![]() The logic works like: (1) solar panel voltage under battery voltage => OFF (2) battery over 14V => OFF (3) battery under 13.5V and NOT (1) => Pulse ON It is "Pulse ON" because we don't want to waste current feeding the relay all the time. Anyway, while I was wrapping up this post I think I've got the magical solution I wanted based on a LATCHING RELAY! "OFF" position will take out the battery completely, "ON" will connect the panel directly. However when the panel at some 15-19V it seems that not much could be gained by playing with voltage changes. Frankly I don't like the sound of that! Another thing: can somebody share some experience with MPPT and such low power setups as mine, is it really worth it (in terms of efficiency, cost up to a point is secondary)? Now if the voltage generated by the solar panel is about the same (or even lower) than battery voltage I can understand a DC-DC converter would push much more current (as opposed to very little or none) and a 5-20% loss in the converter would be irrelevant. I don't particularly like to drop the input voltage just by shorting the panel to the ground but in the last link also it's mentioned that the FET acting as a diode might for a short time actually short out the battery over the other FET that's shorting out the panel! And the protection for this transient event is an inductance in between. In fact current setup has a switch that shorts out the diode in case the panel is connected only during the day (and anyway it depends on the panel and the distribution of enough light/not enough light, it might be better to just forget about the diode and live with the small overnight discharge). Thanks for the answer, yes the diode is a big part of the fight towards "no loss". Any elegant and obvious solution I'm missing? I started thinking about a Low-dropout regulator but I'm having trouble finding one that meets the specs (probably I can do the same just with FET+opamp). It is ok to waste it once the battery is full. I'm not talking about fancy MPPT, just something that kills the charge while wasting almost no energy while charging. Anyway as I said I want this to be basically as efficient as the "manual" option. Now the challenge is to do this about as efficiently as just looking at the voltmeter and pulling the plug by hand (this is in fact easier than one can imagine as the energy balance is usually negative when charging via solar panel, we eat with other loads more than the sun gives us). And preferably charging should resume when the voltage drops under some level (it can be the same threshold if the controlling element can handle it). The controller just needs to stop the charge when the voltage on the battery reaches some level (like 14V or something similar, we don't care about more complicate algorithms or temperature, etc). This charges a deep cycle Pb battery, we're talking about 3+A. Ah well, good old prototypes.I try to come up with a solution for a friend - basically there's a 50W solar panel that can give about 3A at about 15-19V. ![]() The circuit used to work fine, so I didn't suspect that I had damaged it. I really thought it had something to do witih the pulse steering as this is the first time I have implemented it. Sorry it turned out to be my poorly constructed hardware. So I checked for a short, and sure enough, I found a very small connection between the two at the pinout for the programming cable. To find the problem, I set all of the outputs high and then tested the voltages and found that only RA0 and RA4 had a 0.6 V output while all others had 5 V. 10 minutes of work found the problem today. A good night's sleep seems to solve a lot. ![]() I found that probelm first thing this morning. After working with the circuit for a while, I managed to cause a small piece of frayed wire to short RA0 and RA4 to each other. I pinned out the programming pins to a ribbon cable so I wouldn't have to take the chip out of the circuit every time I programmed it. You're right, however the time that the pin was set was long enough to get the right reading. Pommie, that is what I thought you suspected.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |