next »

[mahlness]

OK, here's an electricity 101 question... Is it possible for a solar panel hooked up to an XO to actually DRAIN power from the battery when plugged in, if there is no output from the panel (as in a very dark rainy day)?

It seemed this happened in my classroom today. I had two panels hooked up to 2 XO's (turned off). There was about an hour's worth of power left in the batteries from yesterday. When I tried to turn them on in the afternoon, after no charging taking place (it was a very dark day), I got nothing.

Plugged in and turned on via AC, the batteries were at zero. They then charged fine on AC.

I'm assuming this means I must not plug in my XO's to the panels unless there is a positive power output from the panels.... Whenever clouds appear and the battery is not recharging, is the battery actually LOSING its charge? Any thoughts appreciated! - Mark

This is a follow-up from a solar/battery forum post started a few days ago: http://olpcnews.com/forum/index.php?topic=3943.msg26773#msg26773

[teapot]

AFAIK, no, you can't discharge the battery through the AC power connector. There is a diode before the charging circuit that prevents this (and many other bad things) from happening.

[Dinky]

AFAIK, no, you can't discharge the battery through the AC power connector. There is a diode before the charging circuit that prevents this (and many other bad things) from happening.

Not so fast there.
  A.   There is at least one connection from the xo charging circuit to before the diode used for polarity protection.  Solar panel is different from AC brick, which has its own diodes.
  B.    Power diodes do have reverse currents that can add up over a very long time.
  C.   The radio stays on even though the xo is "off".   Thats why I remove the battery if the xo is idle over several weeks.

Does not add up though -- the experiment bears repeating.

[eden]

I'd like to throw in a comment about my experience today.  I started experimenting today around 9am.  It's winter in Utah but with hardly any clouds today.  I am indoors with the GP Solar 10W14V panel propped in a south-facing window.  I have it plugged into the XO.  I started out with about 75% charge in the battery.  

I noticed yesterday that the charge light came on only once, then went off, but I could hear a faint whining noise from the laptop as long as there was any sun at all.  It was a cloudy day and I started in the afternoon and didn't get any charge.  I lost 2% charge, but that could be because I turned it on and off to check the status.

Today I saw the charge light come on right away but it wouldn't stay on, even though the sun was strong.  It stayed lit for anywhere from about thirty seconds to about 4 minutes, which is is where I'm at right now as I type.  

As soon as the light was off, I tried pulling out the plug and plugging it in again.  The sun was bright but nothing happened.  I then unplugged the panel again, pulled out the battery and put it back, and nothing happened. I plugged the laptop into a wall outlet and the light lit up.  So I plugged in the panel again, and....nothing happened.  So I pulled the panel out of the window into shade.  Then as soon as I put it back up in the sun, the charge light lit up.  Then it became a little game.  Light off, panel down into shade, back up into sunlight, ta-da charge light is lit.  We played this game, my little XO and I, for about half an hour to see if I could get even a one percent charge.  Nope.  I'd lost a couple more percent.  Then, leaving the laptop on, charge light not lit, radio and backlight off, battery display graphic showing, I left the XO alone for a couple of minutes.  To my surprise, I saw the charge light came on and the laptop was charging while powered on.  Okay, not really charging because it lost another couple of percent.  But that dang charge light was on.  So I turned everything off again, panel into shade, back into sun, and the charge light came on an stayed on.  Till now.  I can still hear a very faint whine.  There happens to be a cloud right now, so I will see if the light recovers when the sun returns.

I wish I'd brought my multimeter with me.

Aha!  Light came back on!

I will update this post later to say if I gained any charge by lunch time.  

***A little while later...The sun is out in full force now.  The charge light is off but I can hear a higher-pitched whine.  I played the shade and sun game but can't get the charge light to come back on.  Hmm.

*****At lunch, about three hours after beginning the attempt to charge...I had to reseat the battery to turn on the laptop.  At startup, the battery charge percentage was the same as it was earlier: 72%.  If I gained anything, it was miniscule.  I left the XO alone for over an hour and a half.  During that time the charge light was not lit, but there was a whining sound and the sun was bright.

****Two hours later: the charger light was on for almost two hours after my last edit to this post but I am still at 72%.  Dah.

I have a working multimeter but I haven't found a way to measure the voltage coming from the panel.  If you have read down this far and you know how to do it, please tell me.  I don't know whether to blame the laptop, the battery, or the panel at this point.

[Mike Lee]

So you did all these tests indoors with diffused light coming through a window? If so, try leaving the rig outside a while in bright direct sun.

