DIY inverter software upgrade to combat heavy shading

[curiouscarbon]

for what it is worth, i am one of the unwashed masses, and endeavor to use one (off the shelf) mppt for each series string. addiction to statistics about performance is one part of it.

and threads like this help me learn ?️

i admire anyone who goes and learns how to create the fundamental building blocks of energy collection/storage systems, and goes and builds! ??️

 

[fourtytwo]

Haha, I ran without a battery for quite some time, and even without a solar controller for a while.
My home brew inverter is very forgiving regarding dc input voltage, a water heater would be even less fussy.

But most people here run multiple parallel connected strings to charge a battery, which changes the requirements rather a lot.
So what you are trying to do is kind of unique.

I agree the water heater is less fussy, the advanced course if this works will be to make it work with the grid as a load

The whole false peaks problem , as far as I can see, can only be resolved by doing a sweep of pwm over the whole likely operating envelope.
That would be complicated by the large energy storage electrolytic required between the panels and the buck converter.
The sweep would need to be kept reasonably slow, and even that might get confused by tree branches swaying in the wind, for example.

That's exactly right and that's what the scope shot shows the controller doing. Swaying tree branches would be an interesting case (fortunately I don't have any) if they caused a substantial difference in the portion of the array shaded although this is where true mppt tracking helps as it is not to dissimilar to a passing cloud.

One very large series string poses some rather intractable problems with partial shading.
With really chronic shading, the inverse diodes are supposed to bypass current around a panel, or sections within a panel that are not contributing.

But as we all know, this creates some very peculiar effects with false peaks when the shading is not quite enough to turn the shunt diodes completely on.

Exactly as the plot at the beginning of this thread shows, if you have a hill climber that starts at the high voltage end (most do) it will lock onto a false peak with only a fraction of the truly available power. By taking time out to do a full sweep of voltages the true peak (highest power) can be identified and the tracker forced to begin there, once there the slopes either side of it will prevent the tracker from wandering back to it's old false peak.

Even if some kind of clever circuit was fitted to each panel to somehow replace this simple shunt diode behavior somehow with an algorithm, I doubt if it could ever be made to work properly.

After thinking about it for a while, my first attempt at this might be to split up the series string into completely separate isolated panels.

Fit an individual mppt algorithm to every panel, coupled to an individual boost converter to step up the single panel voltage to drive the water heating element directly. Some testing could be done with just one panel contributing to what you already have.
The eventual system would have each panel contributing individually and independently to the resistive heating load.

It should not be too difficult to measure dc voltage and current at the panel, and establish a suitable duty cycle from a perturb and observe algorithm. A discontinuous boost converter does not care what the dc output voltage is, it just transfers power into a load. The voltage across the heating element is going to vary widely, and the boost converter does not care. Its totally different to an almost constant voltage battery charging load.

There may be a better solution to all this, but its the only practical solution I can think of right now that does not introduce more problems.

As I mentioned in a previous post I had considered per-panel electronics but this raises the cost of a panel substantially, in fact off the top of my head I would think it quite easy to double the panel cost. As I have a strong imperative to make the system pay for itself within a reasonable time I decided against that type of solution but it may be suitable for others less concerned with economics

 

[fourtytwo]

for what it is worth, i am one of the unwashed masses, and endeavor to use one (off the shelf) mppt for each series string. addiction to statistics about performance is one part of it.

and threads like this help me learn ?️

i admire anyone who goes and learns how to create the fundamental building blocks of energy collection/storage systems, and goes and builds! ??️

I feel kinda sorry for people purely reliant upon commercial products because from what I have seen there are many many false claims especially with regard to phrases such as MPPT. The solar industry seems particularly riven with fakes IMOP

Thank you for the compliment, I was lucky to spend my working life in electronics of one form or another, power being just one and it also being one of my lifetime hobbies

 

[Warpspeed]

Happy is the man that enjoys his work.
When work also becomes a hobby and a lifetime interest, it does not get any better than that.

 

[Hedges]

Whenever I think a commercial product is overpriced, brief evaluation of the time (and cost) it would take to develop something different shows what a bargain off-the-shelf is.
Any research lab ends up spending millions for custom capabilities in things which would be much less in volume (e.g. a camera)
But compared to commercially offered services, DIY isn't unreasonable even though it takes me three times as long as they would. I get the benefit of understanding it, and avoiding some compromises "professionals" make.

