I'm a involved in a project together with some friends in Tanzania. They sell small simple solar systems, chargers, flashlights etc in Tanzania, often to people in small villages with very low income.
We have discussed a bit if it is possible to manufacture some simple solar products in Tanzania. A flashlight would be a small first try. Not that manufacturing in Tanzania would be cheaper than buying from China, there are lots of solar products from China sold in Tanzania already. But they often break rather quickly and are highly integrated so hard to repair. And even if the small simple solar products are cheap from our perspective, they are not for someone in a village in Tanzania. And then if it breaks rather quickly and can't be repaired, it is a big loss.
A side effect would of course also be that manufacturing in Tanzania would give work to Tanzanians. And maybe some pride in adding the logo "made in Tanzania"!
Well, this might be a stupid idea. But we want to make a try and see where we land
So the construction has to be simple, stable and cheap in components. Labor is cheap so there is no problem wiring transformers or similar. The solar cells, LEDs and most components would of course be imported.
After some searching I found this page
http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchCircuits-P1.html
In test setups I have tried out "CIRCUIT A" and "LED TORCH CIRCUIT - 1". I used a small transformer from a switched power supply and the circuits did work. But the LED was not very bright and I measured 1 mA (as the current is pulsing I put a 10 ohm resistor parallel with an electrolytic so I go DC and then calculated the current from that). I guess the transformer was not a good match for this, I have to wire my own.
I did some electronics as a hobby many many years ago, but I'm quite rusty. I have always been afraid of dealing with transformers and coils, found it very hard. I would have preferred a solution with a voltage pump using diodes, transistors and electrolytes. But I think the voltage/power loss in the components would be too high when using a 1.2V battery. So using a transformer as in these circuits I think is likely the right thing to do.
I would like to use "LED TORCH CIRCUIT - 1" but with three LEDs in series. I connect a 2-3 volt solar cell with a diode in series directly to the 1.2 V rechargeable AAA battery. The solar cell is weak and with its own inner resistance and the battery pulling the voltage down to 1.2V I think that simple charging would work. Without risk of damaging the NiMH battery or risk of fire or something.
I would be very thankful for some advice and hints?
I'm totally lost when it comes to the transformer. I have a hard time find cores using the F29 material. Do you know some cheap "standard" ferrite core I can use? F29 seems to be very high frequency, do I really need F29 for this application? I don't know the ferrite size I need to handle three LEDs but I guess I can buy some in different sizes and try it out.
In "LED TORCH CIRCUIT - 1" there is a diode in series with the LEDs, but not in "CIRCUIT A", why is that?
Why is a faster diode needed when using three LEDs? I can't find that diode, do you have suggestions for some other fast diode with low voltage drop that will suite the use?
I have trouble to follow why BC338 is good enough in LED TORCH CIRCUIT 1 and 3, but not in 2?
I also have problem to find 2SC3279, any suggestion for a cheap replacement?
Is it true that pulsating current like in "CIRCUIT A" give more light with less power? If that is true, I was thinking if the 100uF electrolytic in "LED TORCH CIRCUIT - 1" could be moved to be over the 39 ohm resistor only, to still give a stable signal to the base of the transistor but let the current pulsate through the LEDs. Guess I just have to try that out. If works, can I remove the the diode that is in series with the LEDs?
The component that breaks first is often the switch. But if making the flashlight reparable and the switch is cheap, maybe not a big deal. Just wondering if you knew of a cheap and small on/off switch that you think would be good for this application in the demanding climate? A non mechanical switch would be perfect, but not sure I can do that with very few cheap components.
I attach some photos. Yes I know, one of the ugliest prototypes ever! But I haven't touched my box with electronic tools and components in many years and I have become old, lazy and got bad eye sight. So I'm kind of satisfied with my prototype anyway
I know using a small microcontroller, like a PIC, to control the charging etc is likely the right thing to do. And if in-circuit-reprogrammable it would mean that not just physical components that breaks could be replaced, but that other defects in function could be corrected by reprogramming. It would really make the "service" part of the Tanzanian solar product stand out compared to a highly integrated Chinese one.
In my next design (if any) for Solar flashlight and phone charger I would use a low cost microcontroller and I will invest the time to learn how to use a microcontroller. But right now I'm after the most simple circuit that is "good enough" for a simple flashlight.
Also not that component cost is critical. Using three AAA or AA batteries would make the resulting flashlight way too expensive.
I'm grateful for any suggestions or hints!
kent
A bit fun for me that my old tools and components came to use again for this prototype.
I had some old veroboard and my Dremel also came to use to cut it. I know the placement is not optimal, lets pretend it was because I was stressed as I had a deadline in a few day when my little brother was to travel down to Tanzania and bring the prototype with him
Was a bit hard to get the board into the box, but veroboard is flexible. The switch was so old that it just broke into pieces when I did hold it a bit too hard with the pliers. Luckily I had one more.
Two $1 solar garden lights was taken apart to get solar cells. The cells were clued to the plastic so hard that I had to use my Dremel to cut them out together with part of the garden light plastic.
