Emulating solar battery function

[greengeek]

I think this device might offer me some options:

https://www.trademe.co.nz/electronics-p ... 129865.htm

I am just waiting to hear back whether this is definitely both Buck and Boost.

Two or three of these in parallel might be enough to stabilise the output voltage of a solar panel array at it's maximum power point while funnelling all that power to an inverter that outputs 240v mains to an aircon.

...with maybe a supercapacitor on the input of the inverter.

That's my theory anyway...

[8Geee]

Frankly, is there any reason you can't just power DC-rated equipment, such as computers, smartphones, etc using a 12V output (or more modernly 5V). I recall that camping and automobile equipment (including stove, 'frige, AirCond, Car Entertainment, etc) are all designed to operate 12VDC (some with Nat. Gas). In theory one could swwitch off the AC completely and go 12VDC.

Just fuel for thought
8Geee

[rockedge]

Good point 8Geee!

12v and 24v systems like on many sailing yachts

[Moose On The Loose]

Good point 8Geee!

12v and 24v systems like on many sailing yachts

Aircraft often use nominal 28V stuff.
When searching for a DC powered thing, this may be helpful.

[BarryK]

And thanks for the link to your lithium project. Lots of good info there.
Especially the comments re lead acid limitations:
https://bkhome.org/news/202004/portable ... mping.html

You have reminded me: another item to add to that lead-acid versus lithium comparison is discharge/recharge efficiency. For lead-acid it is 70-85%, for lifepo4 about 90-92%.

I think that 70% lower figure for lead-acid might be if the battery is becoming a bit sulphated.

That loss of 15-30% for lead-acid will be as heat.

Another comparison item might be charge current. I think lifepo4 can take a much higher charge current. However, I am new to lithium batteries and still learning, don't know much about that one.

Also, the MPPT regulator has some loss, as heat. I have seen a figure of 95% efficient for one of them.

[greengeek]

Frankly, is there any reason you can't just power DC-rated equipment, such as computers, smartphones, etc using a 12V output (or more modernly 5V). I recall that camping and automobile equipment (including stove, 'frige, AirCond, Car Entertainment, etc) are all designed to operate 12VDC (some with Nat. Gas). In theory one could swwitch off the AC completely and go 12VDC.

Most of those - yes - i do already run them from 12v and 24v battery banks.

But I have not seen an aircon unit that runs from a 12v battery - i have only seen units that run from the mechanical drive of the petrol/diesel engine (driven directly via belt).

Although i guess that is one way I could achieve my goal - I could use my solar panels to directly drive a massive DC motor and have that motor coupled to a car aircon pump.

That's difficult to control though - it would be more preferable to run a proper mains powered portable aircon as they have all the control electronics / timers / sensors inbuilt.

And of course that would still leave me with the problem of my panel voltage dropping below it's max power point, which is the wastage I am trying to avoid.

I really need to find an MPPT controller that does not need a battery. It must be available somewhere surely.

Just a "direct connect" maximum voltage tracking DC regulator feeding an inverter.

I know it must be out there somewhere!!

[Flash]

Greengeek, here's a YouTube video that explains the need to run solar panels at their maximum power point and also a cheap circuit to do it. It doesn't directly address battery charging but it's a good start on understanding what's required.

As far as I know, a charger designed for a lead-acid battery will also work for the same nominal voltage lithium battery if the trickle charge feature is shut off. In other words, the charger puts out a constant current until the load (the battery) reaches its fully charged voltage, then the charger simply shuts off.

[Moose On The Loose]

Greengeek, here's a YouTube video that explains the need to run solar panels at their maximum power point and also a cheap circuit to do it.

That circuit is extremely crude. The ideas explained are good but not complete. Watch the video for the ideas and not the design.

If anyone cares enough we can continue this discussion further.

Slightly off topic:
If you are going "off grid" on your power, the power consumption of stuff matters a lot. If you are computing efficiency matters. This means an ARM based machine is better than an X86.

