Measuring Solar Charging Current
Bean
Posts: 8,129
Can I use a normal current sensor chip to measure charging current ? I ask because the current flow is backwards from normal. If you use a shunt resistor you get a negative voltage with reference to ground.
Also can anyone recommend a current sense IC that will measure upto 1 amp ?
This will be used to measure the charge from solar panels into batteries (NiMH·or Lead Acid).
I want to use 2 panels, then every couple minutes measure the charging current with the panels in series, and parallal (using latching relays). Then I will keep them at whichever configuration is providing the most charge current.
Is there any merit to doing this ?
It seems to me that when the sun is bright the panels in parallel would provide more current, then when the sun is not as bright the panels in series would provide more current.
Thanks for any assistance,
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
Also can anyone recommend a current sense IC that will measure upto 1 amp ?
This will be used to measure the charge from solar panels into batteries (NiMH·or Lead Acid).
I want to use 2 panels, then every couple minutes measure the charging current with the panels in series, and parallal (using latching relays). Then I will keep them at whichever configuration is providing the most charge current.
Is there any merit to doing this ?
It seems to me that when the sun is bright the panels in parallel would provide more current, then when the sun is not as bright the panels in series would provide more current.
Thanks for any assistance,
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
Comments
How about using a differential or dual channel ADC measuring the voltage at each end of the shunt resistor, then computing the difference in software?
Mike
Is it a relative value you're after or exact numbers?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·
Steve
"Inside each and every one of us is our one, true authentic swing. Something we was born with. Something that's ours and ours alone. Something that can't be learned... something that's got to be remembered."
The MAX472 will measure current in both directions. One pin produces a current proporional to the absolute current, and another pin is 0 or 1 to indicate the sign. You can scale the gain with resistors to measure just about any current range, with the appropriate shunt.
emesys.com/BS2rct.htm#B_current
I don't have the part numbers handy, but there are several more chips of that type available from Maxim, Zetex and others.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
That's a question that will make you think! Around a year ago I installed a set of 6 panels,in parallel. I wondered the same thing you are as to whether more power might be produced using a series connection at times.But then it seemed there was no valid reason to do this as the power produced,in watts, would remain the same regardless of the voltage arrangement. And after monitoring the panels for a while,it was obvious that there is precious little current to manipulate while maintaining any sort of efficency.Just keeping a small[noparse][[/noparse]10 mA?] drain on the system really hurts.
You might like to see the mounting platform I built that keeps the panels facing the sun.It's controlled by a BS2 Homework board.A 12V linear actuator swings it around to several preset positions throughout the day.Using the sleep mode helps very much to keep the thing from wasting too much current.
Best Regards
Jim Richey
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thanks, Parallax!
The Parallax temperature sensor kit uses a DS2760 chip that is intended for battery charge/discharge monitoring. It has an internal sense resistor (external can also be used) as well as a bi-polar A/D converter and Amp-hour (milli-amp-hours really) accumulators to keep track of the charge in or out. Sound ideal for your application.
To get the very most out of your solar panels you should match their impedance (which varies greatly with incident sun), and dynamically adjust your load to match. This way the most power will be transferred. So to measure their instantaneous impedance you would have a micro that every few seconds switches the panels to a known load for a few milliseconds, and you measure the panel's output voltage. From that you can determine the panel's impedance, and hence the optimum current.
Then comes the tricky part, matching you battery system to that impedance. This would be done by feeding the panel's output to a micro-controlled fly-back style regulator, and make the micro adjust the regulator to run in the "continuous" mode at the optimum current previously determined. I suspect it is important that there NOT be a large (input) capacitor between the panels and the switcher as that would defeat the whole idea of impedance matching.
I have frequently mused about this arrangement, and have some expectation of commercializing this sometime, but I have never built one. Just all my work with fly-back switchers leads me to believe this will work very well. Without doing any math based on emperical measurements, I have the anticipation that such a scheme will significantly (30% ??) improve typical fixed load systems. As you well know, an improperly matched system wastes a lot of power, and a very poorly matched system (say low levels of sun shine) will produce almost no useful energy to the storage system.
Please remember that the switcher must be of fly-back type so the inductor can properly be charged at the optimum current during the ON phase, and then dump its energy to whatever battery voltage you like during the "DUMP" phase.
Should be a hoot to build this!
Cheers,
Peter (pjv)
I may even use 4 solar panels and cycle through "series, series-parallel, parallel" configurations to get the most current.
