DIY Panel + questions

Eric_B
Eric_B Registered Users Posts: 14
I purchased 40, 4"x1.75", cells off ebay over a year ago. Finally got around to making a panel with them.

(picture)
First panel, on left, offers 4 voltage options - 1.5, 3, 6 & 9 volts via the phono jacks. If need be the second panel can be used in conjunction with the first to supply 12 or 18 volts.
http://www.fractalfreak.com/wx/DIY_Panel_1.jpg

(picture)
During storage/transportation the panels are reversed and sandwiched together to protect the cells.
http://www.fractalfreak.com/wx/DIY_Panel_3.jpg

The construction is a little ghetto - plywood, cardboard and lots of glue for the cell support. I slathered silicone rubber sealant over the cardboard supports and then dropped the soldered solar cell chains onto this. The sealant seems to support and hold the solar cells adequately.

Seems to work. The 1.5 volt setting had a little dc hobby motor going. I was easily able to power my old portable boom box/stereo (radio, cd player, tape, etc) off it. It runs off 4 D batteries - I used alligator clips on the battery springs to deliver the power.

I'm hoping to be able to power my laptop. I've got a Dell Mini 9 which doesn't consume much power. Today I spliced the mini's power cord and phono'd it. Thus I can plug it in to either the wall wort or the solar panel. If it's sunny tomorrow I hope the panel will power the laptop. The laptop's AC adapter supplies ~19 volts.

Which leads me to questions.. First I'm trying to figure how many watts my panel is good for. I know it should be easy, however my meter does not measure current > 250mA (quarter amp) so I can't tell how much current these cells are producing. The flea-bay seller told me each cell is good for >1.5 amps. However I'm guessing by how quick my analog meter floored and the 4x1.74" size the cells are good for no more than 1/2 amp (500 mA). They look like poorly cut polycrystalline 'ribbon' cells. So I've a spread of 10-25 watts.. Maybe I should purchase a better meter? (lol) Anyone offer a better guess on wattage?

How do irregularities in cell size effect the current each chain of cells produces?? Is it an average of all the cells' currents, or is the current only as good as the weakest cell?

When I was soldering up the cells I accidentally dropped the probes of the meter and shattered three of them.. (almost picked up the entire panel and threw it across the room at that point).

Was a considerable amount of tedious work to solder all the cells together - I used a 15 watt iron with a very dull tip. And the cells were more fragile than I anticipated. Definitely more fragile than glass. Very easy to crack or shatter. In the future I'll order a professional panel that's properly encapsulated. I didn't realize how these cells really need to be protected!!

Cost? $70 for the cells, another $30-40 in parts, plus a good weekend's worth of work. If the panel can run, or at least charge, my laptop I'll be happy.. Unfortunately I'm not certain I've got enough watts :(

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY Panel + questions

    One thing to be very careful of is fire (Panel Fire Question)... As you have seen, home made panels are difficult to build and seal for long term reliability--And, in may cases, the materials used to make these panels is flammable and can be pretty easy to light off is there is an arc or hot spot in the cells or the panel wiring.

    Do not place the panel somewhere unattended where a fire would be an issue (over dry brush, over a wood shingle roof, etc.).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    BB - thanks for the advice. I'm not planning on using the panel unattended. I don't think I've got enough volts nor watts here to be much of a danger. However the panel is indeed very flammable.

    Most of the time the panel will be in storage no doubt. I live in the dank and cloudy midwest.
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: DIY Panel + questions

    Do not connect solar panel to laptop directly via DC input. Your solar panel will burn up your laptop. You need to feed your laptop a steady regulated voltage. For that you need 12Vdc to 17Vdc (or 19V) dc-dc converter powered from 12V battery to which you can connect your panel via charge controller.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    (picture - attempting to run mini off panel. failure)
    http://www.fractalfreak.com/wx/Solar_fail.jpg

    I tailored this panel to feed my netbook ( to mimic its AC power adapter), so I'm not too concerned about frying it. I agree it's best to use a battery + inverter, however you've got to admit it's a little crazy to go from DC (panel->battery) then AC (battery->inverter), then back to DC (power supply for appliance) for a small setup like I'm using.

    Voltage is a quantum effect and is already in effect 'regulated' .. it's the current that varies dramatically depending on light intensity and the surface area of the cell.

    My Dell mini is an atom based computer and draws around 10 watts most of the time. The 32 Watt hour battery runs the computer for over 3 hours. However if you want to run the computer AND charge its battery the power consumption is more like 30 watts (still quite low for a laptop).

