Simple solar voltage clipping circuit

[SimonSolar2C]

Thanks Tony - I am also thinking a 6.2v zener shunt is the quickest KISS solution. I tried it and I did have problems with the amount of heat it puts out when there is no load. Was worried it would melt the solder holding it.

 

[Tony Stewart]

A series LDO regulator would reduce the waste heat. with a buck regulator being most efficient. (Comparator with choke)

 

[SimonSolar2C]

I havent found a solution. Thanks for the graph Tony - yes I understand MPP of solar panels, I just need to trick the iPhone into starting charge and drawing current.
So my Solar Panel is 12 cells- delivers about 7V oc. The phone will charge if it sees 6v or less, which happens once t starts drawing power.
Ive been trying to come up to speed with LDO Linear regulators.
If I use a 5V output, then there will be probably 1.2v diference under load. Thats more than the LDO value. I dont understand what happens to the missing power. Doesnt this mean it will be disapating some as heat?
That's not the desire. It seems to keep coming back to using a simple zener shunt, but that creates too much heat when the phone is not connected. ( It could burn the person wearing this, or perhaps melt the solder joints)
- Can someone recommend the LDO example I should be using (preferably under $1 in quantity). Shouldnt it be a switch mode version if I want high efficiency?
- Also are the capacitors absolutely necessary?

 

[Pommie]

If this was me I'd try one of **broken link removed**. Just adjust it to give the required output. What can you loose for $1 with free postage.

Mike.

 

[Misterbenn]

I dont understand what happens to the missing power. Doesnt this mean it will be disapating some as heat?
That's not the desire. It seems to keep coming back to using a simple zener shunt, but that creates too much heat when the phone is not connected.

A linear regulator will drop any extra voltage across the device and 'burn' it as heat. The total power wasted will be the voltage dropped multiplied by the load current.
The zener shunt does a similar thing but burns even more energy. A voltage drop is created across the series resistor and the extra power burnt as heat again voltage drop multiplied by output current. But with the zener solution, when you are not charging the circuit will 'burn' all the solar energy created. It will be voltage drop across the resistor multiplied by output current + zener voltage drop multiplied by output current. So the circuit will get very hot!

So - if you " don't desire" to dissipate energy as heat, a zener is not the right approach.
If you want to have the best efficiency you will need a switch mode regulator, these can have have efficiency >95%. They are more complicated circuits but you can buy them from eBay.

If you insist on a simple circuit with reasonable efficiency you need to use an LDO ( as several people have already tried to explain). An LDO will:

• have almost no loss when not charging as there is no load current (most LDO specify a minimum loading <1mA)
• have reasonable loses when dropping from 7V, just voltage drop multiplied by charging current
• have lower losses and maintain good output voltage when the solar panel drops.
• be a compact size (many surface mount and small packages avalible)
• be simple in that it will only require some input & output capacitors for stability

Can someone recommend the LDO example I should be using (preferably under $1 in quantity). Shouldnt it be a switch mode version if I want high efficiency?

It can be switch mode but you seem to have discounted these already because they are too complex?
IFX25001TFV50 is a potential LDO option it gives a 5V output at input voltages down to 5.5V with output currents up to 400mA

 

[SimonSolar2C]

Hello again, I have followed many of you, advising to use LDO regulator.
I tried using a low drop out regulator (https://www.farnell.com/datasheets/...09.680706486.1494294481-1373478196.1490314122) hooking it up in a standard manner.
Sadly whilst this produces 5v at the output, it does not pull the solar panel down at all, into its MPP, so little to no current is generated. (20ma)
I only need to regulate the voltage when no current is being drawn, once the phone starts pulling power, I suppose I want to bypass the LDO regulator.
Im surprised I cant find a simple solution for this.
Getting pretty frustrated now.

 

[Nigel Goodwin]

Im surprised I cant find a simple solution for this.

