DIY solar USB charger - MPPT?

smasher
smasher Registered Users Posts: 5
i just picked up some 25W panels, for making a USB charger. this will be an upgrade from my current solar/USB chargers, which top out at about 4W each (actual peak power output). i'm hoping that with full sun, i can do more charging in less time, and with some cloud cover i can still get 1-2 things charging off of USB... a 5-6W panel at 50% capacity is useless for me, but i'm hoping that this setup can still be useful down to about 30% capacity.

panel specs:
Vmp: 17V
Imp: 1.47A
Voc: 21.4V
Isc: 1.59A

i'll be wiring each of those to a buck-converter with an adjustable output voltage (will be set to 5V, for USB) and up to 10A output (so each voltage converter should be no more than 50% capacity) - http://dx.com/p/10a-dc-cnoverter-buck-4-30v-to-1-2-30v-adjustable-voltage-regulator-power-supply-214277

just wondering... with a setup like this, is there any "real world" advantage of including any type of MPPT circuit, between the panel and the voltage converter? or is that something i shouldn't worry about, as long as i'm not trying to pull more than 4-5A from the voltage converter?

Comments

  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Re: DIY solar USB charger - MPPT?
    smasher wrote: »
    just wondering... with a setup like this, is there any "real world" advantage of including any type of MPPT circuit, between the panel and the voltage converter? or is that something i shouldn't worry about, as long as i'm not trying to pull more than 4-5A from the voltage converter?

    If your load is going to be close to your panel rating you pretty much have to. Without at least an impedance limiter, a straightforward switcher will "crash" the panel under conditions of high load. At the minimum you need a voltage limiter that prevents the panel from dropping below some voltage (should be close to its MPPT voltage.)

    Fortunately Linear Tech now has an IC that does all this for you: http://www.linear.com/product/LT3652

    Question - what USB loads take 10 amps? I have never seen one take more than about 1.5 amps.
  • smasher
    smasher Registered Users Posts: 5
    Re: DIY solar USB charger - MPPT?
    If your load is going to be close to your panel rating you pretty much have to. Without at least an impedance limiter, a straightforward switcher will "crash" the panel under conditions of high load. At the minimum you need a voltage limiter that prevents the panel from dropping below some voltage (should be close to its MPPT voltage.)

    Fortunately Linear Tech now has an IC that does all this for you: http://www.linear.com/product/LT3652

    Question - what USB loads take 10 amps? I have never seen one take more than about 1.5 amps.
    5x gopro cameras can pull >5A. when i outgrow that, i'll get more/bigger panels.

    if that IC could handle 5A output, i'd be tempted. i've never played with those types of battery-charger circuits... i guess if i fudge V-FB i could use it as a voltage-regulated power supply?
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: DIY solar USB charger - MPPT?

    Are you just charging the GoPro's, or are you operating them at the same time?

    Do they lock up when / if a shadow, person or otherwise passes in front of the panels due to low current? And if so, do they stay latched and refuse to charge unless you do a physical reset with a fully illuminated panel?

    I think you would be much better off with a *stable* solar setup consisting of a simple 18v panel, charge controller, and perhaps AGM battery and use that to supply your gopro charging needs with the typical mobile usb adapters and hanging those 5 cameras off the agm.

    If you want to test things out, why not just use a modern 5v mobile usb adapter hanging off your 25 watt panel and test it out that way - most modern ones handle 12-24v input so a quick trip to the local electronics outlet may get you up and running in hours. Just be sure to get a few that handle 1A or more of current.

    I've got 20w Powerfilms, 13w Goal Zeros, no names etc direct charging my toys with high power mobile adapters, but really prefer the convenience and stability of my main panel/cc/agm battery setup for charging.
  • smasher
    smasher Registered Users Posts: 5
    Re: DIY solar USB charger - MPPT?

    charging only, not running at the same time. i also charge phones, 18650s, bike-lights, etc, all via USB. for now, i want to avoid the complexity of a panel/battery combo, and just plug in some small 18650 "power banks" when nothing else needs to be charged.

    can you post some examples of "5v mobile usb adapters"? my brain is processing that to mean wall-warts and cigarette-lighter USB PSUs.
  • LandKurt
    LandKurt Solar Expert Posts: 41
    Re: DIY solar USB charger - MPPT?

