Solar USB Charger

We decided to start this project following a visit to the Goodwood Festival of Speed. We started the day with a fully charged iPhone 4 and it was used constantly, to stream updates to Facebook and Twitter (with location) over a very poor 3G network. The battery lasted little more than four hours :-( This led us to look at ways of extending battery life, when there are no mains power sockets to hand.

Throughout this very sunny day, as we watched the battery rapidly lose power, we were thinking that we could be harnessing the power of sun to keep our iPhone 4 going. There were plenty of opportunities to charge the iPhone, whilst standing trackside. So, the objective of this project is to address this and to build a generic solar power source that can be used to power and recharge a wide range of devices, including phones and tablets.

Design

The iPhone 4 can be charged using a standard USB port that outputs 500mA but the supplied charger and high power ports can also charge the iPhone 4 faster by supplying up to 1A current. In order to achieve this USB port needs to have a special configuration of resistors to 'signal' to the iPhone that it can draw this much power. In a similar way the Apple iPad and iPad 2 can also fast charge, drawing 2.1A from a high-power USB port or plug-in charger.

An iPad would require ~11W to charge at its fastest rate. This is quite a lot to get from a relatively small solar panel, so our design initially uses a single 6V 5.2W solar panel. This should provide enough current to fast charge an iPhone 4 and to also charge an iPad. The design could be extended in the future to use two solar panels and this would then just about fast-charge an iPad assuming the sunlight was consistently bright enough.

solar panel
Solar panels provide lower voltages as the current drawn from them increases, so it is important to use a panel that can deliver a decent current and voltage. The one we have chosen is not cheap but is capable of supplying >550mA at 6V output. It is a fully enclosed solar cell, rated at >8V open voltage and >650mA short circuit. It has a clear epoxy coating with hard-board backing, which means it is quite tough and could be fixed to something like a rucksack using Velcro.

solar panel
These panels are custom made for Sparkfun and can be purchased in the UK from ProtoPic. They are not cheap but, they are much higher power output than those used in most portable solar chargers, such as the PowerMonkey device.

Voltage Rectifier

The panel output obviously varies as the intensity of sun varies and also with the load present. You can't just connect the panel to a device like the iPhone 4 and have to use a voltage regulator to keep the supply voltage at a steady 5Vdc. It is also important that the voltage drop introduced by the voltage regulator as low as possible. Fortunately, you can get 'Low Drop Out' or LDO voltage regulators, which introduce a much lower drop into the circuit.

regulator circuit
For the 5V rectifier, we are using a Micrel Semiconductor MIC29300-5.0WT (datasheet). This is a 3A fixed voltage (+5Vdc) with a very low drop-out voltage. These were purchased from Farnell.

Note: This regulator needs a minimum load of 10mA for the voltage regulator to work properly.

USB Port Wiring

We plan to wire up the USB power output port with the required resistors, to enable iPhones and other devices to draw up to 1A from it. It doesn't make sense to use the iPad configuration (2.1A) as this solar panel simply can't supply that kind of power. We may change this if we decide to add a second solar panel.

Xmultiple dongle
You can buy USB dongles with the required resistors to fool an iPad into thinking it can draw the required power from USB ports that don't have the required resistors. This is the Xmultiple dongle. It is basically a straight through USB to USB connector with the required resistor array.

Progress

Testing the solar panel shows that in bright sunlight it provide over 10Vdc on no load but more typical no-load voltage is around 9Vdc. This obviously drops as higher loads are added. The voltage drop from the LDO regulator is only 0.35V so it should be able to provide some charge when the sun is behind the clouds and also through winter, when the sun is not as strong.

USB test lead
I order to measure the current being supplied or consumed through each USB port, we have taken a short USB extension lead and cut into the power line to allow us to connect a digital ammeter in line. This allows us to see how much current is being provided at 5Vdc and not just the raw output of the solar cell.

This does have limited value when it comes to iPhones and iPads though, as the ammeter changes the resistance seen by these devices, causes them to see the ACD66 (for example) as an 'unsupported device' for charging.

