PDA

View Full Version : Cheap Solar Cells to power Propeller



Philldapill
03-28-2009, 07:26 AM
I was looking on Futurlec.com today for some stuff, and ran across these (http://futurlec.com/Solar_Cell.shtml). It looks like you could easily string 3 or 4 of these cells together, and power the Propeller remotely, without batteries. The first thought that came to mind, was remote data logging out the the woods. Looks like you could do this for about $5-$8.

The low voltage units might be good when used in tandem with the buck/boost chip I talked about here (http://forums.parallax.com/showthread.php?p=793839).

mctrivia
03-28-2009, 08:09 AM
only think you need to remember is that expect only 50-80% of current output under normal lighting conditions.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

mctrivia
03-28-2009, 08:17 AM
not cheap but nice to see that large solar panels are getting cheaper.

15W for $99.99 CAN (http://www.canadiantire.ca/browse/product_detail.jsp?FOLDER%3C%3Efolder_id=253437430 3523074&bmUID=1238203086245&PRODUCT%3C%3Eprd_id=845524443275943&assortment=primary&fromSearch=true)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Cluso99
03-28-2009, 08:27 AM
The trick with solar cells is to use a buck/boost circuit.
I had a friend mount pair of 210W panels and an Outback MX80 multistage regulator on his boat. This is an overkill for this but the concept is the same. The cells produce around 36V and the regulator inverts and steps down to between 13.1-14.4V (we use AGM batteries). This recovers about 20% extra from the cells because of the inverter (after losses).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:

· Home of the MultiBladeProps: TriBladeProp (http://forums.parallax.com/showthread.php?p=786418), SixBladeProp (http://forums.parallax.com/showthread.php?p=780033), website (Multiple propeller pcbs) (http://bluemagic.biz/cluso.htm)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator) (http://forums.parallax.com/showthread.php?p=790917)
· Prop Tools under Development or Completed (Index) (http://forums.parallax.com/showthread.php?p=753439)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index (http://forums.parallax.com/showthread.php?p=778427))
· Search the Propeller forums (via Google) (http://search.parallax.com/search?site=parallax&client=parallax&output=xml_no_dtd&proxystylesheet=parallax&proxycustom=<HOME/>&ie=&oe=&lr=)
My cruising website is: ·www.bluemagic.biz (http://www.bluemagic.biz)·· MultiBladeProp is: www.bluemagic.biz/cluso.htm (http://www.bluemagic.biz/cluso.htm)

Tubular
03-28-2009, 08:35 AM
Clare have some tiny solar cell chips rated at 4 or 8 volts in an SOIC-16. They directly mount on the PCB.

They only put out 100uA each but are cheap and you can parallel them up around your circuit board

http://www.clare.com/Products/SolarCell.htm

tubular

mctrivia
03-28-2009, 08:44 AM
tublar said...
Clare have some tiny solar cell chips rated at 4 or 8 volts in an SOIC-16. They directly mount on the PCB.

They only put out 100uA each but are cheap and you can parallel them up around your circuit board

http://www.clare.com/Products/SolarCell.htm


that is the coolest thing I have ever seen. what can you do with 100uA though. how low of a draw can the prop possibly go?


Not really that cheap though. Digikey lists CPC1824 at $1.7064 CAN each for 25 at only 100uA each you need a few.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Post Edited (mctrivia) : 3/28/2009 1:54:52 AM GMT

Beau Schwabe (Parallax)
03-28-2009, 10:08 AM
The thing to keep in mind when shopping for solar cells is to look at the amount of Watts per dollar. Like any other high volume market they try to trick you into thinking your getting more than you actually are. If you can, "normalize" what is presented to you in a way that is even across the board, you will have much more purchasing power.


Philldapill,

The best buy on that page is the 2V 92mA @ roughly 96.8 mW per dollar, however this would not be the best fit for the Propeller. The best fit for the Propeller, and the best Watt per dollar from that page would be the 3V 45mA @ 84.38mW per dollar.

mctrivia,
If you look hard enough for them you can find 15W panels for about $4 per Watt or $60 per panel .... Even cheaper ($2 per Watt) if you purchase $10K worth of them. (about 334 of the 15W panels) .... $10K really isn't terrible for a 5000 Watt system.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 3/28/2009 3:14:03 AM GMT

Bean
03-28-2009, 10:17 AM
I was just wondering if anyone tried the solar panels I sold at the NorthEastern Propeller Expo ?
I still have about 400 of them.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
There is a fine line between arrogance and confidence. Make sure you don't cross it...

