serial aproach, and is also profitable for victron. I still think a parallel setup will be even better. serial arrays are built to reduce current and save on cabling in large arrays. in a 200w setup not relevant.

I suspect a few percent increase in controller efficiency when going from 17V to 34V. Has to be confirmed though, the only relevant data we got from the forum so far is when a fellow CF-er tested his Voc(!)=16V panels in series and parallel. No efficiency vs Voltage curves from Victron.

Originally Posted by CatNewBee

why add a separate solar system with pwm just for winterizing. why dont you use what is already installed?

I might worry too much, but to my latest info MPPT is not exact science yet, controllers fail every now and then. A reliable source (victron?) PWM is brick and mortar, I don't mind leaving it behind for months. Also the 50W panel stands a bit out, I am not sure if I can fit efficiently to the 4x100W pack and some redundancy doesn't hurt anyway.

Will give it a second thought after the summer.

[evm1024]

These Victron controllers do not supply 12 volts so we should not be doing any calculations using 12 volts.

Depending on mode they will be either outputting 14.4 volts or 13.8 volts (if there is enough solar) in a 12 volt system. Double that for a 24 volt system.

In general solar is just like batteries - parallel the panels and the available current adds together while the voltage remains the same.

Series the panels and the voltage adds while the available current remains the same.

The spec 100/15 says that the maximum input voltage is 100 volts and the maximum output current is 15 amps.

In the case of a 12 volt battery (the controller auto selects the 12 v mode) the maximum power delivered to the output is 200 watts (ok 216W = 14.4V * 15A). For a 24 volt battery the maximum power is 400 watts.


Just added this in because of the error in using 12 volts in some prior calculations.

And just for the systems view:

E(solar) x I(solar) = E(battery) x I(battery) x K

Where K is < 1 and represents all of the losses in the system.

[noelex 77]

Originally Posted by GTom

I suspect a few percent increase in controller efficiency when going from 17V to 34V. Has to be confirmed though, the only relevant data we got from the forum so far is when a fellow CF-er tested his Voc(!)=16V panels in series and parallel. No efficiency vs Voltage curves from Victron.

There is a mistaken belief that charge controller efficiency is best with high input voltages. This is not the case. The highest charge controller efficiency is usually just above (2-3v) above the battery voltage.

For example, this graph from Outback shows how the efficiency decreases the voltage rises.

[GTom]

Originally Posted by noelex 77

For example, this graph from Outback shows how the efficiency decreases the voltage rises.

My guess would be that the optimal voltage depends on the model.

[evm1024]

Whoo boy.

Just to give a look at what goes into some of the engineering read this for your pleasure and edification. (of course it is buck-boost but who is counting?)

http://www.ti.com/lit/an/slyt584/slyt584.pdf


IN general the buck converter is switching at a pretty much constant duty cycle. The duty cycle is scales by the difference in the input and output voltages.

The inductor flux and ripple currents (and the losses that go along with) scale with voltage. And then throw in a whole slew of internal chip losses that scale with voltage and it is easy (hah!) to see why peak efficiency is just a few volts (input) above the output voltage.

(Sorry, I could not help myself - it has been a long time since I was doing any serious design)

[noelex 77]

Originally Posted by GTom

My guess would be that the optimal voltage depends on the model.

Not all companies provide this sort of detail, but when published it shows the same trend from all manufacturers.

It certainly makes sense that the more the input voltage needs converting, the greater the losses inherent in performing the conversion.

So high voltage panels, or series connection, will reduce the controller efficiency although the effect is not great.

[GTom]

Originally Posted by noelex 77

Not all companies provide this sort of detail, but when published it shows the same trend from all manufacturers.

It certainly makes sense that the more the input voltage needs converting, the greater the losses inherent in performing the conversion.

So high voltage panels, or series connection, will reduce the controller efficiency although the effect is not great.

Learned something new today, maybe that 2p/2s configuration won't do me any good and I should stick to all parallel. This case however I should get some extreme low-loss blocking diodes too, otherwise the panels sunny side of the boat will fry the ones in the shade.