This basically start with me needing to charge the batteries even on overcast skys. So I figured I'd use a series connection. But then I though about all the power I'd be missing out on if it was sunny.
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
What if you just put the panels all in series or all in parallel and used them to power a DC-DC converter? These are typically 90+% efficient and could provide a steady charging supply over a wide input range.
-Phil
Peter, I don't think I agree with your analysis of solar panel impedance. The panel is basically a current source, which has a very high impedance up to the knee. The graph of current vs voltage is more or less flat horizonal line, with the same current flowing into a 6 or 12 volt load as into a short circuit. (That is, assuming a panel that has an OC voltage of around 21 volts.) At some higher load voltage, the current goes over a knee and drops off precipitously and is zero at the OC voltage. With less light, the current is proportionately less, and the knee occurs at a lower voltage, but the IV curve has the same basic shape, a horizontal line with a knee. The maximum power point is a curve tracks points just before the knee, so it curves gently upward and to the right with increasing sunlight.
The issue with impedance is that the IV curve is not that of a resistor, so that using a fixed resistor to measure the output impedance as you would for a speaker won't yield useful information. I do think that a switcher that draws just enough power to keep the panel at the maximum power point below the knee is a good solution so long as the switcher can run at high efficiency. A feedback algorithm could dither the load point, and multiply I*V to find the maximum power point, and always center the operation around that point through feedback. I don't see any problem with keeping a large capacitor on the input. That capacitor would integrate the solar panel charge and help to regulate the operating point at the optimum voltage+dither. The switcher would draw down that capacitor as fast as the solar panel fills it up and deliver the energy at the optimum rate for the stuff on the other side.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Clearly we have differing views on some of this, particularly the impedance and the capacitor issue. If my position on the varying impedance is incorrect as you state, then the capacitor issue goes away.
On the other hand, I now think you are right about the capacitor, but by reasoning differing from what you suggested; and that is that the fly-back switcher runs in CONTINUOUS mode, meaning that not all the energy is removed from the inductor in each cycle, and hence we can nicely control the average current it draws, thus matching the optimum point of the panel. This implies that voltage at the panel's output will be (relatively) constant for a specific load at a specific illumination. The capacitor then in fact aids the power collection process by garnering the panel's available power while the switcher is in "DUMP" mode.
I still think I'm correct on the impedance issue, but at high insolation levels is not as severe as I anticipated at low levels, where most of my requirements (winter time in the cold) become the major issue. I do know that commercial panel matching chargers exist..... as to how they work, my theory is as I described.
So I'm going to need to build one of these to prove my point as I think it would be difficult for me to convince you otherwise should I be right, and convince myself otherwise should I be wrong.
More later.
Cheers,
Peter (pjv)
microship.com/microship/techinfo/microshipnet/ppt.html
It does not say much about the optimization algorithm used with the PWM controller but it may be in there to dig for.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
If you were to make a microcontroller controlled DC-DC converter, what parameters would you want to control to get the most output from a solar panel input ? Frequency ? Duty Cyle ?
I've seen some of the BEAM circuits that monitor the charge rate (of the solar panel to storage cap) and trigger when the rate slows to a certain level. Would that be something you could monitor with a micro ? Would it be of any use ?
It seems to me that the solar panel is most effecient when connected to a consistant impedance. For example if the I*V is at the maximum with I=0.1Amp and V=10V then a resistance of 100 ohms would give the most power. But how do you determine that with changing sun conditions. And how do you make the input to the DC-DC converter look like that optimum resistance ? I would guess by varying the duty cycle ?
It's nice to see some lively discussion about this topic. Thanks everyone.
Bean
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
If you were to study the issues around a fly-back DC-DC converter, in particular those revolving around the continuous vs interrupted mode of operation, you would see the benefit of what I'm preaching.
In a fly-back, you can make its input impedance one thing, and its output impedance something else. Thus you can nicely match both your source and your load. You would do this input matching by adjusting the charge-up portion of the duty cycle to get the optimal current (I contend this is not at the maximum current)·that the panel produces with the particular level of sunlight. The output impedance adjusts itself automatically to dump the magnetic charge into the batteries.
My take on the solar panels (yet to be proven) is that their impedance varies significantly with intensity of solar radiation, and hence a specific constant load is only good if it is the right load. If my theory is incorrect as Tracy suggests, then of course my conclusions are wrong. If the impedance varies only a little bit, then there is not much to be gained by my scheme.
I'm going to try to find some time to prove (or disprove) this. Just now, I can't find my solar panels! Oh well.......