 

[fourtytwo]

******************* HAPPY NEW YEAR EVERYBODY *****************
Well Warpspeed as I say I was lucky, still lots of office politics but there was always the engineering to fall back on
Hedges sometimes commercial development costs lots more than it should do because office politics and fools get in the way
at least with DIY neither apply so the job can be done much more efficiently

 

[curiouscarbon]

?️??

 

[fourtytwo]

What passes for sun at 53.5deg North in January did a brief appearance today so I rushed around with the camera and got a pic of the LH side shaded together with a couple of shots of the LCD (watts left, volts right) showing voltage agility picking much more power than if it had stayed fixed!
I don't actually know if it was the normal mppt that followed the panel voltage down (probably) or if it was the new anti-shading scanner as it's just toooo cold to stand in the garage watching the display! Maybe I should get myself a wifi cctv camera just so I can watch it from my desk or even spend some time writing code for an esp8266 thingy I bought for vaguely that purpose some time ago, but like all non-essentials it tends to sit around on the backburner for a long time
That was a lucky strike, 20 minutes later and it's pouring with rain hahaha

 

[Warpspeed]

What passes for sun at 53.5deg North in January did a brief appearance today

Oh my 53.4 North !
I spent a year at a remote weather station on Macquarie Island in the Southern Ocean, 54.50 South as the radio communications guy.
Penguins, seals, snow ice sleet and freezing wind.
Australia has a much nicer climate.

 

[fourtytwo]

Australia has a much nicer climate.

Tell me about it! You guy's have sun when you have the maximum electrical load (Air-con) I wish.......
So we got a little more sun today, rushed about with the camera again and got a range of shadows and operating conditions. It locked onto true power peaks at ~30, ~60 & ~90V representing 1, 2 & 3 unshaded panels completely automatically
Previously power under these conditions would be ~50W as a result of the fake high voltage peak (see post#1 plot).

 

[fourtytwo]

The final part of the solution was being able to check performance from my desk. For this I attached an ESP8266 WIFI processor to the back of the GTI LCD module shown above to capture & decode the LCD data. It runs a WIFI server that downloads a graphing Java script to clients so there browsers display PV volts & Watts, so now I just connect to it from anywhere with Firefox (or any other browser) and get this :-

The spikes are of course the new anti-shading software checking mppt is on the true maximum power point & not some fake caused by shading.
If the sun ever comes out sufficiently to cause real shading I will post the result here. For now at 53.30' North on an overcast day from 1Kwp we have 60-100W

 

[fourtytwo]

Here is a plot from today showing the scanner correcting the MPPT voltage a few times to good advantage (while the panels were part shaded) although most times it runs it doesn't find anything better.

The watch from my desk via the browser is something I should have done years ago rather than sitting in the cold in the garage or worse outside watching a scope trace! Even better it only cost me something like £5 for the ESP8266 board vs ~£400 for the scope (I accept it took me a week to write & debug the code).

 

[fourtytwo]

Greetings! On a freezing cold January day I got this from my 1Kw ground mount

Same old shading problems, I would love to fill in those gaps where the power production is near zero, for those with very poor solar resources every little counts.
EDIT 16th Jan, I found out what the gaps are, they are a breakdown in communications between the inverter and the logger, possibly caused by EMI at higher powers, currently under investigation. Relieved it's not MPPT playing up!!
This is a completely new logger only put into service a few days ago, it logs up to 4 years and display's graphs in various degree of detail.


The latest hour/day/week/month being on the right, all on the same web-page, simple to scroll down.
So I have to look at MPPT/Scanner again, something I have noticed is there seems to be an upward voltage bias in MPPT as often when the scanner parks it at a new starting voltage it will progressively increase despite power falling. The problem I think is traditional MPPT does not have enough history to work from, if it had the trend would be as obvious to it as it is to the human eye on a graph.
The other issue that is hard to crack is on cloudy days often the Insolation of the panels will change substantially during the scan giving false results. This is hard to fix as due to Cin the scan speed is limited especially at low powers.
So I hope there might be something of interest to someone in there Meanwhile a great 2024 to you all.