Hi guys. I found an article on the web where it stated that you can put more panels on the roof than your inverter specs. It was stated due to losses you can go 133% over the specified watt.  It said from 3kw after losses you normally get only 2.4 kw. so you can put 4kw of panels on the roof. Is this true. My mecer inverter specs are 4500w on the roof. I want to go to 5200w. Will it be safe?

20 minutes ago, vulgrim said:

put more panels on the roof than your inverter specs

Not to be pedantic, but it is more about the MPPTs specs than the inverter. For most MPPTs, if you go over the limit, the MPPT simply cuts off the top and runs at whatever its maximum capacity is (sometimes with a bit of temperature derating). Since you only make the peak power for maybe an hour or two in the day, you lose very little. You can gain more production in the rest of the day as well. Visually, you take the normal bell curve, make it fatter and cut off the top. Additionally, since PV modules only run at their STC rated power when conditions are absolutely perfect (and around 80%-90% the rest of the time), it means that most days it's not even going to clip anything anyway.

I usually advise oversizing no more than 20%. MPPTs can experience overcurrent events when solar conditions change quickly and the control algorithm doesn't adjust fat enough. They are designed to handle these things (usually by restarting), but it remains best to design the system to avoid unnecessary overcurrent events. From experience, I can tell you that 50% oversizing gives you lots of overcurrent events! Also, you will be running that MPPT at maximum power for long periods. If it is a quality unit like an outback or a Victron, I'd think nothing of it, but I cannot speak for the others.

I'll see @plonkster, and raise it a step further.

It is a misconception that all panels to an MPPT have to be the same pitch and direction.

Only the the panels in a string should be like that, but an MPPT can have multiple strings.

It is not long in your solar journey before you'll conclude its better to have power throughout the day than a glut of unusable power at noon.

So I'd advise not only over-paneling, but over-paneling in the morning and over-paneling in the afternoon.

So some strings are picking up as others are dropping off, and at noon they're both doing half/half.

So yes, there can be quite a bit of headroom if you do it right.

By the way, the panels in the morning are colder than the panels in the afternoon, and warmer panels are less productive.

Edited April 26, 20197 yr by phil.g00

I'll see Phil and Plonkster and raise them both.

It boils down to a very simple line drawn in concrete, not sand: All MPPT's have a max voltage and amperage one can connect to them. Exceed that and they have a tendency to blow.

As an example, here are some of the Victron controllers absolute max, never ever to be exceeded:

MPPTs are so cheap, just get another one then you don't have to worry.  I think the problem arises where people have an all in one inverter/charger/mppt/coffee machine and you push up against the limits of that one device. Nothing stops you getting an separate MPPT to just do battery charging. 

Edited April 26, 20197 yr by DeepBass9

My installer wired 4x250W, 60 cell panels in 6 strings to two MPPT150/70 units. There was a long DC run from rooftop to battery that was unavoidable. The MPPT would work fine except in cases where it was cool and cloudy - the moment a cloud moves and lets bright noon sun on the panels over voltage alarm triggers. This went on for some time without damage but in the end the array was reconfigured as 3x250W panels x 8 strings and the issue went away. Did not really notice much in the way of lost power due to the lower voltage. Got about 5KW typical max in the day (6KWp panels) - still get that same amount. Highest recorded a tad over 7KW - during the winter.

Will be reconfiguring for a grid tie at around 600V soon - will be interesting if I see much difference in power.

Still waiting for COCT - but got an e-mail with a minor query yesterday and it seems all is well.

On 2019/04/26 at 5:01 PM, plonkster said:

I usually advise oversizing no more than 20%.

For a long time, Axperts didn't provide a realistic maximum PV power specification. If they even provided one, it was the maximum battery-side charge current multiplied by a rounded battery voltage (50 V / 25 V / 12.5 V).

Recently, they have been publishing a number that is different to the above calculation. For example, where an 80 A (battery side) solar charge controller used to be specified at 4000 W, it's now specified at 4500 W. So either they're using a battery voltage of 56.25 V (seems unlikely), or they mean it really is designed with only 100% x 4500 / 4000 = 112.5% of rated output.

So now I say that Axperts should only be oversized by a maximum of 12.5%.

I suspect that if they improved their algorithm a bit (especially, prevented integral wind-up), then they could handle much higher over-sizing.

Does that mean I fold after several players raised the bets?  🤠