Cheers,
Peter (pjv)
www.solar-electric.com/charge_controls/mppt.htm
The attached figure is a graph of a solar panel IV curve, (blue line), and the power curve (green line) taken from the above url. This happens to be a high power (~6000 watt, 100 amp, 60 volt) array, but the shape of the curves is cannonical.
Current is constant, the horizontal blue line, until it goes around the knee, and then drops down toward the zero at the open circuit voltage point. Power is of course zero at both extremes, zero current and zero voltage, and has sharply defined peak power point above the knee. The two main effects are sunlight and temperature. The height of the blue constant current line is in direct proportion to the insolation, while the open circuit voltage shifts slightly to the left when there is less sunlight. Maximum power curve tracks insolation, too, and shifts toward lower voltages in lower light. Temperature of the panel has a big effect, and the peak power increases in cold temperatures and shifts toward higher voltages. Panels do not work as efficiently when they are hot.
Just for an example, suppose the peak power point is 1 amp at 17 volts, 17 watts. If the panel is hooked directly to a battery that happens to have a terminal voltage of 13 volts, that becomes the voltage operating point and the panel might deliver 1.1 amps at that voltage, 14.3 watts. With MPPT, the charger would hold the panel voltage at 17 volts and draw the one amp, and the DC-DC converter convert that to 13 volts at 1.3 amps (17 watts), less the efficiecy losses of the converter. More efficient use of the panel energy.
A microprocessor is in control of the MPPT process, and in measiring input I, V and P, and it can frequently do small experiments (dither) around its set point and seek out and track the maximum power point. With feedback, it can track regardless of the insolation or the panel temperature. I think these systems usually use a step-down voltage conversion, but one could imagine it as boosting the the output of a low voltage panel.
Peter, I defer to you on what converter topology would be most appropriate. The flyback sounds great if it can vary the effective impedance at its input. I agree that there will always be a resistance value that corresponds to the maximum power point. In my example, 17 volts/1 amps = 17 ohms. With half the sunlight, it might be 16.5 volts/.5amps = 33 ohms. My only gripe may be just a question of semantics, and has to do with the notion of measuring the output resistance of the panel. The panel does not have a Thevenin or Norton equivalent circuit overall. The equivelant resistances are small signal paramters, and the slope of the IV curve changes along the curvet. In the horizontal part the dV/dI is large and is best modeled as a Norton equiivallent current source with a large value of resistance in parallel, and the Thevenin equivalent circuit is a high voltage source with the same value of high resistance in _series_ with the load. But the model changes around the knee. The small signal resistance may relate to ohmic power dissipation in the panel itself, but is a minor factor in choosing the load resistor, e.g., the 17 ohms in the previous paragraph. I think that can be done only through the IV = P tracking. Do we have any disagreement about that?
One reason for using a high voltage array like the one in the graph is avoidence of ohmic losses in the wires that connect the panels to the equipment. 100 amps through 0.01 ohm of wire is 1 volt of loss.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 1/10/2007 9:31:18 AM GMT
So, can we generalize and say that Max Power is produced at about 86% of the open circuit voltage and 82% of the Short Circuit Current?
These are both tests that can be done on the fly by the processor, say every 15 minutes. Then the converter/inverter can be adjusted to transfer the maximum power. Or, if the blue curves retain their shape in varying insolation, maybe a satisfactory estimate of MPP can be done with only the OCV.
Cheers
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk
http://www.siskconsult.com
·
What I'm interested in is charging batteries with really low light levels. So I might have a 10Watt solar panel that is only putting out 0.1Watt. But I want that 0.1Watt to be charging the batteries. Again without losing all the power when light levels are good.
Which are more effeient, boost or buck converters ? Or are they about the same ? Is it possible to have both ? Under low light levels the panel voltage will be far below the battery voltage, but at higher levels the panel voltage will be well above the panel voltage. Or is it better to design the panel so only one or the other (boost or buck) are needed (for example the panel voltage never exceeds the battery voltage even in full sunlight, and a boost converter is always used).
I want to thank everyone, I'm really learning alot about solar charging.
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
Tracy;
Yes, power tracking is exactly what I was espousing. My method of determining the panel impedance is what you are questioning, and you might have a point there, given the curves you offered. My issues have generally been with smaller panels and low insolation levels where I believe the effective series resistance of the panel will be much greater than the 100 amp example. Otherwise I think we are in agreement.
