Solar LED Post Circuit Board questions
Attached are pictures of solar light and circuit board. The CB diagram I drew up I think is correct.

Measurements I took with digital multimeter:

1) Current draw from full battery is 55ma and over time voltage drops to 0.8v and the LED goes out at 4ma.
2) The voltage across the LED is equal to the battery voltage and the LED will light even if the battery is down to 0.9 volts (no load measurement).
3) Advertising ‘spec’ says with full charge LED puts out 3.6 lumens, 3300 deg K.
4) With ‘low’ battery: 60 watt bulb 6 inches away solar panel puts out 25ma and after several hours the voltage reading is 1.7 volts or better and when the battery is removed the battery itself is at 1.4v.
5) The solar light is 2 years old so I assume the LED is one of the newer ‘generation’ types.


1) Is the brown cylindrical a diode to prevent draining the battery back thru the solar panel?
2) The manual switch is used when storing the fixture so the battery doesn’t drain to ‘nothing?
3) Are my calculations for the resistor correct?
4) Inductor: Calcs correct? Is the IC switching for the inductor to work? What determines the voltage ‘boost factor’? Construction? Switching speed?
5) 50ma draw from battery: Does the voltage boost inductor decrease the ma going to the LED?
6) When the battery voltage decreases does the large ma current decrease a lot because of the ‘exponential’ relationship in the LED of volts and current or because the battery is ‘empty’?
7) Voltage across LED: Does my multimeter read only the battery voltage because it is unable to read the actual voltage because of the high ‘pulsing’ required for the inductor?
8) What does the light sensor do? Is it a ‘switch’ or does it provide power to the IC 5252F?
9) Is the light sensor also ‘protection’ to ensure the LED does not go on when the solar panel is producing voltage higher than 1.3 volts?
10) What does the 4 pin 5252F IC do in simple terms? (Pins 1-4)
11) Since volts and current are relatively low, does the battery from the digital multimeter affect the readings?
12) Any idea on how a 60 watt bulb 6” away from solar panel compares to bright sunlight?


I would appreciate any help in explaining how/what the circuit board, IC5252 does. I am assuming the ‘clear’ LED is classified as a white led. I am in the learning phase and am continuing to try to understand ‘electronics’ so that I can design my ‘modifications’.

 

4-pin chips are not the most common item in the electronics toolbox so you stumbled upon an odd device for a "learn electronics" project for reverse engineering.

Without seeing the side view of the chip and without trying to reverse engineer myself, I would assume it is either...
a) some type 0f chip that protects the battery from over-charging, or
b) some type of device that prevents the light from turning on when the battery is dead
c) some type of device that does (a) and (b) above.

How old is that lamp (I am guessing 10 or more years). If so, tracking down a datasheet might be an issue, especially if the full part number is not on the chip (missing first few letters).

I would guess something like this chip (voltage supervisor) is possible but, since you don't have a capacitor, not likely this specific chip.
http://rohmfs.rohm.com/en/products/databook/datasheet/ic/power/voltage_detector/bd52xxg-e.pdf


Cheers and good luck.

 

4-pin chips are not the most common item in the electronics toolbox so you stumbled upon an odd device for a "learn electronics" project for reverse engineering.

Without seeing the side view of the chip and without trying to reverse engineer myself, I would assume it is either...
a) some type 0f chip that protects the battery from over-charging, or
b) some type of device that prevents the light from turning on when the battery is dead
c) some type of device that does (a) and (b) above.

How old is that lamp (I am guessing 10 or more years). If so, tracking down a datasheet might be an issue, especially if the full part number is not on the chip (missing first few letters).

I would guess something like this chip (voltage supervisor) is possible but, since you don't have a capacitor, not likely this specific chip.
http://rohmfs.rohm.com/en/products/databook/datasheet/ic/power/voltage_detector/bd52xxg-e.pdf


Cheers and good luck.

GopherT: The numbers 5252F and 1118 below are quite clear. The light is only 2 years old.

 

GopherT: The numbers 5252F and 1118 below are quite clear. The light is only 2 years old.

Most boards like this have a date code, the only one that looks like a date code would be 1118 (18th week of 2011). Therefore, 5252F is the chip number. May be a proprietary chip. Do you have a photo from the side? The little oval top view is not ringing any bells for me.

 

If your schematic is correct then the chip looks like it is driving your LED. If I would take a guess, I would say a switching regulator for the LED but there is no coil. Chip looks way too small to contain it's own. It might me a linear regulator in which case would be very inefficient for a battery application.

 

The 525F appears to be a voltage detector
http://pdf1.alldatasheet.net/datasheet-pdf/view/130683/ROHM/BD5252FVE.html

But it has 6 pins.

