MintyBoost 2.0 and Samsung Galaxy S

[Rednaxela]

So, today I got a MintyBoost 2.0 constructed (ignoring the data line pull-up resistors since I didn't think I'd need it), and tested it with my Samsung Galaxy S (aka GM-I9000M) phone.

Unfortunately I don't have a USB extension cable I'm willing to sacrifice at the moment, so I can't measure the USB current, but based on the current the MintyBoost was drawing from the AAs, it's a real hog it seems.

When I tried with two conventional AA batteries, I found it was drawing 1.5A (!!!!) from the batteries, making them get quite hot. Those batteries seemed to have roughly 0.8 ohms internal resistance put together, thus leading to a huge voltage drop to below 2V. The phone still charged and it increased 7% before I decided I wasn't fine with those batteries staying as hot as they were getting.

When I tried some rechargeable AA batteries I had, they didn't seem to have such a harsh voltage drop due to current, and didn't get so hot, but they were drained quite fast. They were only partially charged to start, but they ran mostly down before I thought of measuring the current from them, so I'll have to wait till they're charged again before measuring with them again.

Thoughts?

[adafruit_support_bill]

The current crop of smart-phones have very high capacity batteries. It would take several sets of AAs to completely charge most of them. Some folks have built up MintyBoosts with C cells for more capacity. Also, the lower voltage output from rechargeables (compared to alkalines) result in a somewhat lower efficiency for the boost converter. Using 3 rechargeable cells in series gets it back into the converter's sweet-spot. http://www.ladyada.net/make/mintyboost/power.html

[Rednaxela]

Yeah, I certainly was not expecting to get a full charge off a pair of AAs anyway.

What I did find interesting though was that the normal alkaline AAs simply couldn't handle the current draw. It makes sense though. At 1.5A of current the pair dropped in voltage from 2.8V to 1.56V, which simply wasn't enough for the MintyBoost to have 5V out (managed to get 4.36V out). If I assume 80% or lower efficiency (lower I'd think, in these conditions), that would mean it's at most outputting 430mA at 4.36V under these conditions. So, it's not as if the device is drawing more than 500mA at 5V, it's just that these normal alkaline AAs drop in voltage too much as the current demand goes up.

About using three rechargeable AAs, that would be good I'm sure, except that the AA charger I have only charges them in pairs, which makes anything using odd numbers of them awkward. So for an even number, if I were to use four rechargable AAs, while they're rated at 1.2V each, I find they seem to charge a little higher to more like 1.26V each in practice, which multiplied by four is slightly over 5V.

[adafruit_support_bill]

Yea, alkalines are not a good choice for high-current applications. You might be OK with the 4 rechargeables. They would total >5v fresh from the charger with no load. But I'd be willing to bet that with any significant load they would sag back to a safe voltage.

[Rednaxela]

Hmm, the rechargeables at 1.26V weren't as fresh as I had thought. I fully charged a set of four and they worked out to 1.39V each. 5.04V I would have trusted enough, but I'm wary of 5.56V. It's not not as if a high current load will always be plugged in, and it wouldn't draw such high current when it's charge gets full too. So... the 4 rechargeable AA idea is out the window I think unless I want to throw in a 5V regulator that would waste power.

Alright, I just tried with two fresh rechargable AAs, each starting at 1.39V, totaling to 2.78V. My phone is at 22% power. The current draw from the batteries is 2.10A (Wow!), and their combined voltage is holding at 2.47V (A drop of only 0.3V, not bad for 2.10A!). The output from the MintyBoost is unfortunately not 5V, more like 4.87V.

The low output voltage despite input remaining strong, plus the 5.2W draw on the batteries, would indicate to me that most likely my device is trying to draw notably more than 500mA. Either that or I'm getting 50% efficiency and the MintyBoost voltage is being sub-par for a different reason, but I suspect it's the device trying to draw more than 500mA.

EDIT:

Performance after 10 minutes is:
- Device battery is at 28% (6% increase, out of a 1500mAh battery in the device)
- Means roughly a 540mA charge rate for it's battery assuming the 6% is accurate, and of course it's drawing more than that from USB considering the device is still on while charging.
- AAs down to 2.33V with 2.20A current. AAs measure at 2.59V when removed.
- MintyBoost output at 4.76V

At this point I really do wish I had a sacrificial usb extension to use for measuring usb current, haha.

EDIT #2:

After 1 hour of charging:
- Device battery is at 48% (26% increase). So an average charge rate of 390mA for the device battery over the course of an hour, assuming the 26% is accurate. USB current still unknown of course.
- AAs down to 1.64V total with 1.38A current being drawn. AAs measure at 2.38V when removed.
- The internal resistance of the rechargeable AAs went up significantly as they got low.
- MintyBoost output at 4.39V

So in conclusion... I don't know how much current the Samsung Galaxy S tries to draw, but seems evident that it's more than 500mA. I really should measure the actual USB current some time.

[Rednaxela]

I ended up giving 4 of the rechargeable AAs in series a try. With that it seemed to work pretty well. The results are as follows:

Before plugging in:
5.50V in at 0.00A
5.38V out

after 0 minutes:
5.38V in at 0.57A
5.10V out
device battery: 28% (initial)

after 66 minutes
4.90V in at 0.61A
5.06V out
device battery: 56% (28% increase over 66 minutes, 381mA average)

after 128 minutes
4.75V in at 0.63A
5.05V out
device battery: 84% (56% increase over 128 minutes, 393mA average)

after 180 minutes
4.77V in at 0.24A
5.10V out
device battery 100% (72% increase over 180 minutes, 360mA average)

after unplugging:
4.88V in at 0.00A
5.01V out


Next step? I plan to compare the result if I use 2 pairs of AAs in parallel. Going to set up a 4 AA holder to be switchable between parallel pairs and series.