·

mctrivia
03-28-2009, 10:21 AM
$99.99 CAN = $80.6893 Canadian Tire is a local store. I never pay for anything at the store. I acquire so much of there fake money because I have one of there credit cards.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

mctrivia
03-28-2009, 10:24 AM
the boost convert mentioned will take 1.8V input and give 3.3 out so the cheapest will work.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

FearTurtles
03-28-2009, 07:30 PM
Yea but at the cost of how many mA? You cannot increase the efficience of a solar panal other then making sure it points directly at the sun and to keep the leads as short as possible. Solar panels are only about 18% efficiant as far as converting the suns energy into electricity. To know how many panels you need figure on the average solar day. Here in Michigan we get about 6 hour days with periods of months of no effective sun. Wind power is a must as an alternative. So figure on 6 hours of sun in my example. If the load averages 500 mW you would need two watts worth of panels and a sizeable battery bank to make it through the night. (500 mw * 24) * 2 So about 24Ah bank. Then multiply that by the max number of cloudy days you expect to have in a row. Over kill is a key to having a solid solar system.

The best deal that I have spotted retail for panels is at Trackor Supply Co. They have a 4-15w panel package for $350. The price of these went up about 8 months ago from $300 so I can say that they are getting cheaper. They sell a single 15w panel for $150.

Earl Foster
03-28-2009, 09:26 PM
I used solar panels in my 2nd high altitude balloon project (http://www.hapb.net/HAPB2.htm) to power the payload. Due to shape of payload there was always 2 panels facing the sun producing ~400ma. The panels ran $34.00 each for 7.2Vdc @ 200ma.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
WWW.HAPB.NET (http://www.HAPB.NET)

"Don't ask yourself what the world needs - ask yourself what makes you come alive, and then go do it." - H.T.Whitman

Philldapill
03-28-2009, 11:54 PM
FearTurtles,
When you say that you can't improve the efficiency of solar panels, you are only half way correct. What you CAN do, is couple the panels to the load better. If you have a 30V panel, and you connect it to a 12V battery, the panel (probably) won't be working at peak output. The reason is simple. Any power source has a certain internal resistance. If the open circuit voltage of the panel is 30V, and you connect it to the 12V battery, the internal resistance will be dropping about 18V. Sure, you will be getting more current the lower voltage of the load, but the POWER transfered to the load is not at it's peak.

What you need, is a means of stepping down the voltage to a more appropriate value for the load, while at the same time, maximizing the current output of the solar panel. That is exactly what a buck/boost converter does. In a way, it acts as a transformer - stepping the voltage up/down, and at the same time, stepping the current down/up. A buck/boost converter is a very efficient means of providing 3.3V or whatever voltage for the propeller, or other electronics. The great thing is, you can take a wide range of voltages, and convert them to your target voltage with very little power loss. Try taking 12V, running it through a 3.3V regulator, drawing 500mA, and see how much of a burn you get when you touch it. Buck/Boost converters should be ice cold to the touch. :)

Beau Schwabe (Parallax)
03-29-2009, 01:14 AM
All,

For most solar applications, the buck/boost converter will be operating in the buck mode allowing for increased current at a lower voltage, while the boost mode allows for increased voltage at a lower current.

Keeping the leads short are important, but the voltage loss due to resistance is negligible when comparing to the internal resistance of the solar panel.