Tom;
No, I don't think you can make that statement. The panel's characteristics change with light levels, loads and temperature. I believe the only way to get the maximum out is by dynamic measurement and subsequent current adjustment to match the peak power point. For optimum results I would do that every few seconds.
Terry;
If you studied the differing topologies, you would discover that the fly-back has the characteristics of combined boost/buck. It can make the voltage higher or lower, it depends only on the load. In my estimation it is the optimum choice. All converters need to be properly designed.... you can end up with high as well as low efficiencies regardless of topology. You just need to do a good job.
Cheers,
Peter (pjv)
If an individual panel has been completely characterized, then a table lookup based on panel temperature and some measure like short circuit current could be used to set the maximum power point in an open loop scheme. However, a closed loop scheme using feedback to seek out the maximum power point would be a lot more robust and would not require the arduous task of panel characterization. Panels characteristics change with age, too, and all of that comes out in the wash in a closed loop scheme.
Peter, I am thinking your scheme with magnetics could be either buck or boost, if the inductor is tied to ground. While attached to the panel side, current would build up as you say, and then when the panel side switch is opened, current would dump through the other side through a diode. An inner feedback loop would keep the panel side voltage at the MPP by regulating the time that the inductor is connected. Is that the sort of toplology you are espousing?
A similar flying capacitor scheme could work as follows for relatively small currents, when the panel OC voltage is higher than the load voltage. This is similar also to the BEAM robot scheme. The panel charges up a large capacitor to the MPP. The A/D converter detects that and switches a small capacitor (low ESR) in parallel with the large capacitor. The voltage on the large capacitor drops say 0.1 volt as the small capacitor is charged. The small capacitor is disconnected from the panel side and is connected to a capacitor on the load side at a lower voltage and transfers the charge contained in Cflying*(Vpanel-Vload). An inner feedback loop modulates the switching frequency of the flying capacitor, to hold the Vpanel at the MPP setpoint. The system ADC monitors both Vpanel and Vload, and the transferred current is the averaged quantity, I = frequency * Cflying*(Vpanel - Vload). An outer feedback loop operating at a much lower rate dithers the set point through three values, V-delta, V and V+delta. If the setpoint is correct, then the center set point will yield higher power than both the + and - deltas. The operation is sitting right at the top of the power curve. If the values are monotonic increasing or decreasing, then the center set point is increased or decreased by delta in order to seek out the maximum.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 1/10/2007 7:39:03 PM GMT
Actually such a buck/boost scheme is most easily handled by transformer coupling and using a low side switch, rather than a plain inductor. During the charge phase, the primary winding builds up current supplied by the panel, while the secondary's series diode is reverse biased, so no secondary current flows. Then when the primary switch opens, the magnetic energy can only escape (find a load) by measns of the secondary, whose polarity now is such that its series diode conducts, and dumps to the load. The beauty of this discharge phase is that the voltage at this dump interval will rise to (almost) whatever required to cause the current to flow, and hence it will go to whatever that battery voltage is.
So, the secondary voltage can be higher than or lower than the input voltage, it all depends on the load.
The analysis is two independent phases: a charge phase with supply voltage, primary inductance, and on duration as the only effective parameters, and a dump phase with only the load impedance, secondary inductance and off duration as parameters. If the off duration is insufficiently long so as not to fully dump the magnetic energy, and a new charge phase occrs, then the converter is operating in continuous mode. However if sufficient dump time exists, then the secondary current will decrease to zero, and a discontinuous, or interrupted mode exists.
More energy can be transferred while in continuous mode as the currents can be made to rise due to effectively lowering of the primary inductance caused by the remaining magnetic energy.
All a good bunch of fun!
Cheers,
Peter (pjv)
I am trying to visualize the voltage across the low side switch on the primary side, and come up blank. I imagine that would depend on the load reflected from the secondary.
Thinking further about the flying capcitor scheme I proposed above, it would be highly inefficient and I don't think it would be a worthwhile approach. When a capacitor is charged from a voltage source, the resulting energy stored on the capacitor is always 1/2 of the work that went into charging the capacitor, and 1/2 is lost in the resistance, even in the limit as the resistance approaches zero. When the open circuit voltage of a panel drops from say, 21 volts to 14 volts, there isn't going to be much current left to harvest. I think the best bet is a low dropout, low quiessent current charger. Or go all the way and use a transformer scheme like you propose that can really harvest all of the available energy. When all the energy is not needed it could throttle down.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
About the transformer..... no, there IS (almost) no relection from the secondary while the primary is charging ! The secondary's series diode is reverse biased and hence isolates the secondary from the primary, and so the front end (low side) switch only sees the inductance of the primary. That's the beauty of this whole fly-back scheme...... two independent activities, charging the magnetics and dumping the magnetics.
