Re: An other ebay mystery...

On a sunny day (Thu, 19 Feb 2015 13:52:31 GMT) it happened Jan Panteltje wrote in :

Anyays, decided to open it up. There is no mike behind the hole marked 'mic'. The chip is a narrow DIL with few pins marked STA801. A websearch for STA801 confirms that that is NOT this chip. The board is marked Y02-DS801-V2 No find either but there exists a similar DS801 mp3 player. The chip date is week 39 2014, so maybe too recent... The other thing on the chip says MQGX269.1 (no find either).

Interesting it becomes when we read on an OTHER sellers site:

Quote: "Built-in Li-ion battery, directly connect with PC for re-charging Note:Do not charge more than two hours, or it easily create MP3 Built-in lithium battery failure and damage "

It does not specify if the damage will be to the battery, the unit, or your house or city,

So that explains the price I guess!

But I have a green one, so...

I am now charging, need to remember to watch that clock. Dunno how much was in it when started...

;-)

So.. I charged the battery for > 3 hours, in the beginning it was a bit warm, but before and after that I never got more than 1 hours 10 minutes play time out of it, with no load (into a high impedance amplifier input).

No bangs or booms or fire. But not the promised (on some sites) 3 hours playtime either.

There is an other model like it witout LCD where they promise 4 hours play time, that one is < 2 dollars:

I ordered some :-) 3$58 for two, inclusive shipping. will make a DC jack on it.

This is the follow up, I REALLY like to get to the bottom of things. Now THIS is well, you name it,.

Could not resist opening it again, take some pictures, and especially measure battery voltage and current, both in play and in charge.

In play it consumes about 17 mA, yes on a 90 mAh lipo that should give

4 hours...

But, after 3 hours charging battery was not full (as indicated by battery symbol, and short play time, and now with digital voltmeter: 3.9 V (4.2 and a lipo is full)..

So, I had a suspicion, it came with the thinnest USB cable I have ever seen.

So measured the charging current (meter in series with battery all the time).

4 mA!!! Well that will take years to get full.

Changed the USB cable for a slightly thicker other Chinese one.

12 mA.

OK, now I had 2 USB cables in series all the time (extention from PC), so plugged the thicker cable in the PC directly:

20 mA

So the Chinese count on the resistance of the supplied USB cable to limit the charging current. mm :-)

The cable from my Canon camera gives 22.4 mA charge current, getting better.

Connected a 5V Raspberry power supply:

31.4 mA

Now that should give about 3 to 4 hours charging time, but but, is there any regulator????

The LCD back light itself uses about 3 mA (set to maximum contrast, should be 'brightness' actually).

Other side PCB, battery mounted with double sided tape, as is the LCD.

This is the only chip I can see:

But there must a LCD controller between layers of sticky tape on the LCD, but too tricky to remove that, as it may tare up the thin flex PCB cable.

Now charging from Raspi power supply with Cannon USB cable.

Wheels of fire, rolling down the road, you know that we will meet again... if you FLASH memory serves you well, or something like that,

DISCLAIMER DO Not trY tHiS aT hOme

Maybe stick the batteries in a metal box while charging. (Blast and fire protection)

George H.

If there is a charge indicator on the LCD there is a charge monitoring circuit. Period.

Nope, plays those pieces at double speed, could be the Samsung chip!

I think the issue with that chip is that it only checks the _first_header in a mp3 stream, and then keeps that as reference for the rest. Of course it should process settings from any subsequent header.

But it _is_ a bit out of the ordinary, so I am not complaining. But it was one of the reasons I wanted to try this one...

The Creative MUVO is falling apart, and that model is becoming strangely expensive on ebay, 189 $!!!

Not what I payed, was new about 30 or 40 Euro IIRC). Will have to re-encode the mp3. Not a big deal.

Playing >4 hours now....

YIKES! Make sure your fire insurance is paid up!

Jon

...

Yup. The apparently "classic" USB LiPo charging circuit for portable players, toys and similar "mass market" goods is a (silicon) diode with a 10 Ohm (SMD 1206 size) resistor in series. Some devices also have a comparator in SOT23-5 case driving a charge indicator LED.

5 V - 0.7 V conveniently happens to be "close enough" to 4.2 V, the upper limit charge voltage for a LiPo cell, and the resistor limits the current into a discharged cell. There's no other regulation, limiting or shutdown. The comparator (if it's even there) only drives the LED (or an MCU pin if the device has a display).

In your case, the cheesy cable and connectors probably had a significant fraction of those 10 Ohms.

The battery itself may or may not have a protection chip. If it has one, you may notice that because sometimes it will latch when left charging too long any you'll have to leave it disconnected for a little before the voltage drops below the threshold, the protection resets and the thing starts working again.

As for safety, I'll leave that evaluation up to you. Particularly if you happen to have an iSomething or other "USB" power supply (genuine or not) that intentionally hits the upper limit of the USB allowed voltage range and "does its best to accelerate charging" with a 5.2 V barely- standard-compliant output.

Greetings Dimitrij

P.S. I don't have the same type of mp3 player as you, so i don't know for sure whether the circuit described above is indeed used there. It seems to be common enough to be a "usual suspect" however ...

On a sunny day (Mon, 23 Feb 2015 00:06:19 +0100) it happened Dimitrij Klingbeil wrote in :

Yup, in the picture you can see the diode,

there is place for 2 diodes, only one is fitted, and indeed it ended at 4.3 V... after a while. The lipos with protecion have a chip visible through the plastic, in this battery there is no such thing visible, and there was stil 5 mA flowing at 4.3 V.

But I am happy with this player, re-encoded my multi-bitrate mp3s yesterday, linux ffmpeg does it right! Those now play OK too. (A 7 hours mp3 imagine, never reaches the end). Need to make some power adaptor.

It kept bugging me that this little player is uncontrollable as far a charging goes. So decided to modify it. After all this is a design group...

So still had some Microchip MCP73831T -2ACI (4.2V) lipo charger chips,

I mean 10 for 1$60, I payed 3$60 for 10 last year... Anyways, this is what I did, left the (it was 6.8 Ohm) series resistor in, removed the series diode, added some tantalum 2.2 uF caps (no they draw no measurable current):

Scratched some solder mask away and mounted the stuff:

And looked up that MCP73831 programming resistor, decided to settle for 22k should give about 30 mA in the datasheet.

So there you are:

It is now chargibg, BUT WILL IT STOP???? SUSPENSE!!! More in the next episode of LOL

PS, had to remove the second tantaal cap over the battery, as leaving it in made it go to 'PC connect mode' every time. Dunno why. The Microchip datasheet goes into the value of this decoupling cap in a long story and recommends (states) that it should be at least 4.7uF.

Well, I cheched with the scope without the cap and I see no instabilities within the scope's range of about up to 30 MHz. Neither does it interfere with FM radio...

So some suspense remains.