You can, such items are standard components used with solar panels - here's one (nice low cost) solution:

https://www.banggood.com/5V-2A-Sola...-Regulator-Module-6V-20V-Input-p-1099540.html

 

[Tony Stewart]

Hello again, I have followed many of you, advising to use LDO regulator.
I tried using a low drop out regulator (https://www.farnell.com/datasheets/...09.680706486.1494294481-1373478196.1490314122) hooking it up in a standard manner.
Sadly whilst this produces 5v at the output, it does not pull the solar panel down at all, into its MPP, so little to no current is generated. (20ma)
I only need to regulate the voltage when no current is being drawn, once the phone starts pulling power, I suppose I want to bypass the LDO regulator.
Im surprised I cant find a simple solution for this.
Getting pretty frustrated now.

How did you test this?

Understand that a PV is a solar controlled current source with a zener limit at Voc.

With no load an LDO does not pull down the PV, rather the LDO only pulls up the output. When that load pulls the PV down V with a rise in current, I call that the incremental negative resistance or simply ESR .

ΔV/+ΔI=ESR I want you to test this and plot V vs I so I analyze your results. Use a heatsink for now and dont worry about heat dissipation yet.

You can measure this using any biased power transistor as an active dummy load with a heatsink dump the heat while you regulated the load with a pot, such as into a Darlington NPN. This will simulate various conditions of charge in battery.

THe simple LDO works best when it's output is 72~80% of Voc which is the mild overcast to full sun MPT voltage. ( although some PV chemistry can increase to range from 80~90% in better versions)

72% of your 7.2Voc= 5.18V making that LDO almost perfect for the job.

Plot your LDO output V vs I with the pot controlled Darlington as I said, and lets fix this simple problem.

recall I said.... If Solar panel is Voc=7v, Isc=280ma it cannot deliver Voc*Isc =2W Get actual curves if you can. My guess is 1.5W max at 6V.

I trust you know what a Darlington NPN is and how to use.

 

[SimonSolar2C]

Tony - I did a very simple initial test - shorting the output through a small resistance with amp meter in series. The supply was the solar panel, therefore current should not exceed the limits of the device. I expected to see the Isc value of 280ma in full sun.
I can try your test with a 10 turn variable resistance and plot IvsV but I'm more of a software guy, so actually not sure how you want the darlington set up.

 

[SimonSolar2C]

You can, such items are standard components used with solar panels - here's one (nice low cost) solution:

https://www.banggood.com/5V-2A-Sola...-Regulator-Module-6V-20V-Input-p-1099540.html

Thanks Nigel - this buck converter looks nice but I need to build this for production into a USB plug, and only need 300ma not 2A, thanks again.

 

[camerart]

Hi,
I haven't read the whole thread, so this may have been already suggested.
I have solar panels, and controllers. What appears to happen is: 12V battery-24V solar panel. The SP charges a capacitor to the Battery charge voltage, then the capacitor uses its charge to charge the battery, then is switched back to capacitor charge and so on. This is happening very quickly especially when it is sunny, but when the panel is just above battery charge voltage it takes longer to charge and switch.
From this description (I can see holes) it may help with your design?
Camerart.

 

[SimonSolar2C]

Hi,
...The SP charges a capacitor to the Battery charge voltage, then the capacitor uses its charge to charge the battery, then is switched back to capacitor charge and so on....[snip]
Camerart.

camerart thanks I had just been thinking the same - that it needs something on the input side doing PWM with the capacitor like it is on the output, so we can set the voltage at the optimum, or better, the max current or MPP
It would need to oscillate the input capacitor's voltage near to the MPP for efficiency, deliver it to the LDO regulator circuit discussed above.
Is there such an IC circuit without getting into the whole MPP tracking discussion?
And if so, can it just be 'bolted on' or is there some impedance matching or some such needed?

 

[MikeMl]

I haven't read the whole thread, but I have a comment, too.