Cheers,
Peter (pjv)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Agfa .... sorry for hoggin' this thread... not so much debating as trying to explain.... sorry!
Tracy......... When the primary's low side switch opens, the magnetic field collapses, and hence the secondary's polarity reverses and can now conduct through its series diode into the load. The only kick you see on the primary switch is due to the primary's relatively small leakage inductance, albeit that is still big enough that it must be handled to prevent damage to the transistor...... easy enough.
If there were no load on the secondary, then effectively the secondary does not exists (other than for some parasitic effects) and all the collapsing energy stays in the primary with a HUGE destructive pulse.
Cheers
Peter (pjv)
I agree. They are way over my head. But it is an interesting discussion, and it IS on topic.
Continue on... I'm enjoying it.
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer
Parallax, Inc.
I'm interested in this topic because I have lots of solar powered systems installed. The main issue is to meet the steady power requirements, year round, but also for various reasons of cost and security, to use the smallest possible panels. Sometimes the sun comes out for half an hour, and it would be great to take advantage of 100% of the available energy. Or, on long dark days with no direct sun at all, to harvest every last milliwatt.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
When panels are hooked up in parallel, the currents add, and each panel contributes the full current it is capable of at the common point voltage. When panels are hooked up in series (as are the many ~0.5 volt cells that make up a higher voltage panel), the voltages add, but the current is equal to the current produced by the weakest cell. That is why shadow on even a corner of a panel can dramatically affect the total current produced by the panel. One cell can be the weak link.
I'm nor sure about the scheme that Bean first proposed, which I believe was to switch a pair of panels from parallel to series connection when the light level is low. Is that a fair statement? It might work, but my guess is that the switching point would occur at a relatively low light level, maybe 1% of full sun, where not much current is available anyway. Whether or not that would be worthwhile would depend on the importance of that small current to the load. The exact amount of additional current that could be harvested by switching from parallel to serial is a quantitative question that could be derived from the solar panel curves.
Different solar panel technologies have different characteristics at low light levels, for example, think of a small solar calculator which is able to function in even average indoor light. Big solar panels produce practically nothing at indoor light levels. I was using some of the beautiful Copper-Indium-Sulfide (CIS) thin film panels--until Shell Solar decided to discontinue manufacture of the small ones due to the fact that business economics favor large panels. CIS panels were supposed to have a better low light response, but as a practical matter I didn't see much difference compared with silicon panels.
This thread transmogrified into a discussion of the maximum power point, which leads to a different and much more circuit-intensive way to get the most out of the panels. It is kind of like operating an engine at is peak torque point, which requires a transmission to move the energy to the wheels.
Sorry about the theory, but I like to qoute Williard Gibbs (thermodynamics guru), "There is nothing quite so practical as a good theory". I'd try to supplement that with some real data.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 1/14/2007 8:09:38 AM GMT
Yes my original concept was to switch from parallel to series under low light conditions.
But now that I understand more about the voltage/current output of solar cells, I can see the advantages of the MMPT system.
I have 10 panels, so now I need to do some experiments.
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com
Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com
Stuff I'm selling on ebay http://search.ebay.com/_W0QQsassZhittconsultingQQhtZ-1
"USA Today has come out with a new survey - apparently, three out of every four people make up 75% of the population." - David Letterman
Me too am interested in PV energy harvesting
specifically for the energy needings in a sailing boat.
Googling over the subject drived me to this site www.tunecharger.com/index-182612.htm
Differently from other sites, in this the author is willing to devulgate..........
I had a torough reading to it and it seems to me that some experiments could be bring forward
with the informations contained in the site, especially for what relate to the extremely simple yet very accurate
method used to find the MPPT point..............
I'm not a software expert, my "expertise" lean more on the electronic side of the matter..........
hope this is of help to who is interested
Saluti
Stefano
( Q = I2 * R * t )... So more amps means more heat and heat is loss... But you need a regulator... We know that regulator have interial resistances and so on... What you decide to choose is your own matter... About mesuring the current of the cell, you can simple use a 555 circuit (well you only mesure volts but I think that suits you, or hope so...)
Hope I helped,
Tricky
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
-Rule your Destiny-
--Be Good. Be Bad. Be Provas--
The hellinic (Greek) robots portal: Greekbotics
Many Projects and Schematics by the users·and also robotic news
(Translate using babelfish)