    The AC power adapter for the Dell supplies 19 volts @ 1.6 amps (close to 30 watts). My panel supplies ~19 volts @ 1 amp, so I was a bit short (around 20 watts). When I plugged in the laptop to the panel it did not recognize it.

    Anyway, I answered my own questions by borrowing a friend's DMM. First off the 4x1.75" solar cells I'm using are indeed good for 1.3-1.5 amps on a good sunny day (we've has clouds and haze all week, there was brief 1-2 hour window of high calorie sunshine yesterday so I was able to test things at lunch). The current in a chain of solar cells is only as good as the weakest cell - it's not an average. This means the broken cells in my panel are significantly effecting its overall power output - I'm getting around 1 amp at the 18-19 volt setting, when I should be getting at least 1.3 amps -- those two chipped cells, which lost 10-20% of their area, are reducing the panel wattage from ~25 watts to around 19 watts!!! DIYers take note.

    I'm planning on creating another panel using 2 amp cells (one watt). This'll be a 36 cell panel, supplying 19 volts @ 2 amps. 35-40 watts. I'm hoping this will be enough to power my netbook AND charge its battery about as quickly as running it off its AC adapter. The extra current will make up for times when its hazy or there are high clouds.

    For me DIY is all about creating something - that's the fun part. If what you create is actually useful all the better. :) These cells are now so inexpensive it's worth doing a little experimenting
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: DIY Panel + questions

    You did it anyways. Oh well, have fun burning up your little computer.

    Look, you need steady source of 19Vdc between 0 - 3A. This means, your power source has to keep it's voltage always at 19V guaranteed stable between 0 and 3 Amps. You netbook should not demand more than 2 - 2.5 Amps. PV solar panel will never be able to achive this requirement by itself, no matter how large or small you size it. Insolation conditions wary, so does load current of your netbook.

    I agree that converting DC to AC and back to DC is silly. That's why there are these things called DC-DC converters. There is one that can take 11 - 16Vdc and convert it to stable 19Vdc and supply current between 0 - 3A for your netbook at decent 80 - 90% conversion efficiency. The one in the middle wil lwork for you. Now you need small gel-cell 12V 6Ah battery and small PWM charge controller with low voltage load disconnect to protect the battery from overcharging and overdischarging.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    First off thanks for the advice on the battery and dc-dc converters

    I don't understand what you mean by stability. If I'm using the panel on a clear day its output will be stable. There's no way I'm going to use the panel+netbook unattended. The regulator in the PC should be able to handle this. The computer is not going to be able to tell the difference between the panel and AC adapter. As far as the computer is concerned it will be running off AC power. At least that's the idea.

    I'm not worried about fire and flames. I'm not going to be push any more watts through the computer than the AC adapter. The voltage and current will match the AC adapters output.

    Yes and watch for any clouds :)
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: DIY Panel + questions

    How are you going to keep panel votlage from jumping above 19V?

    Each polysilicon PV cell is around 0.5Vmp. 36 cells in series = 18Vmp. That is only true when panel is loaded at specific current amount which forces the panel into it's maximum power point at cell temperature of 25 degrees Celsius and insolation of 1000 W/m2. Solar cells also have a property called open circuit voltage. Voc is usually about 1.25 times higher than Vmp. Now your supposedly 18V panel can also ouput 22.5V when current load connected to it is less than what panel can deliver.

    If you want to experiment with direclty connected PV panel to your netbook, at least have voltage controlled current shunt paralleled to solar panel output. If you can find where to buy 1N4555 zener diodes, you can connect 3 of them in series across your PV panel, and they will limit the voltage to 6.8V * 3 = 20.4V up to 6.5 Amps. Aluminum heatsink will be required to dissipate the heat. Or you could build simple zener voltage clamp circuit by using large bipolar transistor as current shunt.
  • Kamala
    Kamala Solar Expert Posts: 452 ✭✭
    Re: DIY Panel + questions

    AntronX is right. Use a charge controller or live & learn.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    Oh no, I wasn't aware about open circuit voltage. I was assuming these cells never exceed .5-.55 volts. It appears I'll have to invest in some zener diodes to avoid possibly destroying my cute netbook.

    Yes I'm still interested in trying to run the computer directly off the panel.

    Thanks for the advice!