For example, if a Solar panel is rated for 15 Watts at 12V, then it's capable of supplying 1.25 Amps under load. The same panel might show 22V without any kind of a load.
In this case, the internal resistance would be about 8 Ohms... Compared to a 100 foot run of #16 wire would only contribute about 0.8 Ohms

You must also consider charging current... With the same setup mentioned above (15Watt 12V panel), if the battery was low, say 10.5V... the 10.5V battery would demand 1.44 Amps across the internal resistance of the panel. Since this is no longer at an optimal power rating of the Panel, the efficiency will drop off considerably. This is where the buck/boost converter will ultimately "kick in" and shine ... The panels want or like to be at their rated current and voltage... so in this case, the buck/boost converter would operate in the buck mode where it would be capable of delivering the lower voltage of 10.5V at a slightly higher current to meet the demand.... Still 15 Watts, but the current and voltage are adjusted accordingly via the buck/boost converter.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe (mailto:bschwabe@parallax.com)

IC Layout Engineer
Parallax, Inc.

hinv
03-29-2009, 02:28 AM
Has anybody thought about making their own. I don't know what is needed to encapsulate them, but tabbing is supposedly really easy for the novice. Check out ebay auction 110366980470.
At less than a dollar per watt, and in discrete cells, you can put together most any voltage/current you want. Since these have 2 tying lines on the front, you could slice them in half to get smaller cells for the same voltage if high current is needed. I don't know if you just need a glass cutter to do it, but it would be good to have a bunch of these around for little projects. They could even be encapsulated on the back side of your circuit board for a nice sturdy package.
Does anybody know where to get the clear stuff to encapsulate these like the ones from Futurelec?

Doug

mctrivia
03-29-2009, 02:36 AM
MG makes clear poting compound not sure if that is what you would want.

I have tried making solar cells smaller in high school. Never succeeded though.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

FearTurtles
03-29-2009, 12:18 PM
Wait a minute now. Buck-Boost converters are only 81% efficient. 90% buck and 90% boost. How do you gain efficiency when you have a 19% loss?

mctrivia
03-29-2009, 12:32 PM
well the ic I am using are all 96 to 97% efficient. why are they more because analog methods are usually only 20 to 50% efficient. also buck converters can have greater current out then in. analog always has less current out then in.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

FearTurtles
03-29-2009, 11:22 PM
What I'm seeing from the two examples above, neither which consider the efficiency of the Buck-Boost . In the first example you are using an extremely inefficient setup to start. Connecting a 30v panel to a 12v battery will kill the battery in no time. So the voltage must be dropped and yes you will increase the charge amps in the process. However the over all wattage will be greatly reduced. It would be more efficient to match the Panels to the batteries.

In the second example you have a starting system which is much more efficient to start. However I believe if you subtract the power that the Buck-boost consumes and increase the charge voltage to one that will actually work you will find that you are still loosing efficiency.

As for the length of wire and size make a big difference the higher the amperage it needs to carry. A 100' run of #16 at 8 amps will suffer a 3 volt drop. If you are trying to charge a 12v battery with 13.8 volts you will have too great of loss to do it. This is a good example of where a buck-boost will work nicely. You could increase the voltage so as to be able to achieve some charge rate. Again this example stems from an extremely inefficient setup whereas if a higher voltage panel or a more suitable wire where to be used to begin with you would achieve a higher efficiency compared to using the Buck-boost.

mctrivia
03-30-2009, 01:01 AM
Example 3. I am a comercial electrician and to start would never use any wire in an example like that smaller then #10. You do have to remember though that wire gets very expensive and at least here in canada the maximum allowable voltage drop is 3%


#10 at 30.5m(100') is 0.2036 ohms

@1.25A - Vdrop=0.2545V (minimum v=8.5V)
@8A - Vdrop=1.628V (minimum v=54.3V)

to use a 12V panel at 8A you need #1 wire that would be very expensive and that is just to be legal.


There is a reason I run all the lighting in big where houses like home depot at 347V and motors at 600V power loss is = IIR and I=E/R so doubling the voltage quarters the power loss.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

hinv
03-30-2009, 02:21 AM
@Fear Turtles: The voltage drop depends on the amperage. It is not like a diode drop, so it would indeed charge the 12.8V battery if you supply 13.8 at the other end of the 100ft wire, just slower.

Philldapill
03-30-2009, 05:59 AM
FearTurtles, I have no idea where you are getting that efficiency rating for buck/boost converters. First of all, I don't think you multiply the buck eff. times the boost eff.
You don't use the output of one converter, as the input of another converter. What I'm talking about doing, is switching to buck mode when the panel voltage is higher than what is needed, and boost mode when it is lower.

