Noise (due to leds?) in MB v2 with Bravo firmware

Home Discussions Your Build Noise (due to leds?) in MB v2 with Bravo firmware

This topic contains 3 replies, has 2 voices, and was last updated by  moshang 1 year ago.

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • #3745

    garretlabs
    Participant

    Good morning from Italy dear Moshang!
    I’ Marco, the technical responsible of Garretlabs, the italian blog of open source *ware ideas (http://garretlabs.wordpress.com). Your MB project is really great!

    I built a Groovesizer MB v2 with Bravo firmware (with the jumper in the right position, I think), and all seem to work well… except for the fact that when leds are blinking (or a I use pots which power on/off the leds in order to view the values for the parameters) a noise appears on my audio channel, which is seems tied to poower on status of leds.
    And when the number of powered-on leds increases, also the noise increases.
    Plus…when the sequencer is ruinng during the steps 17-32 it’s possible to hear a very low noise.

    Could it be tied to the fact I used leds with forward current of 20mA (they are http://www.tme.eu/it/details/l-934srd_f/diodi-led-tht-3mm/kingbright-electronic/l-934srd-f/) instead of your suggested mouser leds (604-WP710A10ID5V, which have direct current of 13mA)?

    Or do you nthink there is some other tricky problem?

    Thank you very much for your suggestion… and thank you again for your very beautiful project!
    Marco

    PS: I saw that another forum user (see http://groovesizer.com/Discussions/topic/background-noise/) had the same problem some time ago, but I don’t use the 386 amp and I have a v2 board.

    #3746

    moshang
    Keymaster

    Hey there Marco – apologies for what will be a very unsatisfactory answer, but unfortunately my electronics knowledge is severely lacking. All I can really say is that I haven’t noticed the problem you mention on one of my own builds yet. The problem of circuit noise leaking into the audio channel pretty much disappeared when the decoupling capacitors (104) were added to all the ICs on the V2 version of the board, but even on the V1 board it was only really a problem when using the 386 amplifier IC. I’m also not sure what impact the forward current of the LEDs might have here.

    I do remember a user having problem with noise on the output of the TB2 that turned out to be related to the audio cable he was using to connect to his amp. Can you try different configurations (using the small jack vs the big jack – headphones vs different kinds of amplification) and see if that makes a difference?

    #3747

    garretlabs
    Participant

    Thank you very much for your answer dear Moshang!
    My amplification setup is very simple: I use the DAC of MB, without 386, and I attached the MB to my home stereo or to my amplified mixer (line in) using small jack or big jack. Both cables are stereo cables (but think it is ok, since I noticed from the PCB/schematic that you duplicate the signal on both channels of the output in the small jack output).
    I tested the cables and they are ok.
    Using earphones directly connected to mini jack out the noise seems not present…but all the volume in the earphones is very low.
    I tried to modify the noise level during the steps 17-18 changing the philosophy of updating led rows in the Arduino code (using three sub-sections instead two):

    if(section==1)
    {
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 5
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 4
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 3
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, LEDrow[1]); // row 2
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, LEDrow[0]); // row 1
    section=2;
    }
    else if (section==2)
    {
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 5
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, LEDrow[3]); // row 4
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, LEDrow[2]); // row 3
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 2
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 1
    section=3;
    }
    else if (section==3)
    {
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, LEDrow[4]); // row 5
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 4
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 3
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 2
    shiftOut(LEDdataPin, LEDclockPin, LSBFIRST, B00000000); // row 1
    section=1;
    }

    It seems that in this way the noise during steps 17-32 disappear.
    Very strange… can it be that CPU doesn’t drive correctly these leds?
    In fact, if I disable in the code the leds update… the sound is very good without any noise (but I don’t know which are the active steps/functions etc.)
    …Now I will try to use different leds (the same you used) or with less power consumption.

    I will update you with my results.
    Thank you again,
    Marco

    #3748

    moshang
    Keymaster

    Hmm, it looks reasonable that it might be related to power consumption then. Incidentally, I had to do the output to the LEDs in stages like this, because there’s no enough power to light them all at once. It starves the ATMEGA and causes a crash if you light them all at once.

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.