    I'll post back here if I ever have success, or for that matter if I toast the computer :(
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: DIY Panel + questions

    Correction on zener part number - 1N4555B
    I cannot find them on mouser.com anymore, try calling your local electronic parts retailers. You will also need mica insulators for them to be mounted on aluminum heat sink, and heatsink thermal grease. But I am not sure if they will work well for your application. You just have to experiment and verify they work before connecting to your netbook.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: DIY Panel + questions

    you might be better off with a linear regulator than a zener.
  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: DIY Panel + questions

    I have an Acer Aspire One ZA3 / 751h. I use one of these:

    http://sangertek.en.made-in-china.com/product/hbRQnajdHqWU/China-Replacement-For-Gateway-Car-Charger-19v-4-74a-YD190-474-.html

    18 bucks off eBay.

    I had originally bought a smaller car charger, but the first one only lasted 2 hours and the free replacement quit after a couple of weeks. They also got very hot. This one is MUCH beefier and works great and barely even gets warm.

    The label on this one says:

    Input: DC 10v-15v max 10a
    Output: 19.4v 4.7a continuous.



    Here's a very strange thing I noticed...

    One day I let the house battery in my camper get too low without noticing it, and in Windows 7 I checked the battery meter and it reported, "74%, plugged in, not charging". Yes, NOT charging. That was pretty weird, but Win7 apparently understood that there was enough power coming in to run the computer, but not enough to continue charging the battery. I have no idea WHERE it got that information.

    When I checked the adapter, the LED was pulsing bright/dim at about 1 second intervals. Whacky.


    For the homebrew thing, I would say get one of these adapters, and stick a small 12v motorcycle battery between the solar cells and the power adapter as a buffer. Then you have a "proper" power supply to handle the voltage regulation to the netbook.

    If you do buy an adapter make sure of the power jack. A buddy of mine bought a brand new Acer laptop (not netbook) off Amazon a couple of weeks ago, and my 12v adapter wouldn't plug in because the center pin on his laptop is a bit different size than the center pin on my netbook.

    I have a 7200mAh battery for my netbook (8+ hours) and am seriously considering getting one of these babys:

    http://www.thinkgeek.com/gadgets/travelpower/e193/
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    OK, I've come to my senses. Thanks for the info on these dc-dc converters/regulators. Pretty slick. I think I'm going to get the one recommended off this link - the 90 watt PST-P90W. It's flexible and should be able to power a variety of gadgets besides my netbook.
    http://www.powerstream.com/ADC-p006.htm

    Another advantage of using the regulator is it allows me to use more commonly available solar cells, like these 3x6" cells which are all over ebay now:
    http://cgi.ebay.com/72-3X6-solar-cells-w-chip-1-8w-short-Tab-DIY-Panel-/140456136657?pt=LH_DefaultDomain_0&hash=item20b3d68fd1

    My latest brain fart.. I'll make my next panel with 24 of the 3x6" cells. Each of these cells is good for ~1.75 watts, so it will be a 12 volt, ~42 watt panel. Plug the panel into the dc/dc converter, and it will feed the laptop a regulated 19 volts, at whatever current it needs (1-2 amps?). Assuming an 80% conversion efficiency from the dc converter it will be able to supply ~33 watts for the netbook. The netbooks AC adapter supplies 19 volts @ 1.6 amps (~30 watts).

    So I'll be running the netbook directly off the panel, through the dc regulator. As far as the netbook is concerned it's running off AC power. This will allow me to use the netbook and charge its battery off the panel as fast as using the AC adapter - about 1-2 hours to fully charge the laptops 32 watt hour battery.

    If a cloud passes over the sun or it clouds over no problem. The laptop switches to battery power and the power indicator light starts flashing meaning there's a problem with the AC adapter (heh, I've already determined this).

    Seems relatively safe.

    I'm going to put a premium on portability on the next panel, perhaps have it break down to four stackable sub-panels for storage and transportation.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions
    dwh wrote: »
    (...)

    I have a 7200mAh battery for my netbook (8+ hours) and am seriously considering getting one of these babys:

    http://www.thinkgeek.com/gadgets/travelpower/e193/

    If I wasn't looking to DIY something that's what I'd get. What I don't like is it takes 8-10 hours for the thinkgeek panel to fully charge it's internal 72 w/h battery.

