Core Electronics Forum

20W Solar Phone Charger

20W Solar Panel - Power Supply

Introduction:
The aim of this project was to use the sun to charge batteries.

Charging LiPo batteries
1W and 2W Seeed solar panels, DFRobot Solar LiPo Charger and a Sunflower Solar Power Manager were purchased. Initially this was unsuccessful because I did not realise the panels need full sun to produce full power. The LiPo was connected such that it was shaded by the panel but it still got too hot for my liking.

The solution was to place the solar panel in the sun connected via a 3 metre cable to the charger inside the house. All the items listed work perfectly for charging LiPo batteries.

Charging Phone, Tablet and Power bank devices
These devices charge via a 5V USB port and can use up to 2A or more. Investigation of my phone showed it would charge at 430mA from a computer USB port and 1.3A from the charger supplied with the phone. So minimum power would be 5V x 2A = 10W.

The rest of this document will describe the Solar Panel 5V USB charger.

Parts: (major items only)


12V 20W Solar Panel with Clips Jaycar CAT.NO: ZM9052
Monocrystalline
Open Circuit Voltage: 22V
Short Circuit Current: 1.21A
Voltage at Power Max: 17.4V
Current at Power Max: 1.16A
Dimensions: 435 x 356 x 25 (LxWxH)
Warrenty: 5 years, 20 years for power output no less than 80%

Adjustable Switching Power Supply Module IN 4V-35V OUT 1.5V-30V LM2596S SKU: CE05572


Input voltage: 4V to 35V DC
Output voltage: 1.23V to 30V
Output current: 3A max
Efficiency: 92%
Dimensions: 48 x 23 x 14 (LxWxH)

Design:

The 20W solar panel was chosen so it would provide enough charge when the sunlight conditions were not ideal; such as clouds or early morning / late afternoon. The panel would also be mounted outside on the house roof, so a long life robust design was needed.

The power supply module was chosen due to its flexibility and power output and the ability to finely adjust the output level using the 20 turn pot. Although it will only be operated at 5V, it can be set to different voltages in the future, if the situation changes. The large input range also allows for future changes as well.

Construction:

Solar Panel and mounting hardware.
The pics show the hardware that was purchased from Bunnings Warehouse and how it was put together. The end result is a fairly robust mount for the panel. The Antsig antenna mast worked very well for angling the panel to face to sun.

The original panel cable was too short to run from the roof to the garage so it was cut and a 2.1 power plug installed. The connecting cable is Automotive / Marine 7.5A power cable and is about 6 metres in length. If this was done again the cable would be soldered to the terminals inside the box on the panel. This would be more robust and reduce the chance of moisture ingress. The 2.1 plug and connecting cable was used for flexibility in testing.

5V USB Power Supply
The switching power supply module was mounted inside a small jiffy box. A USB female connector mounted on a small vero board inside the box and a 2.1 power plug socket installed at one end. A Raspberry Pi 5V fan attached at the other end to remove heat from the box. A few small vent holes were made at the plug end to allow air to flow across the module. A Charge Doctor shows the voltage and current conditions on the USB port. Two small heat sinks were mounted to the module screws for heat dissipation.

A point to note here is the 2 mounting screws on the module are not at the same potential. One is connected to Vin+ the other to Vout-. Shorting Vin+ to Vout- (GND) would not be good. The design of the module does not allow for a heat sink to be easily attached.

Testing:

Solar Panel.
The solar panel produced around 20V with no load and maximum current around 17V with an ideal load. (as per the specifications) 5W & 10W resistors of various values were used to assess the best operating conditions.

Overload condition.
If the panel is overloaded, the voltage drops to a low level and stays there even if the load conditions are changed. ie full sun, then cloud, then full sun. Removal of the load is required to bring the panel back to full voltage. This mainly occurred when the panel was operating close to its full load capacity for the conditions. If it was operating at half its full potential then it would recover correctly if the output dropped briefly for any reason.

The same effect was noticed when the panel was disconnected from the power supply but the load remained attached to the supply. Connecting the panel in this situation would cause it not to work. Connecting the panel first, then the load, it would work correctly. Possibly the startup current of the power supply was too great. Again it happened only when close to its full capacity, less than full and it would work ok.

Power supply heat issue.
Initially the box did not have a fan or heat sinks. Testing showed the mounting screws became very hot when providing the 1.3A to charge the phone for a few hours. Two small heat sinks were added to the mounting screws but did not change the heat output by very much. Adding the fan solved the heat problem. The heat sinks now get mildly warm and the overall temperature is at a good level.

Phone charge current issue.
Initially the supply would only charge the phone at 430mA but was capable of providing over 2A. Why ??

This proved to be because of the way the phone detects the difference between a PC USB port and a Phone charger plug pack. The charger simply connects the D+ and D- pins of the USB A connector inside the charger. The phone detects this connection and switches to a higher charge mode.

So, a jumper was installed on the D+ and D- pins on the vero board. The phone now charged at 1.3A and takes about 2 to 3 hours to charge. At 430mA the phone takes about 6 to 8 hours to charge.

