There are a lot of these “ES500” phone solar chargers on Ebay at the moment. Prices start from about $ AUD 11 delivered upwards. Search for solar phone charger or solar power bank.
The battery capacity claims vary depending on the seller and range between 5000mAh and 20,000mAh. Of course the upper end would have to be a gross exaggeration. I ordered one to keep my phone alive while camping next month. Expectations were not that high at that price!
It arrived a few days later and some tests were done on it to see how true the claimed capacity etc. was.
It seemed to be functional delivering nearly one amp to charge my Samsung S4 phone. So time to do some measurements. The little solar panel on it is supposed to deliver 200mA @ 5V which did seem a bit hopeful considering the size of it.
Repurposed meter, the bottom scale is mA x 100
A 0.5 Ohm current shunt was put on the negative side of the battery and the unit placed in full summer sun. A maximum solar charge current of 130mA was measured. Not bad I guess for the size of it, but not the 200mA stated. After a few minutes, however, the current abruptly dropped to about 70mA. The drop was thermally related and was traced to a faulty solar cell on the panel that seemed to be going short circuit. The faulty cell was isolated by running an ice block over the cells one by one.
Next thing is battery capacity. The unit was cycled twice and then fully charged. The Raspberry Pi with a WiFi adaptor was used for a discharge load. The Pi consumes about 350mA with the configuration it was in. The current was checked with a USB voltage/current monitor. After 5 hours, the solar charger was flat.
This equates to about:
0.35 Amp x 5
= 1750 mAh at 5 Volts.
To be fair, a voltage boost converter in the unit steps up the battery from about 3.6 Volts to 5 Volts. Assuming that the converter has a 90% efficiency and that the source is average 3.6 Volts, the cell capacity then becomes:
(1750 / 90%) / (5/3.6)
= 2700 mAh at the cell.
So, the battery capacity is a bit more than half the claimed capacity and a lot less than what some of the others are claiming.
The controller chip in the unit looks like a clone of this device from Richtek. It has the same four led status indicator, torch function and I/O.
An Ebay sourced switched mode boost converter from 5V to 12V was connected to run my 40m QRP transceiver. 
It ran quite well in receive, but the transmit would have to be limited to one or two watts.
The little solar panel on the charger was not quite sustaining the 180mA 5V load, so the battery was supplying the rest.
The boost converter must be rather inefficient as the load at 12V is 56mA. (~ 75% efficiency). 
This is not a very efficient way to run the radio as there are step up losses in charger’s boost converter, further losses in the external step up converter and even more losses in the linear regulators in the transceiver that step down to 5V again!
Update: A bit disappointed by the performance of the unit, especially on solar charge, I grafted a larger solar panel on the unit. The panel provides 400mA of charge and makes a the unit a bit heavier, but it works just fine! The panel was fixed to the unit with silicone adhesive.
Another identical solar charger was tested and the charge output was 60mA. The original solar panel consists of two series strings of solar cells in series. These strings are connected in parallel. By shading alternate halves of the panel, it was obvious that only one string of cells was working. This is also the problem with the first unit, although it does work briefly until it heats up. Maybe the cheaper units on Ebay are QA rejected units?
Some experiments powering the MTRV2 from the USB powerbank and DC-DC step up convertor:
Single cell 18650 powerbank running the MTRV2. Note receive current is 70mA at 5 V and transmit (800mW) is 1.47 A at 5 V. The DC convertor is set to 8V output. The powerbank simply can’t deliver any more current at 5 V. The same test again below with a higher power USB powerbank.
This time the output of the convertor is set to 10 V. This powerbank can supply more current at 5 V and the DC current at 5V is now 1.64 A and the RF output is 1.5W. It can’t supply much more current though. The 5 V output isn’t dropping as much as the smaller powerbank.
VK3YY 
Like your technique of sliding an ice cube over temperature sensitive faulty solar cells! That sounds like something you learned from an old timer. So the bottom line is the product is a workable solar charged battery for a smartphone but not more. Have you looked at any other solar panel/regulators to charge your QRP radio’s battery whilst camping?
Hi Paul, the icecube idea came from a lack of freeze stuff in a can! The charger is sort of useable for a couple of phone charges with the measured capacity, shame about the faulty solar cell though, hardly worth returning! I have a few solar panels that do charge my phone directly- https://vk3yy.wordpress.com/2013/03/22/solar-charger-for-mobile-phone/ . Also have a nice little solar phone charger from Bunnings (Eiger brand) that has a built in torch and panel light. It charges the internal battery at about 90ma though, so it would take over 16 hours of sun to charge the internal 1500mAh battery! Have used this on a couple of camping trips successfully for a light and a phone boost charge. The same unit is sold by National Geographic under their brand for about $100. http://www.amazon.co.uk/National-Geographic-4-in-1-Solar-Charger/dp/B007WRY8KM . It was only $15 at Bunnings!
Hi Glenn,
I think you’re onto something when you suggest that these are QA rejects. It doesn’t surprise me at all. There are lots of LED arrays being sold on eBay at the moment that are also rejects. If you power them up with a variable supply, you notice that the leds fire up at different voltage levels and in some cases, not all of the individual cells in an array light. When you turn them on by applying the rated voltage, they appear to be lit, but they are not outputting their rated intensity.
I also came across a battery scam a few years ago. Chinese sellers are selling grossly overspec’d batteries which clearly were never going to be even close to what they were supposedly rated for … check out my findings at http://www.radcomms.net/gallery/thumbnails.php?album=9
Regards,
Ben VK3KBC.
Hi Ben, OK on your experience with suspected QA rejects as well. It certainly makes it hard to sort the wheat from the chaff. I also had experience with Ebay lithium cells that were grossly over stated in capacity! Cheers, Glenn.