Are the analog inputs protected?
So if I connect a negative voltage or a voltage that’s higher than 3.3V , will it limit it (using internal diodes) or will it break my Spark?
I’m connecting an analog audio signal to it (just so know). Not for the audio, but for its amplitude.
To answer my own question, I think it is.
According to the microchip’s scheme on page 79, there’s voltage clipping: http://www.st.com/web/en/resource/technical/document/datasheet/CD00161566.pdf
But if someone can confirm this, that would be awesome.
Hi @SamDecrock
You are correct that like all modern ICs, there are protection diodes for ESD and some amount of over/under voltage protection, but you need to be careful.
The STM32F102 has 5V tolerant pins and and non-5V tolerant pins–the analog inputs are not 5V tolerant. There is a table on page 37 of that spec that shows that for the 5V tolerant pins, the max and min allowable input pin voltages are Vdd+4.0 and Vss-0.3, where Vdd is 3.3V for the core and Vss is 0V, that means -0.3 to 7.3V on those pin, but that just means that the chip will survive, not function at those levels. For the non-5V tolerant pins, the upper limit is 4V and the lower limit is the same -0.3V. The larger the over or under voltage you put on the inputs, the more current is flowing “backwards” through the device, until finally the diodes fail.
So if you drive your inputs too far above or below the range, the STM32F102 will be damaged and unfortunately it has already happened to one or two users here on the forum, so be careful.
If you want to give details of what you are trying to do, we can help you more.
OK, thanks.
Am I correct that if I place a 12k Ohm resistor in front of it, I’m able to go as far as 60 Volts (in both directions (thats -60V and +60V) ?
As on page 38 it states that it can handle up to 5mA. (60V = 5mA*12kOhm).
Or do I get it all wrong? 
I’m creating a sound detector. I’ve done al the coding using the MAX4466 but now I want to hook it up it to the XLR-output of a sound mixer. I don’t need the waves, just the amplitude. I’m using the “hot” wire of the balanced XLR-output.
Hi @SamDecrock
No, I would not try this. It might work for a while but you are asking for trouble.
If you have balanced XLR out, you should probably be looking for a step down transformer or better-yet an op-amp solution. If you only use one side of the balanced output you are asking for ground loops.
Why do you have 60V signals on there? Are you talking about phantom power for a microphone? That is easily blocked with a DC blocking capacitor. Or are you talking about a high-voltage speaker system like some buildings use mounted in the ceiling? That is a unusual signal level.
Here is a schematic to get you thinking:
That came from here–no affiliation it was just the first picture that looked right to me on google.
You really won’t get a good result from just measuring the hot output of a balanced line-out from a mixer.
I’d suggest a Jensen JT 10KB transformer to convert from balanced to single-ended in a reliable way.
It will also fully insulate you from phantom (which won’t be present on line-out anyway.)
http://www.jensen-transformers.com/datashts/10kbd.pdf
Alternately, I have had great experience with the INA134 from TI:
Thanks a lot guys, I’ll look into it.
The 60V was just a quick calculation of how far I could go. The output of the sound mixer goes no where near that. Well, that’s what I think. I’ve measured the output of the sound mixer using a regular (AC) multimeter and only got it to go up to 10 volts.
Line out from a mixer should only have a a low voltage signal ~1V RMS, when terminated correctly (think 600Ohm, 10K etc.)
Were you measuring between pins 2 and 3 of the XLR, or between pin 2 and ground (pin 1) ?
Even the high input impedance of the multimeter will produce misleading results (too high.)