Purpose of this project is to create batreryless light for my balcony .I plan to harvest energy from solar cell to supercap using MPPT IC. Then this energy should go to boost-converter and ultimately constant-current power source for couple of LEDs. Project should contain no MCU and will also have some 3D prints included.
Okay, i assembled first version. Fabricating and soldering went well despite some doubts that i had due to elements and track sizes
Few issues were found quite to my expectations. Mostly due to mistakes in my design.
1. Enable pin didn't work on mcp1640, it stays on forever, for now not sure why. Possibly i did get ic version without this feature (if that's even possible).
2. More importantly regulation didn't work as expected. I didn't put current regulating resistor in series with output LED, therefore i effectively tied regulator output to forward bias voltage of LEDs making its work impossible. Due to this current went well above my calculation and current shunt sensor sent ~2v to feedback pin of voltage regulator. This is well above required 1.2v and therefore regulator saturated driving whole chain to balance point way above my expectation.
Here is updated schematics
First of all, i want regulator to drive it's output to ~5 volts as it is designed to. Therefore to keep power wasted on resistor down i put LEDs in series, with this configuration i expect voltage on the ouput to be ~5.5v, with total voltage drop of 5v this shoul leave 10 Ohm resistor .5v, therefore driving whole chain towards 50mA current. This will give me 50mv voltage across shunt resistor, and with gain factor 50 i'll expect 2.5 volts on feedback pin of INA199. Therefore i send it back to regulator through resistor voltage devider that i'll set on 1/2 for now and adjust later.
Here is updated pcb layout and i'm currently working of fabrication of second prototype
Next step is to prepare first sample and test schematics.
I prepared first version of PCB, size is 34x42 mm.
Smallest part is ina199 in SC70 package (2.00 mm × 1.25 mm, 6 legs), slightly worrying me to be honest.
Then i prepared home-made smaple using photoresist film technique. I'm not superhappy with the result but really want to try assemble it before order more from PCB supplier. Normally it takes few iterations before all issues are found. This still lacks some post etching work and hopefully it will get slightly better before assemble.
Next i did some calculation and changed my mind regarding schematics. Seems that with this tiny cell that i plan to use, light would be either not very bright or it will work for an hour at best. Therefore i did some research online and prepared schematics with MPPT IC and constant current driver.
(NB: this schematics didn't work quite as expected, pls see updated one later in the project log)
First part is SPV1040T driven MPPT charging driver. Scematics is directly from datasheet.
Second part is MCP1640 boost converter driven by INA199 current shunt monitor. Simple idea here is that INA199 monitors current through shunt resistor and as voltage from cap drops it sends higher feedback voltage to boost converter to boost it up even more. And this goes until cap is almost dried out (MCP1640 will work down to 0.35 volts on its input)
Last part is photoresistor driven enable pin on boost converter. This should start discharge only when it gets dark.
NB: This is not yet tested and few adjustment will be made in the process.
When this project started i planned to use most simple schematics with direct charge from solar cell to supercap and discharge through simple photoresistor driven mosfet switch to LED. Therefore it all started from the case. And this model was created. It should have place for small 30x55 mm cell and place for photoresistor sensor. To it's sides there is a place for 2 60F 2.7V caps that should get connected sequentially to boost max voltage at least to 5.5v.
I've prepared my model in FreeCAD and after few adjustments printed that with semitransparent filament.