Strange problems using SD Card and VS1003B based board

Using Arduino Audio
#1

Okay, so this is really weird, I could not find anything on the subject. I tried to fix it whole day, but it really gave me headache :frowning:

So the problem is basically this:

When I get close to the cables, it plays great. When I remove my hand, it stops :fearful: Also it stops when I connect my other hand to ground.

In general, it stops processing everything at all, not just the sound. It looks like the interrupts are stopped or something like this.

I’m using info based on this http://arduino-cool.blogspot.com.es/2012/10/arduino-mp3-player-i.html but I have adapted it for Arduino Mega 2560, as I’m using this in my project.

I’m attaching the schematic I’ve come up to, sorry for the bad quality, but in general, it’s the same as the project, I’m only using MC14050BP instead of voltage dividers, to bring levels to 3.3v. That’s the only difference.

These are my connected pins:

#define MP3_XCS 35
#define MP3_XDCS 39
#define MP3_DREQ 37
#define MP3_RESET 33
#define SD_SEL 31

#define SPI_MOSI 50
#define SPI_MISO 51
#define SPI_CLK 52

I’m using the same library AGMp3 as in the thread. I’ve tried other libraries, but the result was the same, so it’s not something in software. I tried modifying the SPI speed values inside the project a lot of times, without success.

I really have no idea what’s going on :frowning: Please, help me.

2014-04-13 21.44.30.jpg
2014-04-13 21.44.30.jpg2450×2146 306 KB

#2

Wow. Weird.
I was fixing my fathers old radio. It had a problem where only by touching the sound came out. I didn't understand what was going on until I dug out the datasheet of the ancient amp ic. It turned out the amp ic had 2 operational modes depending on whether 1 of the ic pins was grounded. When I touched the radio, the pin was grounded and the sound was output to the speaker circuit. The original contact to ground had been lost due to a dried out capacitor.

I have no clue what is wrong with your sound breakout, but I have had problems with cheap chinese modules where not all pins of the ic were properly tied to ground. The fact that your hand affects the sound output suggests to me that some pin or some ground that controls the playback is 'floating'. Just my 2c.

#3

I really could not find the problem. No one seems to have it. First, I replaced all cables with brand new. No success.

Then I started from scratch, even used voltage dividers instead of IC, but the result was exactly the same. Already suspecting faulty mp3 module, because I already checked SD module (have few of them) and they had no problems.

Right now I'm moving everything to universal pcb, without much cables and with connectors, so when my replacement mp3 module comes, I will check it.

Already thinking of doing one from scratch with VS1003/VS1053 IC by myself ... might be better :slight_smile:

#4

Problems where your hand proximity causes a change are always due to floating inputs. That is inputs that are connected to nothing. Inputs should be high or low not floating. If an input is not used it should be connected to the supply or ground not left floating.

#5

I really don't have such floating inputs, at least none I can tell of. How I can check them? Everything looks connected, I tested with multimeter and everything is connected ...

#6

The problem might not be that the wires that you are connecting are floating. I am suggesting that one of the "legs" of the IC that controls the VS1003B is not traced to ground = is floating. This is because sometimes breakout manufacturers are lazy/careless.

I had a similar problem once with a FTDI chip on a cheap Nano clone = I had to put a jumper connection from the tiny IC leg to ground to force it low.

I suggest you check the datasheet of VS1003B, see what pins are supposed to be pulled low/high. Test if pins are connected to ground / high or not.

#7

kkazakov:
I really don't have such floating inputs, at least none I can tell of. How I can check them? Everything looks connected, I tested with multimeter and everything is connected ...

Then test some more. Your problem is one of floating inputs period.
Just because you haven't found it yet does not mean you haven't got them.
Do not use a resistance meter on a powered circuit. Also it is not much use on an unpowered one most of the time.
Measure voltages both on the wires and on the pins of the device.

#8

Hi Smultron,

I’m curious - did you find out about this or dd you change your project to a proximity activated mp3 player :wink:

I’ve got one coming in the post next week…

Best,
G

#9