Hello All. First time poster and EE freshman student here.

A while back I bought a fun little portable solar phone charger similar to this one. It eventually died, but I held onto it. Recently, I took it apart and tested the panels to see if they still work, which they do!

I thought it would be fun to slap them together into an array and regulate their voltage directly into a USB charger (no intermediate battery like the original product.) Each panel outputs 5.5-6.9 V, and 100-200 mA depending on sunlight.

Instead of putting them in series two-by-two to get up to 12V only to regulate back down to 5V for USB, I hoped I could simply get a low-dropout linear voltage regulator and go from ~6V to 5V. This one seemed perfect, I bought three, but they're all letting the current pass through at whatever voltage it darn well wants, both above 5V and below.

My circuit was originally more complicated (with various indicator lights and transistors to control them,) but when I didn't get the result I wanted, I reduced it to the simplest example I could find: Two capacitors on either side of the regulator. Still no luck.

I'm thinking my rookie eye for datasheets has resulted in me missing a crucial stat that I failed to understand. Any ideas?

Thank you.

 

You are running it close to the dropout voltage, but it should work at 200mA. It's not complicated - connecting it backwards is usually the issue. Have you tried hooking the panels in series and running the regulator at 12V to see if it is a voltage headroom problem?

 

I actually hooked it up to a 9V battery with a potentiometer and used as a poor-man's power supply. As I turned the dial, the voltage-out adjusted to match, bottom to top.

I also noticed that although the component page has some info, the actual model number isn't mentioned in the datasheet they include. Is that a red flag?

I'd be happy to buy a regulator with an even lower drop out, but it seems lower drop out means lower max current.

 

Are you sure you have it connected correctly?
The following is from the data sheet:

LM2940-N, LM2940C SNVS769J

7.4.1 Operation with Enable Control

The LM2940 design does not include any undervoltage lockout (UVLO), or enable functions.
Generally, the output voltage will track the input voltage until the input voltage is greater than V OUT + 1V.
When the input voltage is greater than V OUT + 1 V, the LM2940 will be in linear operation, and the output
voltage will be regulated. However, the device will be sensitive to any small perturbation of the input voltage.
Device dynamic performance is improved when the input voltage is at least 2 V greater than the output voltage.

 

Oh fascinating... I was unaware of the lack of UVLO. I'll definitely be needing that. Actually, I wasn't aware of the term until now, so thank you for the education.

Is it odd, or standard practice that the LM2940T-5.0 is not specifically named on the data sheet?

 

I actually hooked it up to a 9V battery with a potentiometer and used as a poor-man's power supply. As I turned the dial, the voltage-out adjusted to match, bottom to top.

That doesn't sound like such a good idea. What resistance was your Pot, and how did that tie to your load+Iq? Did you meter the voltage of the pot wiper/regulator input as you adjusted it?

If your pot resitance was too high, the regulator+load would pull down the regulator input voltage, and if the resistance was nice and low it would load the battery.

 

Oh my… seems I was out of my depth. It was a 10k pot in series between the positive terminal and the rest of the circuit. And I only metered the voltage after the regulator. Error on my part?

 

Yes. A resistance in the '000s will have reduced the input voltage considerably given even just the quiescent current of the regulator, not even including the load.

 

The LM2940T-5.0 regulator should regulate at 5V.
Double check the pinout configuration and your wiring.

 

The dataset shows a minimum load of 5mA for regulation. Did you have a load on the output when you measured it? Try putting a 1000Ω resistor across the output.

Bob

 

Aahhh... there it is. Thank you Bob, and thank you all. Still learning here.

 

Follow-up question: Now that the USB out will be protected from over voltage, shouldn't I have something that prevents under voltage? Or do phones/devices simply stop charging if voltage dips too low?

 

Follow-up question: Now that the USB out will be protected from over voltage, shouldn't I have something that prevents under voltage? Or do phones/devices simply stop charging if voltage dips too low?

Yes, they will likely just stop charging.
I highly doubt there will be a problem if the voltage drops below the charging voltage.

 

Most low dropout regulators do not work properly if the output capacitor is too small. the datasheet for the LM2940-5 says 22uF minimum so use a 33uF or 47uF.

 

I used a 1W zener diode with a 1W solar panel to clamp the voltage. This has problems of its own but works ok for my hobbyist purposes. You could add some circuitry as I have that allows the battery to charge only when the zener is clamping to the desired voltage and when enough current is flowing to maintain regulation.

 

Interesting... On the other end of the circuit: What about putting different panels of slightly different voltages together in parallel? I have seven different panels, salvaged from a total of four different amazon products over the years.

I know different voltages in parallel is generally a no-no, but does the regulator get me out of that bind? Here's a pic of my "ideal" janky layout. If it's a no-go, I'll just pull off the three oddballs and stick with the four identical ones.

 

Interesting... On the other end of the circuit: What about putting different panels of slightly different voltages together in parallel? I have seven different panels, salvaged from a total of four different amazon products over the years.

I know different voltages in parallel is generally a no-no, but does the regulator get me out of that bind? Here's a pic of my "ideal" janky layout. If it's a no-go, I'll just pull off the three oddballs and stick with the four identical ones.

Never mind... found some info on this. Thanks again to everyone.