Help - My system is set for 1 day a year!!!

[LLTHB]

Hi to all.

Please excuse any errors I make here, I am an amateur at this.

We have installed a Goodwe hybrid grid tie inverter, 5Kw unit.

It was supplied with 20 panels - 2 strings of 10 connected to the 2 MPPT inputs.

Due to the construction of our house in a "H" shape, we throw shade over the north facing panels till about 09h00.

To get around this we have added a 3rd string of 10 panels on our east facing roof.

 

So now at night we turn off one of the north facing strings and turn on the east facing string. So still only 20 panels connected at any one time to the inverter.

At about 09h00 when the full sun is on the 20 north facing panels, we turn off the east panels and turn on the north facing string that was off, so only 20 panels on at a time.

All well and good, works like a bomb.

BUT, BUT, BUT - we have now learned on an overcast day, we can turn on all 30 panels and make more power from less light. All well and good.

But what will happen if the clouds suddenly disappear and all 30 panels are producing max power?

The installer has explained to me that basically my system max spec is a day in October where because of roof angles etc, that is the day my 20 panels will make the most power ever if the sky is clear and that is the maximum input the inverter can handle. Lets call this the "perfect day".

 

So, all other days we can make less power because of angles, cloud etc etc. This seems wrong.

On a cloudy day, we could do with 40 panels, not 20. Why can we not just put more panels up than needed on the perfect day, but each other day we could make the "perfect day" if we could have 40 panels on line all the time, but the inverter must only take what it can handle max.

 

Does this make sense or am I lost here? Why are we limited like this?? Or are we?

 

Any advice appreciated here.

Cheers.

[Antony]

30 minutes ago, LLTHB said:

But what will happen if the clouds suddenly disappear and all 30 panels are producing max power?

Good day LLTHB and welcome to the forum,

I asked Goodwe this very question, should the total max power exceed the MAX input of the inverter, it will blow the interver, see copy of email below.

Oliver Shen	Wed, Sep 26, 8:17 AM
Hi, Mr. Naude,

 

When the PV array operates at full load 6600W, the inverter will be broken, please reduce the power of PV array.

 

 

Best Regards

Oliver Shen

Overseas L2 Technical Support Engineer

GOODWE POWER SUPPLY TECHNOLOGY CO.,LTD.

—————————————————————

M  +86 15062346311    T  +86 512 67375149

A  189 Kun Lun Shan Rd, Suzhou, Jiangsu, China

 

发件人: Antony Naude [mailto:[email protected]] 
发送时间: 2018年9月25日 17:53
收件人: Oliver Shen
主题: Re: Reply: Correct version

 

Good day sir,

 

If you say the MAX is 6500 and my PV array is 6600, what will happen?

 

[Antony]

50 minutes ago, LLTHB said:

It was supplied with 20 panels - 2 strings of 10 connected to the 2 MPPT inputs.

What panels are they?

[Jaco De Jongh]

9 hours ago, LLTHB said:

But what will happen if the clouds suddenly disappear and all 30 panels are producing max power?

Firstly, do you have the loads to consume the production of 40 Panels. If you have, I will suggest you rather make use of external MPPT's to connect the 20 extra panels to your system safely. Running the risk to blow your inverter is not worth while.

I am not sure about this specific MPPT, but not all MPPT's were designed to "protect themselves" against overloading, and will take whatever is available to a point where something blows.

If Your Mppt does have the ability to "Protect itself"  and only allow through what is safe, it will mean that a big percentage of the panels on the roof will not get used, bringing me back to the point of having external MPPT's to ensure the full capacity is available for use should it be needed.  

[___]

We had a long discussion about this earlier, about the maximum current a charge controller has to be able to handle and how that relates to the Isc (current, short circuit) of the array

I think there is something important that we're missing in this discussion, and that is that there is a fundamental difference between some MPPTs in the sense that while the majority of them are buck converters, some are not.

In yet another thread I opined that the Goodwe inverter most likely injects its PV energy directly into the high voltage DC side (with a buck/boost pipeline down to battery from there). If I am right, then the MPPT in the Goodwe is likely a boost converter.

In a boost converter the output current is lower than the input current (and the output voltage, conversely, is higher). This is different to a buck converter where the input current is lower, and as long as you match the capability of the switch component to the inrush current of the inductor you can get away with almost anything, even feeding it from a battery! :-)

So I will have to develop this thought some more (aka study it!), but I'm beginning to have a suspicion that this might be the reason the buck-type MPPTs have such generous limits, while the boost ones seems to want you to size your Isc below some limit.

It could also just be economics: High voltage high current FETs are expensive. That's one reason why other components are usually used on the high voltage side, eg Traics on the AC side, or IGBTs (bipolar transistors with a FET on the gate essentially).

[LLTHB]

Hi to all and thanks for all the input so far.

 

In reply to Jaco - no, I cant handle all 40 panels at once. We are set up to use about 5Kw at a time.

This we get from 20 panels on a bright sunny day, no problem.

But when it is overcast, we can only get about 3.5Kw as an example. We then turn on the extra 10 panels, bringing 30 panels online and this brings us up to 5Kw again. But if I go out and forget to turn off the 10 extra panels and the sun pops out, I risk damaging my inverter as emails from Antony in earlier post show.

So in a perfect world, 20 panels will be great if the sun shines every day. But this is not true so we want to put up extra panels safely.

Bright sunshine - 20 panels = 5Kw

Overcast - 30 panels = 5kw

So we have to look at external MPPTs.

In the old days you set the jets in your car for Jhb or Durban. So if you jetted your carb's for jhb, as you went down hill you could either lose or gain power as things changed, but your carburetor was dumb and was not variable. Modern fuel injectors sense the amount of air pressure to work out O2 levels and and adjust spray to optimize so no matter the altitude your car is efficient.

 

This is the thrust of my argument - how can we put up excess panels to allow for variables like overcast cloudy days. Safely that is.

 

Cheers.

[___]

15 minutes ago, LLTHB said:

So we have to look at external MPPTs.

I think you have your answer. Of course this is less efficient, the power goes down to battery voltage and then back up again via the inverter to line voltage, so you will be taking around  5% hit in efficiency. But it is safer. Depending on your battery, you might be able to just leave it permanently connected (assuming it can take the whole 7.5kw on one of those rare slightly-overcast days with huge amount of sunlight+backscatter.