    The panel I'm envisioning will be able to charge my netbook's 32 w/h battery as fast as the AC adapter - about 1-2 hours. And run the computer.
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: DIY Panel + questions

    No, it won't work without 12V battery between solar panel and dc-dc converter. You do need an energy buffer in between, that converter was not designed to work like you want. The battery does not need large capacity, you could get by with only 3 Amp/hour 12V battery. But you cannot use gel-cell lead-acid because it cannot handle high charge rates like 1C or 2C. C stands for capacity which measured in Amp/hours. When I say charge rate is 2C it means 2 * C = 2 * 3Ah = 6 Amps charging current. You need Lithium battery, like LiFePO4 for it's higher thermal stability to reduce risk of battery catching fire. But they require their charging voltage to be limited to no more than 3.6V per cell and they cannot be discharged below 2 - 2.5V per cell. And all cells have to be balanced, or balance charged.

    Another mistake of your's is only using 24 cells in series. You need at least 32 or better 36. Because solar cells reduce their voltage (Vmp) as their temperature increases. When panel is lit by full sun, cell will be hot. For example, my Evergreen panels are rated 170 Watts at 25 deg. Celsius at 1000 Watt / meter square solar light intensity. But usually I get about 85% of full power due to in the full sun, they are almost too hot to hold. This power loss is mostly due to lower Vmp. So you can (roughly) estimate that 18Vmp * 85% = 15.3Vmp at full sun and hot panel. Add some wire resistance voltage drop, and your lead-acid battery sees around 14.5Vmp - near highest battery charging voltage. This is why most (if not all) solar panels meant to charge single 12V lead-acid battery, have 36 individual cells in series.

    If you still want to charge your netbook directly from solar panel, I think it can be done, but I don't think it will be as efficient at utilizing available energy from solar panel compared to "PV > CC > 12V batt. > dc-dc > netbook" method. Few things you need to make sure, to make it work. First, size your panel to have Amp rating of at least 4A in less than perfect sun, or about 30 degrees off axis. This will require 6"x6" cells rated at 8A. Then you will need your panel Vmp to be at around 21V / .85 = 24.7Vmp. That means you need at least 50 cells in series. You can use 3"x6" cells, but you will need to make two panels of 50 cells series each, and parallel the two for 8A maximum total current capability.

    The final and most important step is to limit this wildly varying PV voltage and keep it locked at 19V during sun. This can be done with simple voltage regulator based clamp circuit rated at least 15A current capacity. High power zener diodes are not practical in this scenario due to high current capacity and voltage precision required. This is how it will work: Your 24.7Vmp panel will be more like 21Vmp panel when it heats up to it's normal operating temperature in warm summer weather. But the voltage clamp circuit will essentially force the panel voltage output down to 19V by absorbing panel current. The power dissipated into heat when no laptop load is connected, and the sun is bright, can be 19V * 8A = 152 Watts. You will need big heat sink with fans to keep power resistors and pass transistor in clamp circuit from overheating. But in most typical hazy sky you will see 4 - 6A. When you connect your laptop, it will further "take away" 2 - 3 Amps, and power dissipated into heat by clamp circuit will be in 19 - 95 Watts depending on conditions.

    The last scenario will work, but solar panel must be grossly over sized in order to obtain enough current "reserve" in case of varying solar insolation. If your panel output current falls below laptop's instantaneous current demand, the voltage will drop below 19V and laptop will stop charging internal battery, thinking you just disconnected the power supply.

    Now, there is a proper best way this should be done. Your laptop's internal battery needs to be connected to solar via properly designed charge controller with cell balancing included. This will require disassembling battery pack case to attach wires to each cell terminal. Then wires will be connected to lithium battery balancing circuit and cell voltage monitor circuit to prevent cell overcharging. This method is best and most efficient, and will only require small solar cell, smaller than you have now. But this method is most complex and requires electronics knowledge with lithium charging process understanding. Hey, here is a business opportunity for some EE to make modified laptop computer battery packs with necessary circuitry to charge from solar panel directly without dc-dc converters, neither via the laptop internal charger. Maybe I will take a stab at this in the future.

    This post is getting long without me explaining fully every implication of your project idea. There is a lot to this stuff than just simply hooking up some cells to the netbook.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    Perhaps my latest brain fart is no good?

    You say 'that converter was not designed to work like you want'.. Can you elaborate on this, and have you ever attempted to use a similar converter like I'm attempting? I was assuming the converter would work with a varying load on the input side and the voltage would remain stable on the output side.

