| Electronics > Projects, Designs, and Technical Stuff |
| Driving fans from organic solar panels |
| (1/2) > >> |
| Greg J:
Hi guys, I've never posted questions about my own projects so here we are. I got a bunch of organic solar panel strips from InfinityPV - great stuff. They don't seem to be too efficient (3-4% tops) - but I thought it will do. The project I'm working on, is to drive two small fans (or more, but I'm sticking to two for now) - Pi-FAN 5V, 0.12A each using the panels to generate the electricity. (EDL3007S05) The surface these are on is of an unusual shape - think about a elliptical shape pipe, where the panels are on the outside. Rather large area. Both sides are exposed to the sun. So I thought that having couple of these panels in series, producing about 15V without load, drive a DC-DC converter, which would output 5V - and this could drive a fan or two. The fans are just there to keep the surface under the panels cooler. I struggle with this. The fans do sometimes come on, when voltage reaches about 3V from the DC-DC converter. Sometimes I have to nudge them, and then they stay on. So obviously, the panels don't seem to be generating enough current to get it over the initial hump. I keep on stringing more and more panels in series, but I was wondering if there's anything else I have missed? The circuit diagram would be rather simple. Panels in series -> DC-DC converter -> fans in parallel. Parallel with fans I got a 3300uF 6.3V electric cap - to smooth out the demand when the fans are starting. The questions I got: - current requirement. Could panels be connected in parallel instead of in series? - should I add anything else to only switch the fans once the circuitry reaches certain voltage? What sort of circuitry would it be? - am I mad trying to drive small motors using these panels? I purposely don't want to use batteries that charge, because that complicates the matters. But I'm open for full level of schooling here. |
| CiscERsang:
--- Quote from: Greg J on June 01, 2019, 01:46:27 pm --- I keep on stringing more and more panels in series, but I was wondering if there's anything else I have missed? --- End quote --- hi there, what about sunshine in Foggy Albion? :-// |
| Greg J:
It's actually quite sunny in Kent :-0 |
| mariush:
Try moving the capacitor in FRONT of the regulator, and use a smaller capacitor after it. Of course, can't move the 3300uF 6.3v in front, as you'll kill it with the 15v coming from panels... but consider something like 470uF 25v polymer or low esr capacitor in front and maybe 100uF 10..25v or something like that on the output, you won't need more. With a big capacitor at the output of the regulator, you're forcing the regulator to pull a burst of current in order to fill the capacitor with energy and that's in addition to the initial demand of the motors of the fans. you have 5v at 0.12v so two fans will be 5v at 0.24a ... that's 5 x 0.24 = 1.2 watts ... assuming the panels give you 15v then the buck regulator will probably pull around 1.5 watts from the panels, or 1.5w / 15v = 0.1 A (buck regulator will only be around 90% efficient, so account for that) If your buck regulator has an enable pin, you could see if maybe you could use a voltage divider (2 resistors) to send the enable signal only when the input capacitor's voltage goes above some threshold. For example, if the regulator turns on only when you have 1.2v or more on the enable pin and turns off at less than 0.3v or 0.6v , then you could choose two resistors to get a 10:1 ratio. This way, when the input capacitor is charged enough to have more than 12v, the regulator turns on and produces 5v. You could also have a time delay circuit added to each fan... for example a combination of resistor and capacitor plus a mosfet Pick resistor and capacitor values in such a way that voltage ramps up from 0 to something like minimum 2v in a second or two. Connect this to the gate of a mosfet which you basically use as a switch. When the voltage on gate is high enough, the mosfet turns on and therefore powers the fan. For the second fan, pick the resistor and capacitor to ramp up voltage much slower, let's say giving around 2-5 seconds more. This way first fan turns on and starts spinning for a couple of seconds and the initial current burst is gone, and now second fan can start and pull some current. Here's tutorial and explanations on how to use RC and how to calculate time : https://www.electronics-tutorials.ws/rc/rc_1.html Here's part of the table on that page: Time Constant RC Value Voltage Current 0.5 time constant 0.5T = 0.5RC 39.3% 60.7% 0.7 time constant 0.7T = 0.7RC 50.3% 49.7% 1.0 time constant 1T = 1RC 63.2% 36.8% 2.0 time constants 2T = 2RC 86.5% 13.5% Let's say you have a 470kOhm resistor and a 10uF capacitor and the regulator outputs 5v. The time constant is T = R x C = 470k x 10 uF = 4.7s ( because you convert uF in Farads) At 0.5 time constant (~2.4 seconds), the voltage on the output of the 10uF capacitor will be around 39.3 / 100 x 5v = 1.96v At 0.7 time constant (~3.3 seconds), the voltage on the output of the 10uF capacitor will be around 50.3 / 100 x 5v = 2.5v For the second fan, you can pick a higher resistor value and a bigger capacitor, to get something like 6-10 seconds for the time constant. Or, you connect the first fan directly to the output of the regulator and have it start right away, and use this RC + mosfet scheme to delay the second fan startup a few seconds. Now, you need to pick a mosfet that has minimum turn on threshold voltage somewhere above 1.8v but which can also handle up to 5v on the gate (which shouldn't be a problem) You have to look at the Vgs(th) aka Voltage gate source threshold in the datasheets ... that tells you the minimum voltage that must be between the gate and source pins of the mosfet for the mosfet to turn on. For example BS270 has a typical 2.1v gate threshold voltage (max 2.5v) but also a minimum of 1v , so you may have to get a bunch of mosfets and test them and hand pick the ones which only turn on at higher voltages: https://www.onsemi.com/pub/Collateral/BS270-D.PDF (digikey link: https://www.digikey.com/product-detail/en/on-semiconductor/BS270/BS270FS-ND/974203) A safer choice would be for example FQN1N50C which has a minimum threshold voltage of 2v : https://www.onsemi.com/pub/Collateral/FQN1N50C-D.pdf (digikey link: https://www.digikey.com/product-detail/en/on-semiconductor/FQN1N50CTA/FQN1N50CTACT-ND/1923111 ) However, this one also has a higher Rds(on) resistance of around 6 ohm ... it's like having a resistor in series with the fan... and this can cause fan to slow down a bit. BS270 would be a bit better because it has only 2 ohm internal resistance so it will heat less. now you can simply have +5v ---[fan]---- [drain pin ]--- [mosfet]----[source pin] ---- ground ... and the gate of the mosfet connected to that RC circuit 5v --- [ resistor in series] ---- [capacitor between r and ground] ---- [gate ] Here's a mosfet tutorial / explanation : |
| Greg J:
The DC-DC converter I'm using does have an enable port. This is the one I'm using: https://www.droking.com/5-pcs-lot-mini-power-supply-module-car-adapter-dc-4.5-24v-to-5v-3a-buck-converter-adjustable-voltage-regulator-driver-module-charger Not quite sure how to use the enable pin on this buck converter, the description says something like "Just direction connect VO+, GND, IN+ and if there is no high or low electric level control, it will be no need to connect EN. There is an internal Enable control on the IN+, if you want to connect it separately, please connect it on the EN". This doesn't quite make sense in english. But I'll experiment with the enable pin first. I think its the fans that just require a burst of energy to start turning initially - is what's killing it. Secondly, I don't think the panels produce enough current - so the output voltage drops to below 5V on the input with even single fan. I need to email InfinityPV people and ask few questions. Thanks Mariush, that's amazing. I've used some of these techniques before, but it's been literally over 10 years. |
| Navigation |
| Message Index |
| Next page |