 

1) Is the brown cylindrical a diode to prevent draining the battery back thru the solar panel?
YES, the thing between the red and white wires. It's more likely black, with a silver band on one end.
2) The manual switch is used when storing the fixture so the battery doesn’t drain to ‘nothing?
YES, and to keep the light off, if that matters.
3) Are my calculations for the resistor correct?
YES
4) Inductor: Calcs correct? Is the IC switching for the inductor to work? What determines the voltage ‘boost factor’? Construction? Switching speed?
YES, YES, needs a longer answer.
5) 50ma draw from battery: Does the voltage boost inductor decrease the ma going to the LED?
The time-average should be similar, just as the average voltage is the battery voltage.
6) When the battery voltage decreases does the large ma current decrease a lot because of the ‘exponential’ relationship in the LED of volts and current or because the battery is ‘empty’?
It drops because the boost circuit cannot boost voltage enough to drive the current anymore.
7) Voltage across LED: Does my multimeter read only the battery voltage because it is unable to read the actual voltage because of the high ‘pulsing’ required for the inductor?
YES, it would peak much higher if you could see it with an oscilloscope. The meter can only show you the average.
8) What does the light sensor do? Is it a ‘switch’ or does it provide power to the IC 5252F?
It changes form a very high resistance to a lower (but still high) resistance when it sees light, and this switches off the LED, under control of the IC. The CdS sensor does not switch the power itself, the IC does that based on the resistance change.
9) Is the light sensor also ‘protection’ to ensure the LED does not go on when the solar panel is producing voltage higher than 1.3 volts?
Nope, just turns it off when the sun comes up. Back on at night.
10) What does the 4 pin 5252F IC do in simple terms? (Pins 1-4)
Oscillates to run the boost, and turns the oscillator on and off with changing light.
11) Since volts and current are relatively low, does the battery from the digital multimeter affect the readings?
In general, yes, ALL measurements are affected by the process of measuring. How much? Probably not much here.
12) Any idea on how a 60 watt bulb 6” away from solar panel compares to bright sunlight?
I forget, I think it's lower than real sun. You can find this answer online, though.

 

It's a switcher: it uses the inductor ("L" on the schematic.)
Doesn't look like there's any battery-charge control. Maybe the PV-cell is too small to overcharge a battery.

 

Looked and looked did not see an inductor in the schematic. Though I see it in the test of the message. So I was right it is a switcher.

 

The QX5252 is an IC specifically designed to run solar lights. I'm quite surprised by the schematic, as these chips have the battery isolating diode built inside, and they use the solar panel voltage as the dark detector.

More info here.

 

...these chips have the battery isolating diode built inside..

Good catch. Some of these 4-pin ICs do not have that diode, and my bet is someone made a substitution without taking advantage of this "improved" IC. They squeeze pennies but missed this one.

It would be interesting to bypass the diode and confirm it still works, and it should work better.

 

Good catch. Some of these 4-pin ICs do not have that diode, and my bet is someone made a substitution without taking advantage of this "improved" IC. They squeeze pennies but missed this one.

It would be interesting to bypass the diode and confirm it still works, and it should work better.

Wayneh: For understanding purposes, why would it work better? By bypassing the diode wouldn't the battery drain thru the solar panel and also defeat the purpose of the manual switch?

 

As ErnieM noted, the blocking diode is already inside the IC. Adding another one externally adds no benefit but introduces a small voltage drop - and consequent power loss - between the solar panel and the battery. Removing it should add a few percentage points the overall efficiency. I think you'd only see the improvement if you carefully monitor the time the light stays on following a controlled dose of sunlight.

 

I just barged thru Jan's schematic with the 5252 data sheet. The on board diode is required due to a repuropsing of some of the inputs to the IC.

The sole charge path for the battery is from the panel thru the diode. The "normal" use of the internal diode is not used since there is an external light sensor (CDS cell?) is being used to "fool" the chip into seeing that as the voltage source. It may offer some improvement over the "intended" design for dark and under voltage sensing.

 

As ErnieM noted, the blocking diode is already inside the IC. Adding another one externally adds no benefit but introduces a small voltage drop - and consequent power loss - between the solar panel and the battery. Removing it should add a few percentage points the overall efficiency. I think you'd only see the improvement if you carefully monitor the time the light stays on following a controlled dose of sunlight.

Wayneh: The battery and solar panel are 'hooked up together' and the potential diode in the IC is quite far away. In trying to understand stuff I don't understand how a diode in the IC would prevent the battery from back feeding into the solar panel and discharging the battery when it is dark?

 

OK, I was looking at the schematic linked in #10 and not at your drawing. They are not the same, as Ernie just noted in #14. Sorry about the confusion.

 

If your schematic is correct then the chip looks like it is driving your LED. If I would take a guess, I would say a switching regulator for the LED but there is no coil. Chip looks way too small to contain it's own. It might me a linear regulator in which case would be very inefficient for a battery application.

A while back I bought a bicycle headlamp in Poundland - it has a black-blob chip driving 3 actual white LEDs from 2x AAA cells.

If there's an inductor hidden in the black blob - its a very small one!

 

A while back I bought a bicycle headlamp in Poundland - it has a black-blob chip driving 3 actual white LEDs from 2x AAA cells.

If there's an inductor hidden in the black blob - its a very small one!

Ian:
In my schematic and in the photo of the CB there is a green thing which I think is an inductor. Is tha the 'coil' you are looking for?

 

Ian:
In my schematic and in the photo of the CB there is a green thing which I think is an inductor. Is tha the 'coil' you are looking for?

Sounds like it could be - but like I said; if the light I mentioned has it hidden in the black-blob - its a very small one.

 

Yup, the green thing is an inductor and you'll find one in almost every solar light. There are other ways to boost voltage without one, but I believe they are less efficient.