    Also, I think there's a slight misunderstanding. I'm not attempting to charge the laptop's battery directly - much too dangerous. I'm attempting to run the laptop directly of the panel - the laptop already has sophisticated logic/circuitry to charge its battery. What I have in mind is a panel, or panel + converter | voltage clamp, which will power the computer exactly the same as its AC adapter.

    I thought my netbook was a good candidate because it consumes relatively little power. However as is nearly always the case these things are more complicated than one assumes.

    I appreciate you taking the time to help me with this!
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    Darnit it's cloudy here, again. Been cloudy all week. Makes it hard to test stuff :(
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY Panel + questions

    I believe one issue with most all voltage converters connected to a solar panel without any energy buffering in the system is this...

    Solar panels output a constant current rather than a constant voltage.

    Virtually all of the power supplies that you can purchase (with the exception of MPPT solar charge controllers and linear current boosters for water pumps, etc.) all assume that the amount of energy on the input is >> than the amount of energy the voltage converters/regulators will draw.

    With solar panels and constant current output--that is not a good assumption...

    As long as the converter is drawing less than Imp (or whatever current is available with the current sunlight conditions), the Voltage will be above Vmp and the current will be below Imp.

    As soon as the voltage converter/power supply draws 0.0000001 amps more current than is available from the solar panels, Vpanel-out will start to collapse. If you are using a switch mode power converter of some sort, its input power is P=V*I--

    So what happens when V starts to collapse, the converter wants to draw more I--(to keep Power=needed energy for conversion)--So the Vpanel will continue to drop as the converter tries to make up for lost voltage by drawing more current... This is a "positive feed back loop"--Basically, the converter will draw down Vpanel to its minimum input voltage--and either stick there, or "turn off"... So, either the converter will not get enough power (and may continue to supply some output current--an OK thing) or it will turn off, Vpanel will rise, converter will restart, and the whole process start all over again.

    For "smart" charge controllers, the constant resetting caused by Vpanel collapse (and local microprocessor resets) can cause the battery charge controller to forget the current charging state--and try to recharge the battery from some sort of initial condition state... And probably will result in less than ideal charging performance.

    A PWM type charge controller just lets whatever current is available / appropriate for charging the battery, through to the battery for charging. And the battery really sets the Vpanel output voltage (charge controller only reduces current as the battery approaches full charge--by monitoring the battery voltage, temperature and possibly other conditions). There is no collapse involved and no fancy electronics tracking Pmp=Vmp*Imp going on...

    A MPPT type charge controller, on the other hand, has sort of two sections... One is the input section that monitors voltage and current from the panel and controls the current draw based on the MPPT calculations and current panel voltage--And and output stage that monitors the directly connected battery (or utility power in the case of a GT Inverter) for proper output conditions.

    With a Solar panel to DC Voltage Converter to LapTop input Charge Controller, you have a very difficult task to design a circuit that can meet the input requirements of the laptop battery charging/power management circuit (that probably assumes one of two state--Either unlimited power, or no power/unusable power input).

    If you had a DC Converter that did the MPPT calculations to only take Pmax (based on Vmp*Imp)... And did its best to keep the Voutput voltage at 19 volts, but let it drift down as input solar power drops--You may get something to work (really dependent on laptop input power design--can it take 100% at 19 volts input and 50% at 18 volts and 0% at 17 volts--and still be happy?).

    By you sticking a solar panel with Vmp > ~17-19 volts, and placing a ~20 volt clamp / shunt regulator on the output--You may not have the most elegant MPPT type charge controller--but you may have at least something the computer's DC input could work with--Again, if it gracefully drops charging current as available voltage/current slowly drops.

    You could do the same experiment with a bench top lab power supply (something like this one will do 30 volts / 3 amps, or 20 volts * 3 amps = 60 watts for your testing--~$200; or a smaller one that will do 30vat 1.2 amps, or 20 volts * 1.2 amps = ~24 watts--will not work at full power, but will allow you to understand your computer's low energy input ability--only $90 or so)...

    Set the DC output voltage to ~19 volts--And then start "playing" with the output current regulation settings. See if the computer can take significant power when the DC current regulation is set to less than the input current it is demanding at that time (battery charging and battery charged, and just supplying energy to the computer operations, screen off, screen on, idle, sleeping, etc.).

    A little experimentation will tell you a lot before you try and build a large linear shunt controller (or regulated digital shunt controller with load resistance) for your solar panel (which would have to dissipate your full panel wattage of XX watts--not a small controller).