LED lights are more efficient than CFL but their power supplies tend to impose a capacitive load on the mains power. Some inverters do less well working into capacitive loads. Generally a bigger brighter LED light will do this less than several smaller ones that add up to the same.

[Flash]

You're right that the video should be watched for the ideas, but I wouldn't call the circuit crude, just very basic. It elegantly does what it is supposed to do, but only at a particular level of illumination of the solar panel. The potentiometer would need to be adjusted as the sunlight level changed, to keep the circuit oscillating about the maximum power point. Perhaps the addition of a CdS photocell somewhere in the potentiometer circuit of the NPN transistor and pointed at the sky at the same angle as the PV cell, could keep the circuit at the maximum power point as the sunlight level changed. The circuit doesn't need to be perfect, it only needs to work well when the sun is high enough in the sky to generate usable power.

[Moose On The Loose]

but I wouldn't call the circuit crude

That's OK, I did it for you

The pot needs to be adjusted from light level, temperature and load.

The switching of the PMOSFET is slow leading to losses.

It uses the inductance of the motor as its working inductor.

It didn't use even one 555 timer chip

[Flash]

Despite all those failings, that simple circuit succeeds in putting far more power into a useful load than the solar panel by itself was able to. And he does mention that it might need an inductor if the load isn't inductive. Also, I think the hysteresis provided by the feedback resistor provides a snap action that speeds up the turn-on and turn-off times. Without that feedback resistor, you'd be right about the slow turn-on and turn-off times. Only an oscilloscope could say for sure. I'm surprised the guy didn't use one. I'm sure he must have had one handy.

[Moose On The Loose]

Despite all those failings, that simple circuit succeeds in putting far more power into a useful load than the solar panel by itself was able to. And he does mention that it might need an inductor if the load isn't inductive. Also, I think the hysteresis provided by the feedback resistor provides a snap action that speeds up the turn-on and turn-off times. Without that feedback resistor, you'd be right about the slow turn-on and turn-off times. Only an oscilloscope could say for sure. I'm surprised the guy didn't use one. I'm sure he must have had one handy.

LTSpice can be used to put such circuits to the test. It is very good for this sort of circuit. I haven't tried a model of this myself but it would be fairly easy to do.

The hysteresis won't quite cut it in my opinion. It can only start to act once a quite significant voltage change has happened at the drain of the MOSFET. This means it starts off late in trying to speed up the gate drive.

Is the guy who started this still here?

If so I may spend a little time on making a design for a circuit that auto adapts to the panel and load etc.

[Flash]

Greengeek is still active in the forum. I don't know if he's following this thread any longer. He's not the only one who would benefit from a simple circuit that keeps a solar cell at its maximum power point while doing something useful like charging a battery. Barry posted in this thread too. So please do give it a try. Your effort won't be wasted.

If you post any schematics, please use a format that the forum will show, such as jpeg. I've always admired hand-drawn schematics with pencil on graph paper. Computer-drawn schematics just don't seem to have the right proportions. For one thing, the text that labels the components is generally much smaller than it needs to be.

[greengeek]

Greengeek, here's a YouTube video that explains the need to run solar panels at their maximum power point and also a cheap circuit to do it.

Thanks Flash - awesome video and exactly the sort of focus I am looking for. And the step by step explanation he gave is very helpful. It confirms my feelings that very few components are required to allow the panel to float at/around it's max power voltage - especially when I just need raw voltage output rather than specific voltages/currents for charging of lead acid batteries.
To be able to harvest max power and drive it into a motor load (or storage capacitor) is exactly what I need.

Is the guy who started this still here?

Yep, still here and still keen to get max power out of my panels.

[Moose On The Loose]

Yep, still here and still keen to get max power out of my panels
.

Today I work.
Tomorrow I am off so I will take a run at it then

What is the nominal voltage of the panel?
What is the nominal voltage of the battery?
Initially, I won't include any "charge controller" section.

I will use www.digikey.com as "the part is easy to get"
That is true for US. They will ship by mail

I won't use a micro

I will try to make it easy to understand.