I have a boat and truck parked at a remote location that get used infrequently. I needed to keep several batteries from self-discharging in hot weather while the equipment sits unused, sometimes for months at a time. I use a solar panel (~10W, 21V open-circuit voltage) to keep three large 12V flooded-cell batteries up. I use a shunt-regulator, and a multiplexor to connect the charger to the three batteries on a round-robin basis. That way, if one battery shorts a cell (the most common failure mode), the other two batteries are uneffected.

I have all the time in the world, so operating the panel at MPPT is never a requirement. I found that the shunt regulator is the simplest, and very reliable. It uses a NPN power transistor in series with a wire-wound power resistor to load the panel voltage down to 13.6V when the battery is fully charged (i.e. the long-term float voltage for the batteries when the battery is not drawing any charging current).

I chose the power resistor so that it dissipates about 7W while the NPN transistor dissipates about 3W to absorb the 10W output of the panel when no power is required to charge the batteries (~99% of the time when the sun is shining). I use a TL431 to sense the battery voltage and to control the base current of the NPN to create the precision shunt-regulator.

 

[camerart]

camerart thanks I had just been thinking the same - that it needs something on the input side doing PWM with the capacitor like it is on the output, so we can set the voltage at the optimum, or better, the max current or MPP
It would need to oscillate the input capacitor's voltage near to the MPP for efficiency, deliver it to the LDO regulator circuit discussed above.
Is there such an IC circuit without getting into the whole MPP tracking discussion?
And if so, can it just be 'bolted on' or is there some impedance matching or some such needed?

Hi S,
What is MPP?
How I see it, is: As soon as the capacitor1 hits the correct charging voltage, the Solar panels switch over to 'say' another capacitor to charge, while the first one is charging the device, cap2 is charging up to the correct voltage. As soon as the voltage of cap1 is below the charging voltage, switch back to cap2, and so on. The speed of the switching is controlled, by the caps voltages.
I haven't thought it through, but just my assessment of how my chargers work, after observation. Of course currents, over voltage and safety etc need to be taken into consideration.
C.

 

[MikeMl]

What is MPP?...

Maximum Power Point. Every panel has a "sweet spot" where the product of the panel voltage and the panel output current hits maximum Power output. If you are trying to recharge a battery in the shortest possible time, then you would like the charge controller to operate the panel at Vmppt.

The Vmppt for a 21Vopen-circuit panel is typically about 16V, so if you load the panel down to the battery charging voltage (~14V), it takes slightly longer to charge the battery than it would if the panel output voltage was kept at 16V. However, the decrease in charging time by using an expensive MPPT controller vs just connecting the panel to the battery with either a series or shunt regulator is so small so as not being worth the trouble...

 

[camerart]

Maximum Power Point. .

Hi M,
Thanks.

 

[SimonSolar2C]

Today the sun shone and I plugged in a real load and the LDO is working. Somehow the Isc test I did previously made no sense, but actually using it to charge my phone works. !
Many thanks to everyone for their comments here.
As for mefficiency, with the solar panel connected directly to the phone, in slightly cloudy sun, I get less total power to the phone than with the LDO in circuit. That doesnt make much snese unless the sunlight moved or the Maximum Power Point is actually better with the LDO in circuit.
I'll do some more tests tomorrow with a more robust test platform.
Tony Stewart spec Misterbenn If you had any thoughts on improvements that could PWM the solar input side also, so we can hard wire the MPP in average conditions, that would be a nice improvement.
BTW The iPhone seems to have some sort of control in it, where it starts up drawing 150ma, and I'm assuming if the voltage drops too low, it drops back its current draw to just 89ma. Tried to search on info about this to no avail.

 

[camerart]

Hi s,
Don't forget to 'aim' the panel towards the sun.
C.

 

[Spek]

Hi I'm new ! Say, how about just a 7805 voltage regulator? Takes any (?) dc input and holds the output at 5v (maybe slightly higher).

Spek

 

[Tony Stewart]

7805 has a 1.5 to 2V dropout . FET or CMOS LDO’s >50mV]