I also got use for my old conical hand reamer, it was used to widen the hole in the box.
We have discussed a bit if it is possible to manufacture some simple solar products in Tanzania. A flashlight would be a small first try. Not that manufacturing in Tanzania would be cheaper than buying from China, there are lots of solar products from China sold in Tanzania already. But they often break rather quickly and are highly integrated so hard to repair. And even if the small simple solar products are cheap from our perspective, they are not for someone in a village in Tanzania. And then if it breaks rather quickly and can't be repaired, it is a big loss.
A side effect would of course also be that manufacturing in Tanzania would give work to Tanzanians. And maybe some pride in adding the logo "made in Tanzania"!
Well, this might be a stupid idea. But we want to make a try and see where we land
So the construction has to be simple, stable and cheap in components. Labor is cheap so there is no problem wiring transformers or similar. The solar cells, LEDs and most components would of course be imported.
After some searching I found this page
http://www.talkingelectronics.com/projects/LEDTorchCircuits/LEDTorchCircuits-P1.html
In test setups I have tried out "CIRCUIT A" and "LED TORCH CIRCUIT - 1". I used a small transformer from a switched power supply and the circuits did work. But the LED was not very bright and I measured 1 mA (as the current is pulsing I put a 10 ohm resistor parallel with an electrolytic so I go DC and then calculated the current from that). I guess the transformer was not a good match for this, I have to wire my own.
I did some electronics as a hobby many many years ago, but I'm quite rusty. I have always been afraid of dealing with transformers and coils, found it very hard. I would have preferred a solution with a voltage pump using diodes, transistors and electrolytes. But I think the voltage/power loss in the components would be too high when using a 1.2V battery. So using a transformer as in these circuits I think is likely the right thing to do.
I would like to use "LED TORCH CIRCUIT - 1" but with three LEDs in series. I connect a 2-3 volt solar cell with a diode in series directly to the 1.2 V rechargeable AAA battery. The solar cell is weak and with its own inner resistance and the battery pulling the voltage down to 1.2V I think that simple charging would work. Without risk of damaging the NiMH battery or risk of fire or something.
I would be very thankful for some advice and hints?
I'm totally lost when it comes to the transformer. I have a hard time find cores using the F29 material. Do you know some cheap "standard" ferrite core I can use? F29 seems to be very high frequency, do I really need F29 for this application? I don't know the ferrite size I need to handle three LEDs but I guess I can buy some in different sizes and try it out.
In "LED TORCH CIRCUIT - 1" there is a diode in series with the LEDs, but not in "CIRCUIT A", why is that?
Why is a faster diode needed when using three LEDs? I can't find that diode, do you have suggestions for some other fast diode with low voltage drop that will suite the use?
I have trouble to follow why BC338 is good enough in LED TORCH CIRCUIT 1 and 3, but not in 2?
I also have problem to find 2SC3279, any suggestion for a cheap replacement?
Is it true that pulsating current like in "CIRCUIT A" give more light with less power? If that is true, I was thinking if the 100uF electrolytic in "LED TORCH CIRCUIT - 1" could be moved to be over the 39 ohm resistor only, to still give a stable signal to the base of the transistor but let the current pulsate through the LEDs. Guess I just have to try that out. If works, can I remove the the diode that is in series with the LEDs?
The component that breaks first is often the switch. But if making the flashlight reparable and the switch is cheap, maybe not a big deal. Just wondering if you knew of a cheap and small on/off switch that you think would be good for this application in the demanding climate? A non mechanical switch would be perfect, but not sure I can do that with very few cheap components.
I attach some photos. Yes I know, one of the ugliest prototypes ever! But I haven't touched my box with electronic tools and components in many years and I have become old, lazy and got bad eye sight. So I'm kind of satisfied with my prototype anyway
I know using a small microcontroller, like a PIC, to control the charging etc is likely the right thing to do. And if in-circuit-reprogrammable it would mean that not just physical components that breaks could be replaced, but that other defects in function could be corrected by reprogramming. It would really make the "service" part of the Tanzanian solar product stand out compared to a highly integrated Chinese one.
In my next design (if any) for Solar flashlight and phone charger I would use a low cost microcontroller and I will invest the time to learn how to use a microcontroller. But right now I'm after the most simple circuit that is "good enough" for a simple flashlight.
Also not that component cost is critical. Using three AAA or AA batteries would make the resulting flashlight way too expensive.
I'm grateful for any suggestions or hints!
kent
A bit fun for me that my old tools and components came to use again for this prototype.
I had some old veroboard and my Dremel also came to use to cut it. I know the placement is not optimal, lets pretend it was because I was stressed as I had a deadline in a few day when my little brother was to travel down to Tanzania and bring the prototype with him
Was a bit hard to get the board into the box, but veroboard is flexible. The switch was so old that it just broke into pieces when I did hold it a bit too hard with the pliers. Luckily I had one more.
Two $1 solar garden lights was taken apart to get solar cells. The cells were clued to the plastic so hard that I had to use my Dremel to cut them out together with part of the garden light plastic.
I also got use for my old conical hand reamer, it was used to widen the hole in the box.