The following is from a recommendation standard for phone charger plug packs.

TELECOMMUNICATION STANDARDIZATION SECTOR ITU-T L.1000 (03/2010)
Universal power adapter and charger solution for mobile terminals and other ICT devices

Ideal charging time.

Power is only available while the sun is shining. Testing has shown the usable power to charge the phone is from 9:00am to 3:00pm, at this time of year, where I am located. Most people use their phone during the day and recharge at night. But this is not an issue for me because of my situation. (retired)

Conclusion:
The solar charger has been used to charge a phone, a LiPo Christmas decoration and as a power supply for Arduino and Raspberry Pi projects. (the Christmas decoration has a USB charger installed)

It has worked successfully in all instances.

The savings in electricity to charge the phone is about $1 per charge, not a lot. It will take some time to recoup the cost of the project, but given the 20 year life span of the panel it should pay for itself many times over.

When using solar panels consideration should be given to the available sun light and location of the panel. A management device is recommended to ensure the maximum power is delivered from the panel for the current conditions. The amount of output can vary significantly during normal daylight hours. Even a small shadow can reduce the output by a significant amount.

This is fairly simply project; a solar panel connected to a power supply, adding some form of management device would be a future improvement.

Overall I have learnt a lot from this project about solar panels, the advantages and limitations.

Links:

4 Likes

Nice write up James :slight_smile: You should send it through to them https://core-electronics.com.au/create-project

2 Likes

Wow, that’s great !

4 Likes

Wow, that’s great ! The solar charger has been used to charge a phone, a LiPo Christmas decoration and as a power supply for Arduino and Raspberry Pi off grid projects. When using solar panels consideration should be given to the available sun light and location of the panel. A management device is recommended to ensure the maximum power is delivered from the panel for the current conditions. The amount of output can vary significantly during normal daylight hours. Even a small shadow can reduce the output by a significant amount.

3 Likes

Yep an MPPT is a good option too for optimal performance. Here’s really good, very effective, but very simple solar tracker :slight_smile:

2 Likes

This is what I found.
Afternoon sun caused the TV mast to shadow a small part of the panel drastically reducing its output. So it was moved to a different position unaffected by the TV mast shadow. The angle is optimal for midday sun, motorising it to track the sun would produce more power over the day.

A friend has solar panels installed and a Wifi app that shows the power produced over the day; best output is only a couple of hours in the middle of the day. But to make all the panels track the sun would be a huge cost increase, guess that is why they don’t do it for domestic solar arrays.

Cheers
Jim

Hi Jim,

Nice project! :partying_face:

Also, another much simpler method rather than integrating tracking to improve your performance may be to add a mirror or other reflective surface to capture sunlight from more angles. Depending on how you’ve got it mounted you may need to get a little creative with the positions for the mirrors, but it could be another potential addition :grinning_face_with_smiling_eyes:

(This is used mainly for heating up salts, low melting temperature metals and similar to power steam turbines, but the exact same idea could be integrated into a small scale project such as a remote power setup)

Thanks for sharing your project with the community!

1 Like

Hi Jim

I think this was actually done. I seem to remember a solar hot water system where each “collector” element was about 1metre or so long and consisted of a long parabolic (I assume) reflector with a black tube mounted along this which sat near the focal point of the reflector. These “collectors” were mounted as an array into a panel type structure. The panel was not rotated as a whole but each individual “collector” was rotated in the panel to track the sun. To me it seemed to be a great idea at the time but maybe it was consigned to the too hard basket. I don’t know any details of the rotating system or whether this type of thing is still available but it did exist.

Whilst in Townsville many years ago we witnessed an ABC segment being recorded for the “Inventors” show. Do you remember that one? These 2 guys had invented a solar tracking panel with no electrics or such but was operated pneumatically with air. Delightfully simple. Two air tubular “tanks”, one on each side of the panel painted matt black and exposed to the sun with a shading strip alongside each one. Each of these connected to a pneumatic ram sort of facing each other. The idea was that when the sun heated up one tank more than the other the rams rotated the panel until equilibrium reached which occurred when the panel was directly facing the sun. So simple.
The ABC did a follow up to these inventions to find out any results. This one was so sad. These guys could not find any backing in Australia at all to help develop this idea and possibly produce something. Not so the Americans. Thee were sent 2 first class return air rickets along with a cheque for $10,000 for “spending” money all just to come over to USA and “Talk” to them. I don’t know of any outcome to this but something similar may have been used in the water system above or used in some other completely unrelated project or maybe electronic development overtook it before it could get going. But it was sad when others were prepared to throw this sort of money around just to talk when no local interest was forthcoming. I am talking here of over 40 years ago when $10k was a considerable sum.

Cheers Bob

1 Like

Hi Robert,

Super interesting stuff in this post, you’d think with so much barren, sun-covered land we’d be at the forefront of solar panel technology, maybe that abundance of space means that it’s simply cheaper to place more panels than create a sun-tracking mechanism. I love the simplicity of a pneumatic system and I’ll definitely hunt for that New Inventors segment when I get the time.

-James