Oversizing panels but not in sun.

[tricky74]

Have been looking at output data on pvoutput and have an idea for poor weather.
As an example lets say the scc can handle a maximum of 1kw of input on a sunny day and we get 4 kw/h per kw.
That is 4 kw/h produced on a sunny day
On a crap day roughly 1 kw/h per kw in my area.
Same system does 1 kw/h.
So why not put 4kw of panels that are facing away from the sun so on a sunny day the output is still low.
4kw of panels on bad day.

Anyone doing something like this?

 

[gnubie]

You can put more wattage on the controller than it is rated for, over-paneling, a decent controller will current limit once it hits full output. There is a catch or two. Depending on how well the controller can manage its output side it may not be able to back off enough, or outright turn off, to properly control the current into the battery with excessively large array power. You also need to think about what happens if there is a short in the controller, ie it may be built to contain the consequences of a short at 20 amps, but 50 ... ?

So over panel, but don't go nuts, and fit a fuse on the PV input to the controller that is suitable for the controllers ratings. If it says max 30 amps, a 40 amp fuse is probably the right size. The fuse needs to be rated to break the Voc of the solar array.

 

[Mart Hale]

I agree with the fuse idea. All it takes is one day you forget, the fuse would be good insurance.

I have over paneled because the length of wire to my charge controller looses a fair amount of energy, and I had extra panels that output still fit inside of what my charge controller could handle. Think about the amps the cable can take and the amps the connectors are rated for.

 

[tricky74]

I was thinking about having panels mounted north (southern hemisphere) etc just switched off until a crap day.
But if you forget you have issues.
This idea however I am thinking mounted say vertical so 90 deg away from optimum.
They should never ever give much output unless the sun moves in which case we are toast anyway.
We have some average weather coming so I might do some testing and compare to what the optimum production angles are.

 

[svetz]

low panels on a string will bring down high panels just like shade. If they're not on the same string then I doubt it makes a difference.

 

[seven7]

I've always wondered why SCC cannot protect against overvoltage. They seem perfectly positioned in the system to do just that.

And if a simple fuse is a suitable insurance policy against overvoltage due to over paneling, Why don't SCC include them to protect the SCC and the batteries? Is this true for all SCC or just the cheap ones.

As a new comer to the PV world I'm always baffled when I hear that SCC can't protect batteries from over voltage. It just seems odd from the perspective of an uneducated bystander. Thank You to everyone for helping me understand the function of all these components.

Mike

 

[svetz]

...I've always wondered why SCC cannot protect against overvoltage....

Not an EE, but I think they do to a certain extent...you only get into trouble if you go over their design rating.

 

[seven7]

Svetz thanks!
The degree of patience you guys have for us beginners is inspiring.

 

[gnubie]

I've always wondered why SCC cannot protect against overvoltage. They seem perfectly positioned in the system to do just that.

And if a simple fuse is a suitable insurance policy against overvoltage due to over paneling, Why don't SCC include them to protect the SCC and the batteries? Is this true for all SCC or just the cheap ones.

As a new comer to the PV world I'm always baffled when I hear that SCC can't protect batteries from over voltage. It just seems odd from the perspective of an uneducated bystander. Thank You to everyone for helping me understand the function of all these components.

Mike

The input side of the SCC can be protected, clamp diode + fuse or any number of other ways, since we know the maximum permissible voltage but what about the battery side? That's not so easy. Yes you can do the diode trick there, and many do, but where do you clamp? 12v, 24v, 36v, 48v, 96v? Many (if not most) SCCs these days will work with different voltage batteries so any dumb clamping must happen above the highest battery voltage. A good a quality controller can be trusted to keep voltages where you want them through its normal operations.

The question is can you rely on a cheapie to stick to the specified voltages during normal operations? My own testing has shown several simply can't and I'm not talking 0.4v like the MakeSkyBlue controllers either, more like 5V over. The $20 blue PWM controllers are well known for not doing such a good job with regulation, and this piece of garbage is essentially a death warrant for your battery.

What happens when things go wrong needs to be considered too. If the load goes away, say lithium BMS disconnects, and there is a reasonable amount of current flowing at the time the output voltage can spike up. Usually it's only a problem with cheapie controllers because their control systems can't react in time to keep things reasonable. A 12V inverter being hit with a 50V spike may result in failure.

 

[efficientPV]

I have a small battery and over paneled. One of my charge controllers has a nasty problem with overshoot when a load is removed. I'm talking a small load like a 150W soldering gun. When I click off, it goes into over voltage >15.5 and it turns the inverter off with a fault 50% of the time. When I'm doing soldering I have to switch the panels off that CC. I don't understand it either. I've done similar charge controller designs and always included a clawback for over voltage. Simplest piece of software you can write. over voltage, immediately remove FET drive. I communicated with the actual manufacturer in China, lumiax.com. They never responded.

 

[gnubie]

I put it down to there simply being no actual technical expertise behind the products, assuming who ever is selling it actually had anything to do with the manufacture of it in the first place. It's not unusual to find pointless components, ie does nothing useful, in all sorts of things out of China because at some stage along the copy of a copy of a copy that is how things are done in China someone has tried to fix a problem and everyone downstream from there has simply copied it. The same thing would apply to firmware no doubt.