Secondly, many converters are upwards of 98% efficient. When you are talking abou gaining 20% more energy from the panel using a step-down/up method, the 2% loss is well worth it.

Your point about connecting a 30V panel to a 12V battery is sort of correct. If the panel puts out more current than the float current of the battery, yes, you will kill the battery ONLY after it has charged beyond the float voltage. If you have a car battery and connect a 15W 30V panel to it, the panel voltage will drop to the voltage of the battery. The only problem is AFTER the battery is charged, it will overcharge, damaging it.

Do some research on Maximum Power Point Trackers, or MPPTs. These do exactly this. They make a psuedo-load using a buck/boost sort of converter, and constantly watch the output of the panel. These devices essentially do a good job of matching the ideal load for the panel, extracting the peak amount of power from the panel.

In your third example, the BETTER way to charge the battery and lose the least amount of power, is to wire the panels in SERIES so that they make a higher voltage, lower current. Then, on your battery end, put a buck converter between the panel wiring and your battery to step up the current. It may not be for every application, but if the power lost in the buck is less than the power lost in the wires using your original method(high current, low voltage from panels), then go for the buck. Otherwise, just stick with the high current and eat the R*I^2 loss.

FearTurtles
03-30-2009, 07:57 AM
All of the buck-boost boards that i have seen tend to be anywhere from 40% to about 85% efficient . It depends on the input voltage compared to the output. The greater the voltage difference the lower the efficiency. Just as I stated above. I did find one that boosted 97% efficiency it but did not indicate at what voltages. I will continue to search. Other then that I'm finished with this topic. LOL When you have two people attempt to correct you by elaberating on what you have already stated you know its time to move on.

@Hinv You're right I stand corrected.

Cluso99
03-30-2009, 09:57 AM
For MPPT see the article on the Outback MX60 (do a google search). This is what we use on boats and it works really well - we get > 20% more power from the solar cell.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:

· Home of the MultiBladeProps: TriBladeProp (http://forums.parallax.com/showthread.php?p=786418), SixBladeProp (http://forums.parallax.com/showthread.php?p=780033), website (Multiple propeller pcbs) (http://bluemagic.biz/cluso.htm)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator) (http://forums.parallax.com/showthread.php?p=790917)
· Prop Tools under Development or Completed (Index) (http://forums.parallax.com/showthread.php?p=753439)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index (http://forums.parallax.com/showthread.php?p=778427))
· Search the Propeller forums (via Google) (http://search.parallax.com/search?site=parallax&client=parallax&output=xml_no_dtd&proxystylesheet=parallax&proxycustom=<HOME/>&ie=&oe=&lr=)
My cruising website is: ·www.bluemagic.biz (http://www.bluemagic.biz)·· MultiBladeProp is: www.bluemagic.biz/cluso.htm (http://www.bluemagic.biz/cluso.htm)

Tracy Allen
03-30-2009, 10:00 AM
EDN has the Design Ideas column, and quite often there thought provoking contributions from Stephen Woodward. A case in point is last December:

Solar-array controller needs no multiplier to maximize power (http://www.edn.com/article/CA6619019.html?text=woodward)

The circuit trick is two transistors in series that effectively generate a signal proportional to the power being drawn from the solar panel. That signal is used in the feedback loop of a buck-boost regulator to dither up to the maximum power point. No microcontroller needed. You have to have a load that is able to absorb that power, which is of course not always true in battery charging, and an output control loop is not implemented in the Design Idea.

The LTM4607 regulator used there claims to obtain an efficiency of up to 98% at certain voltage & current points. It will be a while before I use one though (LGA package).

I have a similar problem as FearTurtles with the efficiency issue. I use panels for low power instrumentation, and the time efficiency is most needed is on short winter days when there is only a few milliamps available. I don't want to through that away on the quiescent current of the regulator.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com (http://www.emesystems.com)

mctrivia
03-30-2009, 11:24 AM
solar panels are always in multiples of 0.5V I guess you could run the prop at 3.5V but a regulator of some sort will be needed to get 3.3V Switching is always better then linear and some switching regulators can even be shut down to draw almost no power.

Oh if anyone buys that Ebay item I would like to buy 10% of it off you if you can ship to Alberta, Canada at a reasonable amount.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.