We put some resistors directly across the panel and measured the voltage and current output:

  • 820Ω = 8.9V and 11mA in partial cloud
  • 430Ω = 8.7V and 20mA in full cloud
  • 270Ω = 8.9V and 34mA in partial cloud
  • 100Ω = 9.7V and 140mA in sun
  • 47Ω = 9.5V and 200mA in sun
  • 39Ω = 9.3V and 230mA in sun
  • 18Ω = 8.6V and 400mA in sun
  • 12Ω = 2.45V and 180mA in cloud
  • 12Ω = 8.0V and 670mA in full sun, and burnt out the resistor

So we have a decent benchmark of what this solar panel can do. In full sun it should be able to supply over 650mA at 5Vdc through a USB port. If you wanted to charge an iPad directly, then we would need two of these panels.

Charging Tests

Vero board with USB port
This is a test board we have built with the voltage regulator, USB port and resistor network to convince an iPhone that it can draw 1A from this USB port.

Having built this board we tested it to ensure everything was working as planned. This led to the discovery that the voltage regulator output was simply tracking the input voltage :-( Having carefully re-read the datasheet, we realised that this voltage regulator needs a minimum load of 10mA, something we had missed before. To avoid an over-voltage situation and to make sure all devices are not subject to voltages higher than 5Vdc, we decided to put a 560Ω resistor across the output. This is slightly annoying in that 10mA is now wasted heating up a resistor but in the scheme of things, this is quite a small amount and protecting the devices from over-voltages is more important.

The first charging test we did was with an iPod Touch (3rd generation). Even with a cloudy sky, the device would happily charge through the above circuit. We did notice that the voltage regulator was starting to get warm, so we are going to add a heatsink. We could also charge an older Android phone (HTC Hero) with full cloud cover and heavy rain falling.

Like our iPod Touch, our ACD66 is also very happy to charge from the above circuit and would do so even with a cloudy sky. This is very good news, as this is our preferred route to keep our device going using the power of sunlight alone.

Resistors?

The next set of tests plan to determine whether we should be using the resistors to signal to the iPhone 4 that it can draw 1A from the USB port. We are currently testing without the resistors in place to see if this works better. In real-world use, it unlikely that the solar panel will supply much more than 500mA anyway.

Further tests show that the iPhone 4 is also happy charging from this our circuit if we put the required resistors in but, if we put a current meter in-line or don't have the resistors in circuit, it is not happy and displays a 'charging not supported from this device' warning and will no longer charge.

So the final decision is to go with the resistor network in place.

Enclosure

Enclosure front
We have redesigned the enclosure slightly and the voltage rectifier is not mounted to thin metal case that is fixed to the back of the solar panel. This makes it more compact and the case acts as a large heatsink for the rectifier. This means that the panel can now be mounted to the outside of a bag or rucksack and it now has a flying USB lead that can connect to devices inside the bag.

Enclosure back
This aluminum enclosure is only 5mm deep and also acts a heatsink for the voltage regulator IC. Despite being used all day, it barely got warm.

Summary

We are still writing this bit up as this is a project still in progress ...

The use of a standard type-A USB port makes it very easy to connect different devices to our charger. As well as charging devices directly, it can also be used to charge a battery charging pack such as our ARCADIA iMILCH ACD66 External Battery Pack.

As part of this project, we have decided to see if we can run our iPhone 4 (soon to become an iPhone 5) on solar power alone for a whole year. Will will be using this solar charger to directly charge our iPhone and to also charge our ARCADIA iMILCH ACD66 External Battery Pack, which will then also be used to charge our iPhone. The ACD66 will in effect be a solar energy capture and storage device whilst the iPhone is being used elsewhere. The only time the iPhone will be plugged into other power is whilst it is synchronised to our laptops and this will be for very short periods of time.

Come back soon for progress updates.

Share ...
We are on ...
Facebook Twitter
YouTube Flickr Follow us on Pinterest
Buy It Now