    If you’re getting 4-5A at 5V then you’re already acting like an MPPT to pull up to 20-25W from a 25W panel.

    Will a buck converter be able to provide move than the Imp of 1.47A at five volts on one of those panels? If it acts like a PWM charger you’d be limited to 1.47A at 5V = 7.35W under ideal conditions.

    I imagine it would be useful to include a small 12V battery and a little PWM charger so you could at least get 14 – 20 Watts into the battery. Then with the battery providing plenty of amps you could get 3 – 4 A at 5V out of the converter steady state.

    Or do I fail to grasp how these sort of converters work?
  • smasher
    smasher Registered Users Posts: 5
    Re: DIY solar USB charger - MPPT?
    LandKurt wrote: »
    Or do I fail to grasp how these sort of converters work?
    it's been a while since i've played with hardware electronics ;)

    IIUC... the buck converter has a DC input (let's say 15V) and a DC output (let's say 5V). in between the input and output is an AC section, this one running at 150KHz. between the input and output, there's ≈90% efficiency, this one "up to 98%".

    if we pretend that the converter is running at 100% efficiency, and the input voltage is three times the output voltage, then the input current is 1/3 the output current.

    so a 25W panel (under "ideal conditions") would be able to supply 5V*5A through one of these buck converters (if /a/ the buck converter was 100% efficient and /b/ the input impedance of the buck converter adjusted itself for MPPT).

    if i wanted to do it the way i learned in high-school, i'd use a 7805 voltage regulator... then i'd get a regulated 5V output, but my current would be limited to the panel's Imp (or perhaps nearer to Isc). heat dissipated by the regulator would be = (Vin - Vout) * Imp

    or, in round numbers...

    (17V - 5V) * 1.5A = 18W

    that would provide (in round numbers) 5V * 1.5A = 7.5W.

    so doing it "the old fashioned way", yeah... my output current would be limited to the panel's Imp (or Isc) and the voltage regulator would waste more than twice as much power as it could provide.

    but like i said, it's been while since i've had much hands on with DIY hardware projects, so when i get it together then i'll find out if i'm right :p

    as always, wikipedia is a good place to start and a bad place to finish - https://en.wikipedia.org/wiki/Buck_converter
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: DIY solar USB charger - MPPT?
    smasher wrote: »
    can you post some examples of "5v mobile usb adapters"? my brain is processing that to mean wall-warts and cigarette-lighter USB PSUs.

    Sure. Most modern 5v usb mobile adapters don't use 3-terminal regulators, (not very efficient dropping an 18v panel all the way down to 5V) but use buckboost switchers. How about a large one by Anker capable of 24W, with 2.4A on each output:
    http://www.ianker.com/product/71AN2452C-WA

    Since like most others, it is capable of 12-24V input, you could easily put this directly across the typical small panel with a female cigarette lighter adapter. Or attach to a much larger battery to charge whenever it is convenient, and not have to rely on the sun. Here, Anker is just an example - there are some other quality manufacturers out there, but as always one has to be vigilant about not getting burnt on counterfeit products.

    The thing you want to use looks like it may work too.

    Also, how is your usb cabling? Does the go-pro limit itself to only a 500ma charge when in the presence of a data/sync cable? Do you have to rely on so-called "charge only" cables to allow for the full 1A input? (Ie, these are typically cables that have had the data+ / data- leads shorted facing the device, simulating an AC charger).
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY solar USB charger - MPPT?

    I just purchased the auto charger from Anker--Have not tried it yet (has Apple and Android configured ports). Looks like a nice mate to a "12 volt" solar panel.

    However, I have used a couple of their 5 port AC to USB chargers and a 12,000 mAH (12 AH) at 5 volt USB charger battery pack--Both have been working very well (last few months).