[greengeek]

Awesome, thanks Moose.

Nominal MPP of the panel is 17 volts more or less (some variation between panels but that's a good value to begin with).

My battery banks use 12v lead acid batts with one array of 24v made up of the same 12v batteries in series. Max charging voltage approx 14.4v per battery (no extended gassing or equalisation).

BUT - I really don't want to build this circuit around the gluten free sensitivities of lead acid chemistry - I just want max power out if the panels into a Super capacitor and then directly off to a load.

One question in my mind is how to regulate the output to 13.2v or similar if it is feeding straight into a 12v-230v inverter. I guess a standard DC/DC converter.

[Moose On The Loose]

Awesome, thanks Moose.

Nominal MPP of the panel is 17 volts more or less (some variation between panels but that's a good value to begin with).

My battery banks use 12v lead acid batts with one array of 24v made up of the same 12v batteries in series. Max charging voltage approx 14.4v per battery (no extended gassing or equalisation).

BUT - I really don't want to build this circuit around the gluten free sensitivities of lead acid chemistry - I just want max power out if the panels into a Super capacitor and then directly off to a load.

One question in my mind is how to regulate the output to 13.2v or similar if it is feeding straight into a 12v-230v inverter. I guess a standard DC/DC converter.

There can be a fairly big problem with going from a panel to a DC-DC and then to an inverter.

Lets imagine that the load is 1W so my math is easy.
We will also go with 100% efficiency.

The DC-DC needs to draw 1W from its input.
At a 10V input, it needs to draw 0.1A
At a 20V input, it needs to draw 0.05A

Lets imagine that the panel can only make 0.9A
When the sun first hits the panel, the voltage on the panel starts to rise.
The voltage will get up to where the "under voltage" circuit of the DC-DC allows it to start. When the DC-DC starts, it draws 0.1A and thus the voltage falls.
The voltage will fall until the "under voltage" circuit shuts down the DC-DC and then it will rise again.
The system will lock into oscillating in this range and never have the panel rise up to near 17V

In am still "mid-ponder" on the suggested circuit.

I have the beginnings of a circuit that uses:
3 each LM311 AKA LT1011
1 dual op-amp of some sort
P-MOSFET

My goal is to make this a circuit "mere mortals" can build, understand and modify. Thus there will be no programmable parts involved nor will there be any "weird circuits" that can't be easily explained.
I am going to base the design on the idea that it only needs to work with an input well above 13V but less than 30V

There will be one slightly odd thing.
Given that 17V is getting near the 20V limit for the gate of a MOSFET, I intend to make all the workings run between the nominal +17V and a voltage of {17-8} V. That is as far as the circuit is concerned, the (-) pin of the panel is not circuit ground.

https://media.digikey.com/pdf/Data%20S ... y2012.pdf

[Flash]

Moose, if it's any help, I'm pretty sure that the output voltage of a solar panel (at its maximum power point) doesn't vary by much between when it begins to put out usable power in the morning and when the sun is directly overhead at noon. https://en.wikipedia.org/wiki/Maximum_p ... ure_Method

I wasn't able to find representative graphs with a quick search of the Internet, but I've seen such graphs and as the light varies, the output voltage at the maximum power point doesn't vary nearly as much as does the current. This makes sense to me, since the open circuit output voltage of a solar panel is a function of the somewhat limited range of wavelengths of the photons that make the photocells work, while the output current is a function of the number of photons thumping into the photocells.

Useful reference:
Maximum power point tracking

[greengeek]

Just want to post a reference link re MPP choice:

https://blog.voltaicsystems.com/select- ... ontroller/

I'm trying to understand if this has some info to offer regarding how i can achieve my intended result.

In addition, the solar charge controller board mentioned above has spot for a custom MPP voltage to be set. Whether you’re using it for a different panel, or our panel in a very specific environment, this option allows you to tailor the board using the information provided in this article.

[greengeek]

Also want to post a possible example of a controller that seems manually adjustable for MPP voltage:

https://www.trademe.co.nz/business-farm ... 977459.htm