MP3 Player Hack

[kam]

Hey all,

I have a difficult one (I think).

I want to hack into an MP3 player using its switch interface to control it.

basically i got a normal MP3 player from an electronics store to open up and see if i can do something with it.

It runs of a 1.5V AAA batt.

after opening it and checking voltages from my multimeter. I found out that Side A of the power button it gives 1.4V and on Side B it gives 0V. after turning on, it gives 3.3V on Side A. Side B is 0V again after letting go of the switch (obviously)

thing to notice here is that it using a step up. Why? probably for ttl logic, i guess.

So i think that would mean that side B is being pulled high, right?

if thats true, then the rest of the switches are being pulled low. cuz on one side they have 3.3V and on the other side they have 0V and after pressing the switch, Side A falls down to 0V.

Now i was using a LM1117 which is a 3.3V regulator. and to supply the MP3 player with 1.5V on its terminals i used a 10K Voltage divider to get 1.5V.

Problems:

1) The Mp3 wont turn on with that setup. after testing the voltage, it turns out that as soon as i press the power button, the voltage drops down to 0.9V. thats not enough to drive the player. maybe the voltage divider should be of a lesser value, but then i noticed that the resistors got hot.

2) After putting in the 1.5V batt and keeping the GND common,i turned on the player and then i tried the 3.3V wire coming out of the regulator, and it wont switch it high or low.

What am i doing wrong???

i cant post any pictures yet (cuz i got nothing) well i just started. but i hope that i can get some response from what i've just told.

Thanks

Kam

 

[hippy]

You have three separate problems -

1) Powering the unit. Using a voltage divider can be problematic, especially trying to use 10K R's. You would be much better off using a variable-voltage regulator configured to give the desired 1V5 or you could try using the 3V3 with diodes in series with the supply-out to drop that to 1V5. You may also be able to use an R plus Zener diode configuration.

2) Forcing button pushes other than power-on. If the buttons are commoned to 0V then they are probably active-low, short to 0V, with a pull-up to internal +V. Putting an external 3V3 to those connections may not be a good idea and you would need to short them to 0V to emulate a button push anyway.

3) Forcing power-on button push. This may again be the same as the other buttons, active-low, short to 0V, but the circuit may be more complicated than that, trying to inject 3V3 may damage the MP3 player.

To determine if the buttons are short to 0V. Disconnect the battery and as much as possible ( remove MMC/SD Cards etc ). Measure the resistance between 0V to both sides of the switch. If they are short to 0V, one side should be close to zero-R, the other should be much higher but drop close to zero-R when the button is pushed. Check all switches to make sure they are the same. Set the meter to a low R range to hopefully limit the current the meter puts out.

All suggestions are only that, they may work, may not, may cause damage, may not. Any tinkering with electronic assemblies without a circuit diagram has to be done at your own risk. It is possible to damage things and you have to accept responsibility if that happens.

 

[kam]

Thanks Hippy!

I'll look into it.... after i repair the wire that i snapped out while testing just now.... and confession, i did damage a previous mp3 (same model) by accidentally injecting 5V into the system
ah well we learn, or do we

 

[BeanieBots]

I have great concern that you say one of your 10k potential divider resitors got hot. I think you might be mistaken. It would take about 50 (fifty) volts to get a 10k resistor more than a few degrees above ambient

It is very likely that your device does indeed use a boosting regulator to increase the internal voltage and it is also very likely that the push-buttons control a matrix and do not connect directly to either supply or 0v. It is also quite possible that there is an internal voltage which is below the battery 0v reference. Only a good thorough browse around with, ideally a 'scope but at least a DMM will give any clues.

If the keys are indeed of a matrix type, then you will NEVER be able to control them with a simple high/low type signal. The simplest solution would be relay contacts or for a non-mechanical method use an analogue switch such as a 4066.

To test if the switches are a scanned matrix without a 'scope, put your DMM on AC volts. If there is any AC voltage, then they are scanned and cannot be driven with simple high/low signals.

 

[kam]

Hmm, analog switch.... havent worked on that before but by the looks of it, doesnt it seems like a normal transistor?

Ok, say i was to use the switch 4066. how am i to connect it, i guess i would bypass the mechanical switch with the ins and outs and then switch it with a high signal from the PICAXE. right?

Maybe that would be an easier way of doing this. gonna give it a shot, seems like fun.

About the resistors... i was surprised too, maybe cuz they were low watt ones, but i'm not sure, i'll get some high watt ones in case, and if that doesnt work, Diode Chain!

 

[BeanieBots]

An analogue is MUCH more than a simple transistor switch.
Yes, replace the two switch terminals with in/out.

Higher wattage resistors????? DO THE SUMS!!!!!!!
(1.5V across 10k = 0.000225W. Even the samllest resistor will cope with 0.2W)

Use a diode chain to drop the excess voltage.