    Solar panel+12/24 volt usb charger+12 AH 5 volt USB battery...

    The warning about the USB cables with larger USB devices (and batteries). I have some (I thought) perfectly good USB cables from the local computer store, and they will hardly recharge the 12 AH battery pack (charge a standard cell phone just fine). I use one of the cables from Amazon for their Kindle Fire (color tablet)--And then everything charges quickly.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • smasher
    smasher Registered Users Posts: 5
    Re: DIY solar USB charger - MPPT?

    @PNjunction - i've wondered about those car-charger ones... specifically, i've wondered which ones might use more efficient (and more costly) switching voltage regulators, and which ones may use less efficient (and cheaper) linear regulators.

    in a typical automotive application, who cares if one of those runs hot? but if it's got a total power throughput of 24W, then it's almost certainly using a switched regulator.

    based on the specs of the buck converter i've got on order, it should give me 5V output when the input reaches about 6V. not sure how much current i can pull with 6V from a 17V panel (eg, cloudy day), but may well be enough to charge something. those car-charger USB ones... who knows what input voltage i'd need to get 5V out? rated 12-24V input, but may (or may not) work at 6-24V input...?

    worth a shot, though, to just tie a cigarette-lighter plug to the panel and see how one of those works. thanks for the idea!

    of course, i still wouldn't get any MPPT to match the input impedance to the panel :(

    yeah, i think the latest gopros limit to ≈500mA when connected to a computer, but can definitely pull >1A when connected to a "dumb" power supply.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY solar USB charger - MPPT?

    If you want to build your own MPPT based switch mode battery charger from Texas Inst and another from Linear Tech:

    www.digikey.com/product-detail/en/BQ24650RVAT/296-27699-1-ND/2352706
    www.digikey.com/product-detail/en/LT3652EDD%23PBF/LT3652EDD%23PBF-ND/2138757

    Some general information:

    http://www.digikey.com/us/en/techzone/energy-harvesting/resources/articles/designing-a-solar-power-battery-charger.html
    ...The most basic approach for maintaining peak power point voltage (VMP) sets panel voltage to a constant voltage level, based on the manufacturer-specified open-circuit voltage (VOC). The rationale for this approach is based on the fundamental electrical characteristics of a solar panel. A typical solar panel comprises a series of cells that, electrically, are simply forward-biased p-n junctions. Consequently, engineers can approach solar cells as common p-n diodes -- with similar electrical characteristics including VOC and temperature coefficient (typically about 2 mV/°C).

    For a solar panel, VMP can be approximated as a fixed voltage below VOC, with a temperature coefficient at VMP essentially the same as that at VOC (and considered linear across normal environmental temperature ranges). Designing an efficient solar battery charger then becomes a matter of augmenting the battery protection capabilities mentioned above with a simple temperature-compensated resistor network designed to set panel voltage at VMP.

    The Texas Instruments bq24650 and Linear Technology LT3652 each offer MPPT based on this constant-voltage approach.

    For the TI bq24650, engineers can set the solar panel voltage to the peak power point using a resistor network across the device's VCC and MPPSET pins. The device's input voltage regulation circuitry responds when the input voltage drops as the solar panel loses the ability to provide the total power of the system. When the device senses that voltage on the MPPSET pin drops below 1.2 V, the IC maintains the input voltage by reducing the charge current. For MPPSET pin voltage below 1.2 V, the bq24650 stays in its input voltage regulation loop while the output current is zero. Engineers can also disable charging completely by pulling MPPSET below 75 mV.

    Along with a full range of battery protection capabilities, the bq24650 supports diverse battery chemistries including Li-Ion/Polymer, Lithium Phosphate, and lead-acid batteries. Based on the TI bq24650, the TI bq24650EVM Evaluation Module offers a complete solar power battery charger with multiple test points and jumpers for experimenting with this device.