    Not saying it cannot be done--just that we don't know enough to understand if it can be done with Brand XYZ laptop computer. :confused:

    -Bill

    Or that it may be simpler / easier and more convenient to purchase a MPPT (or PWM) charge controller and a "smallish" AGM battery to build a solar PV power system and use the LVD (low voltage disconnect) available on some charge controllers to cut the power to the laptop when the battery is nearing full discharge.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: DIY Panel + questions
    Eric_B wrote: »
    If I wasn't looking to DIY something that's what I'd get. What I don't like is it takes 8-10 hours for the thinkgeek panel to fully charge it's internal 72 w/h battery.

    The panel I'm envisioning will be able to charge my netbook's 32 w/h battery as fast as the AC adapter - about 1-2 hours. And run the computer.

    My ZA3 with 7200mAh battery runs about 8 hours and takes about 4-5 hours to recharge from the AC adapter if I'm using the computer at the same time. I haven't checked to see how long it would take if I didn't use the computer at the same time.

    I also haven't done anything to see how long it takes to recharge from the DC adapter, though I should check my watch one of these days and find out.

    In Linux, I was seeing acpi report 9000mw-13000mw power consumption depending on CPU throttle (800mhz-1300mhz) when the battery was fully charged. There was a bug however, since when the big battery was charging acpi was reporting consumption of 714,000mw (no, that's not a typo), which can't be right.

    I recently upgraded my Back|Track install from BT4 to BT4.1 and haven't checked the acpi since doing that, so maybe the bug is fixed.

    Then again, maybe not since I have noticed since the upgrade that with the battery charging, acpi sometimes does, and sometimes does not report the time remaining to full charge. This was happening in BT4 as well so maybe they haven't changed acpi at all.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    BB - thanks for the information. I'm starting to understand why this is a more difficult problem than I imagined. I'd assumed PV was constant voltage, and it was the current that varied. The reality is current is constant (assuming steady atmospherics) and the voltage varies.

    You lost me with all the descriptions of various regulators, though I've been doing some googling. :)

    I'll have to do more homework. First I need to get a decent DMM and meter how many watts the laptop is actually consuming. Since it charges its 32 W/h battery in less than in hour its probably consuming > 40 watts when running and battery charging. I'll also have to meter how much power the computer consumes when not charging the battery. It varies dramatically depending on CPU utilization and graphics use - you can feel the computer heat up and cool down.

    Once I know its power consumption I'll likely attempt to build a panel with some sort of voltage clamp/shunt circuit and see if this works. It also helps to know the details of laptop's power management circuit, as you mentioned. What I'm attempting is a little risky.

    I'm also starting to understand why the PV panel is just a fraction of the equation. If you want to do this properly - charge controllers, inverters, and some sort of battery bank are essential given the intermittent nature of solar power.

    Again, thank you people for your help!
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY Panel + questions

    If you are interested in DC power--These DC Amp*Hour / Watt*Hour meters are pretty nice (used by radio control hobbyists).

    To start with regulations--Read about:

    Linear Regulator
    Buck Regulator

    There are, of course, many more regulator topologies but the 3 Terminal Regulator and the Buck Mode Regulator are two to study first.

    The Linear Regulator is a very "quite" but very inefficient form of regulation.

    The Buck Regulator is very simple and efficient, but can be quite "electrically" noisy.

    The each have their place--but in general, the Buck Regulator (and other switching type regulators) are going to be what you find in the majority of applications these days.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    These people also have done what I'm attempting to do.. I'd get this setup with the 25 watt folding (CIGS) panel:
    http://www.21st-century-goods.com/product/SolarSolutionForMacBooks.html

    Specifically this device, which apparently can be charged directly off a fickle PV panel:
    http://www.21st-century-goods.com/product/Tekkeon-My-Power-All-Plus-3450.html

    The Thinkgeek charger also looks good.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY Panel + questions

    They both are an external battery pack plus electronics (charging/power supply):
    Electricity provided from solar panels provides variable voltages due to clouds and inconsistent weather conditions. Since laptops require a clean and consistent power source, power from solar panels must be run through an external battery that can charge from inconsistent voltages but provide a consistent power source to the sensitive electronics of a laptop. This is why the Tekkeon my Power all Plus is a perfect external battery for your laptop. It takes all the charging it can from solar panels with vary voltages and gives your laptop the clean consistent power it requires.

