Simple solar voltage clipping circuit

[SimonSolar2C]

Hi Spec - is there an alternative to the Si8824EDB ? Any FET or is that one specifically needed. I think that's the circuit I want to test first.
I'd love to talk to you about LED drivers sometime. You seem to understand the concept of a simple circuit. Shall I start another thread?

 

[spec]

Hi SS,

Hi Spec - is there an alternative to the Si8824EDB ? Any FET or is that one specifically needed. I think that's the circuit I want to test first.

That NMOSFET is special and not many other NMOSFETs have the right characteristics along with an extremely low gate threshold voltage.

I'd love to talk to you about LED drivers sometime. You seem to understand the concept of a simple circuit. Shall I start another thread?

Yes, best to go for a new thread for a new topic- that improves readability no end. There are many members on ETO who understand simple circuits. But thanks anyway.

spec

 

[SimonSolar2C]

Hi spec,
Ive been trying these circuits, thanks again. The option 4 using FET seems ideal for my needs, but Im getting unexpected results.
The voltage on each side of the FET is 6.2 and 4.2
I didnt expect this, because the idea is that the phone will see a voltage below 6.2,. so it considers the line okay to draw power, so it will start drawing power and pull the voltage down on the solar panel to about the ideal voltage of 5.5v at which time I would see 5.5v on both the source and drain of the FET. Then I know the FET is hard on and all power is going to the phone.
Have I misunderstood the workings or calibration of the circuit? It seems it may be dissipating power somewhere, which is not what we want when charging from a solar panel.
I just checked your comment about low gate voltage - and the FET I tested with is not the one you specified. Couldnt get it here. I used a standard 2N7000.
Would the NFET you specified avoid this issue Im seeing?

 

[spec]

I used a standard 2N7000.
Would the NFET you specified avoid this issue Im seeing?

Hi SS,

Afraid the 2N7000 is not the NMOSFET for this job. (Vgt= 3V, RDss= 20 Ohms @ 100 Deg C, Thermal resistance = 313 Deg C watt)

spec

 

[SimonSolar2C]

Can I use the 2N7000, or similar easy to find part, if I drive it with a BC54x like in circuit 3?

 

[spec]

Can I use the 2N7000, or similar easy to find part, if I drive it with a BC54x like in circuit 3?

Short answer is no. If you are going to use an NMOSFET circuit #4 has it. But the NMOSFET must have a low threshold voltage and a low RDsss (resistance drain to source when turned on).

spec

 

[spec]

Hi SS,

You live in New Zealand so you have good access to components, RS for example, so why not get the specified components (RS do free postage).

spec

 

[SimonSolar2C]

Hi Spec,
Thats a fair question. RS dont have it. My goto here is usually https://nz.element14.com/ and mouser.com to check global pricing.
Im a bit nervous about the Microfoot. I'd like to find something easily available that I can prototype with and will fit in my USB plug, like SOT-23-3. Im a bit rusty on reading datasheets to be confident of getting the right one.

 

[SimonSolar2C]

On circuit 3, sometimes the iPhone I use for testing with doesnt draw enough power - only 89ma @~6.2V even though full sun could provide up to 250ma. Its almost like it settles into an equilibrium with the zener.
I havent done enough testing on this to find the cause/fault in my design. Obscuring the solar panel then exposing to the sun and it correctly draws 250ma 5.5v

 

[spec]

Hi SS,

If I were you at this stage I would go for an adjustable low dropout voltage regulator in a suitable surface mount package.

That would give you a Rolls Royce performance and will be the quickest cheapest and most sure way to achieve your objective.

For example, the AMS1117 (LM117) is freely available and dirt cheap.

http://www.advanced-monolithic.com/pdf/ds1117.pdf

**broken link removed**

http://nz.element14.com/webapp/wcs/stores/servlet/Search?catalogId=15001&exaMfpn=true&mfpn=LM1117DT-ADJ/NOPB&showResults=true&categoryId=&langId=64&searchRef=SearchLookAhead&storeId=10189

spec

 

[SimonSolar2C]

Thanks Spec - I ordered some of those parts and the Si8824EDB.
In reading the datasheet for the LM117, does it always maintain a 1.2v difference between input and output? Because that would be less than ideal for my design, because as the phone draws power the voltage of the supply(solar panel) comes down to a point where we prefer the power to be transferred directly, e.g. at between 5.5 and 6v. That would be the most efficient solar charging. When I see a heat sink on the LM117 device I assume it is wasting power somewhere. Am I right or wrong on these points?

 

[mbarazeen]

you can simply put 2 or 3 diodes in series with output of panel, that will drop the voltage as required.

 

[SimonSolar2C]

Hi
Thanks for the suggestion but I only want the voltage to be lower when there is no load. When the load comes on the voltage drops from the solar panel anyway. So I don't want a permanent lower voltage

 

[SimonSolar2C]

So I'm mreading up on the LM117 and in this discussion they talk about it 'burning off' the excess power. I certainly dont want that - I want all my solar energy. Comments?
https://www.adafruit.com/product/2165

 

[MikeMl]

Do you know the difference between a linear regulator and a switch-mode regulator?

Do you know the difference between a low-drop-out linear regulator and a regular linear regulator?

 

[SimonSolar2C]

I know I think I want low drop out and switch mode, but not enough to find/select/build. Also price is a major component for this design.

 

[SimonSolar2C]

Hi MikeMI ( nice plane BTW - is that a 180 or 185? )
Do you have a better solution for my problem?
I dont want to drop the voltage permanently - the solar panel does that when current is drawn - I just want to lower the voltage to <6.2 when no current is drawn.

 

[MikeMl]

Piper PA20 Pacer

Use a 5V low-dropout linear regulator (with resistor feedback to trick it into putting out 6.2V) between the panel and the load. When the load takes all the current that the panel can deliver, the voltage differential across the regulator will be a few tens of mV. As the load stops drawing current, the output voltage from the regulator will stop rising at 6.2V, while the panel voltage will rise toward the panel's open circuit voltage.

 

[SimonSolar2C]

Okay ! Sounds good. I see 14 LDO regulators at mouser: ( The LM1117 I was lookoing at claimed it was LDO, perhaps I need ultra-low?)
https://www.mouser.com/Semiconductors/Power-Management-ICs/LDO-Voltage-Regulators/_/N-5cgac
Some of those at 43cents, but others are >$3 !!
Im not familiar with 'tricking it', or is that using normal adjustable resistor bridge?

 

[Tony Stewart]

You may not get what you expect. I suggest you test it without a regulator and keep an eye on the voltage.

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 think a simple series 1A Silicon Shottky diode would be plenty protection. You could use 2 power WHite LEDs in shunt to make a 6.2V zener. +/-0.1 with some sources like Edison rated for 250mA.

Consider this larger PV array and divide voltages by 3


With 4V charging you can only expect 1W max.and MPT usually occurs at 75 ~82% of Voc.