    For the Linear Technology LT3652, engineers can program the peak power voltage for a solar panel by setting the required values in a resistor divider across the VIN and VIN_REG pins. Designed specifically for solar power battery charging, the device features multiple battery protection modes and supports diverse battery chemistries including Li-ion/Polymer, Lithium Phosphate, and NiMH/NiCd. The Linear DC1568A is a demonstration kit that provides a complete solar power battery charger based on the Linear LT3652.

    With their use of resistor networks for setting the fixed voltage for MPPT, both the TI and Linear devices enable engineers to employ a relatively simple approach for implementing temperature-compensated MPPT. Because the temperature coefficient for a typical solar panel is essentially linear, engineers can implement basic temperature compensation by augmenting the resistor network with a 3-terminal temperature sensor such as the National LM234...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: DIY solar USB charger - MPPT?
    smasher wrote: »
    worth a shot, though, to just tie a cigarette-lighter plug to the panel and see how one of those works. thanks for the idea!

    Just make sure it has in input voltage rating of 12-24v and not something on the edge like 12-18v. I use a few Radio Shack 1.5A types that include the built-in cable - that way I'm not tempted to go hunting for a random usb cable that is wired as a data/sync cable and artificially limit my rate to only 500ma.
    of course, i still wouldn't get any MPPT to match the input impedance to the panel :(
    True, but personally, the use of MPPT really shines when dealing with panels larger than 200 watts, cold temps etc. In our little solar toy charging scenario, I think MPPT is total overkill, although a cool project. The links that BB points to are pretty exciting for DIY, but for me that would be incorporated into a much much larger system to reap any benefit from it.
    yeah, i think the latest gopros limit to ≈500mA when connected to a computer, but can definitely pull >1A when connected to a "dumb" power supply.
    Well, definitely hack a standard usb cable, short the data leads (typically green and white inner leads) facing the device under charge, and put a multimeter inline with the say the red power leads and measure charge current when the battery has been previously discharged more than 20% from full. (Charges that reach 80% capacity or more are in the absorb stage where current is naturally tapering by the battery itself.) Obviously only do this if you know what you are doing - safety first!

    WARNING - despite some products touting safety features like short-circuit, temp protection and whatnot, things do go wrong, and we are still dealing with LI-on battery (or a variant of Lithium) and they DEMAND respect. It is even more dangerous with counterfeit stuff. Some of the blatant safety violations I see with online video-tutorials makes me shudder.

    Also much of the charge current depends on how "smart" the device is. For instance, this "charge only" cable with the shorted data leads may not actually allow smarter devices capable of say 2A of current from receiving it. They may depend on a combination of resistive networks, current sampling, voltage floating on the data pins etc to actually allow for the absolute fastest charge. Then again, I don't charge my toys directly but sometimes charge a Li-Ion or Li-Poly external pack that accepts usb input, and they are typically dumb and are happy with the simple hacked charge-only cable.

    Note - some commercial cables that are labeled as charge-only may not actually be wired up as described above. All of this is kind of why I recommend those just getting into solar to cut down the variables and set up a small classical panel > CC > AGM battery. That way the smart devices won't make a newcomer tear out their hair when they do all the math and the charging actually takes twice as long or more than initially calculated.
  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Re: DIY solar USB charger - MPPT?
    smasher wrote: »
    5x gopro cameras can pull >5A. when i outgrow that, i'll get more/bigger panels.

    You mean 5 Gopro's can pull more than 5 amps? I can see that. However in general it's a bad idea to just wire them all in parallel. Those connectors are only good to about 2 amps, and if you exceed that significantly you can damage the connector. (Some USB devices rely on the charger 'folding back' above a certain current.) I guess a bunch of 2A fuses would protect the connectors.
    i guess if i fudge V-FB i could use it as a voltage-regulated power supply?

    If efficiency isn't important just use a regular buck converter and add a circuit to keep the input voltage from ever dropping below the peak power point of the panel (which you can measure beforehand.) You will not get "true" MPPT performance, nor will it compensate for temperature, but it will get you in the ballpark (and be relatively cheap.)