    Nice little setup--But know your power requirements and what you will get from the sun... And remember, that the solar panels (at least) need to be out in full sun 6+ hours per day to get most of the energy from the sun.

    So, you end up with a bit of security issue too (panels+battery pack+laptop charging in the middle of day).

    Certainly can be done--and if you are off-grid for long periods of time--then solar is not a bad solution. Taking solar on hikes for a few days/week at a time with a laptop/note book computer--An external Lithium primary cell pack (like CR123 cells or something similar) would probably be easier and more predictable--and would let you recharge when you need the power instead of having to setup in the middle of the day.

    Also, looking a a real low power computer that does just enough for your needs (no DVD's in camp)... Conservation will cut down on the weight you have to pack around.

    If this is a fixed setup--then look at standard solar panels instead of the fold-ups.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bmet
    bmet Solar Expert Posts: 630 ✭✭
    Re: DIY Panel + questions

    Borrow your friend's DMM again, and you will see just how UNSTABLE a panel's voltage can be a a clear day.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: DIY Panel + questions

    It is not just that solar panel voltage is "unstable" during the day--it is the basic problem of the solar panels being a current source and that the loads need to accept the characteristics of being operated from a current source.

    A simple PWM charge controller and battery work OK together because the battery will simply accept all of the current the panel can provide as long as the panel output voltage is above the battery charging voltage (plus a little bit for voltage drops in the system).

    A solar panel, more or less, maximum output voltage is reached when weak sunlight is falling directly on the panel (and near zero amp output). As the sunlight strengthens, the available output current goes up. The battery will accept any current from zero amps to the maximum available from the panel.

    For active loads (like electronics and battery chargers)--They assume (more or less) a voltage source (fixed voltage, and unlimited current up to the rating of the system/device).

    When you place such a load on a solar panel--it will work as long as there is enough sun to generate more current than the load is using (the voltage will be somewhere between Voc and Vmp (voltage open circuit and voltage maximum power). Once the circuit tries to draw more power than is available, the solar panel will act like a current source and its output voltage will crash--giving the active loads no useful energy (unlike the above example where the battery simply accepts the available current from the solar panel--a lot or a little).

    Of course, devices can be designed to operate directly off of a solar panel (well pumps are a popular example--pump water into the tank when the sun is shining--pump a lot of water in bright sunlight, pump less water in in cloudy conditions).

    However, usually people put a battery bank in the middle... People generally want to run their power day or night, sunny or cloudy weather. Charge the battery when sun is available and store the energy for use at any time by standard DC or AC devices.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • James
    James Solar Expert Posts: 250 ✭✭
    Re: DIY Panel + questions

    Listen to these smart folks. You need a regulated power source for sensitive electronics.

    I know, I have done everything wrong at least once! :)

    When I was a younger man, I tried to connect a 100 watt PV panel directly (no battery or regulation) to one of those cheapie 300 watt vector inverters...i smoked it in 10 seconds...input voltage to high.

    Lesson learned.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: DIY Panel + questions

    You need to examine the specs of the DC-DC converter, what it's acceptable input range is, how will it behave if there is little load on the output ....??
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • dwh
    dwh Solar Expert Posts: 1,341 ✭✭✭
    Re: DIY Panel + questions

    FYI: Just checked my rig in Linux and acpi reports 13752mw consumption when charging using the DC adapter, with the big battery fitted and with my Alfa AWUS036H plugged in and cranked up to full 1w power output mode.

    That 713,000mw error was probably due to some problem with the crappy little adapters I had before.
  • Eric_B
    Eric_B Registered Users Posts: 14
    Re: DIY Panel + questions

    Well, I've given up on trying to power my laptop directly from the laptop. :(

    In addition to the Thinkgeek charger (already discussed in this thread) I came across this device ( Tekkeon MP3450):
    http://www.tekkeon.com/products-mypowerall.html

    It's got a 50 W/h battery, and can be used to power/charge a variety of devices - cell phones, PDAs, netbooks, DVD player, etc. It's got a variety of voltage options. For $20.00 you can get a solar panel charging adapter which will allow one to charge the Tekkeon off a 10-40 watt panel with at least 15 volts (open circuit). Apparently the device can be in use and charging off the panel at the same time. The panel should be able to recharge the device in 2-4 hours.

    This place has some small foldable panels, and uses the Tekkeon:
    http://www.powerenz.com/store/index.php

    (No, I don't represent either of these places)

    For those looking for small portable solar options this may be worth looking into.