Mike

[eden]

Yes, indoors.  I measured the panel with your helpful instructions (stick red probe inside barrel, black probe touches outside of barrel) and found that the panel was putting out 16.5V at 9am without any load.  I have 17.1 now at 10am.  And yes, I'm still indoors.  I want to get outside and measure but I am stuck in meetings at lunchtime.

The panel is rated for 14V, which is supposed to be enough to charge the laptop.  Since I am apparently getting more voltage than that, why in the world isn't the laptop charging?  I have the latest 8.2.0 build and firmware.  This does happen to be the laptop with the developer key but I'm still running the regular Sugar stuff.  It charges fine from a wall outlet. 

When I tried yesterday to run the laptop from the solar panel with no battery, it was no go--the charge light flickered and there was a little clicking sound but it wouldn't turn on.

I thought the XO would accept DC voltage from 11 to 18V.

The amperage listed on the panel is 715mA.  I saw something on the wiki about charging the NiMH batteries under 760mA but it didn't say anything about the Li-FePo4.  The battery itself reads that it is rated 6.5Vdc--3.1Ah with a limited charge voltage of 7.5V.  I have a battery with a yellow sticker and the wiki said I have a BYD LiFePo4.

Does anyone have an idea?

[Mike Lee]

Interesting. Now you've got me thinking I should do a test myself with the panel as I wonder if the firmware version matters. I have what came with the release of 767, but I hear there have been several updates since.

But in your case, you defintiely want to avoid trying to charge from behind a window as that adds one or two panes of glass that cut the solar energy. Also the light coming in your window is likely diffused from trees or other buildings. And your laptop is off?

You are also just measuring voltage when output current (amount of power) is the other factor. See this PDF:

http://apps1.eere.energy.gov/education/lessonplans/pdfs/solar_1photovoltaicpvpoweroutput.pdf

If you haven't seen it, the Solar UMPC blog has done a great job over the last year covering real world use of solar panels (and outboard battery packs) with portable computers (unfortunately not the XO).

http://www.solar-umpc.com/

I'm still surprised that no one has written a practical guide to using commonly-found solar panels with the XO. Part of the reason is that until recent months, the firmware was in flux with frequent updates, and I imagine there are still bugs yet to be resolved. Maybe everything is stable enough now (or soon) to do some tests that would inform an article that would be helpful to a lot of people. Another future project for me.

Mike

[Mike Lee]

As much a note to myself: The PDF linked in the previous message from me is from a set of classroom lessons on understanding solar energy. What's nice about them is that advanced topics are covered. The material is produced by the Florida Solar Energy Center and also promoted by the U.S. Department of Energy. I think it might be a useful exercise to adapt these lesson plans for use with the XO and the 10watt panel.

http://www.fsec.ucf.edu/en/education/k-12/curricula/use/index.htm

Mike

[eden]

Fascinating.  Thank you, Mike.  I will look at that.

I did run outside for a bit around 11am and tested the panel at 19V and 20.1V, depending on the angle of the panel.  Not too much more than what I got through a window.  Fortunately, I am on the top floor of a taller building so the sunlight just beams right in.  Yes, the laptop is off.  I tried to charge with it on too but just continued to lose charge.

I am considering rolling back to 656 and the older firmware just to test the panel.  Or maybe upgrading to joyride something or other.  Hmmm.  Or I wonder what else I could charge with it...probably blow out my eee pc 701...

[mahlness]

Mike and Eden,

In over 2 months with two 10w panels and multiple xo's, I'm pretty clear on how to get the puppies charging, if there's enough light...

- unplug the panel from the xo
- reseat the battery (you'll see a brief orange battery light flicker when doing so)
- plug the panel back in

works every time, if there's enough light, needs to be done pretty much every time you want to start up charging process

pretty sure it's not a firmware issue, I wouldn't waste my time there

good luck - mark

[eden]

Thank you, Mark.  I am going to try your method today.

[eden]

Aha!  In 20 minutes time outside with the panel at about a 60 degree angle and the laptop turned off I managed a 5% charge.  This was after the mahlness method was completed.  Verrry nice!

I am now back inside with the experiment continuing in the window...

[Mike Lee]

Mark,

I'm going to make your method my standard!

Mike

[Dinky]

Mark:

Ah, so it is the cessation of charging that trips the battery controller ?  Then you have to reseat battery and plug in the solar panel while it is in the sun?

[mahlness]

Dinky, Great question... what exactly is it that trips the battery controller?

My experience says it may be several things:

- unplugging the computer from the panel and putting it "away" for the night
- sunlight fading, sun going down

Every morning I must reseat the battery before a solar charge will start. And at least once more during the day.

Bottom line, plugging in to a solar panel is not anywhere close to plugging in to AC for power. Never mind the efficiency of the panel. - Mark

Advanced search