Usb audio constant noise

I’ve had the pro set for the synth kit for a while but I’ve never used the usb audio until today. It basically sends a constant saturated high pitch noise even when no synth bits are connected. As you can see on the attached the input is constantly on red. The screen grab was taken with just the usb bit connected to the laptop, nothing else. This also happens when I connect other synth bits to it. In this condition it’s completely unusable. Is it faulty? Is there a solution?

I have the same issue.

The shape you see in the picture is constant and I can only get rid of it using a high pass filter. It seems to be the driver itself rather than the hardware.

Since I don’t have this USB module, can you tell me the frequencie of this noise ?
It seems to me that it could be the sampling frequencie :open_mouth:

I’ll have to test this out this evening. What I can say right now is that for me at least, I can remove the lowest frequency that FL Studio’s PEQ2 EQ high pass filter will let me set to the most steep curve and it seems to get rid of it completely. In other words, it’s an extremely low frequency.

This is FL Studio’s spectrum analyzer:

^That yellow line is the frequency. It’s at the absolute bottom of its range which when I hover the mouse over it says it’s 2Hz, but it might be even lower than that.

Thoughts?

And this is what the driver looks like outside of my DAW:

Hi,
Strange indeed.
But when you save your recoded audio as a wav file, is it still there ?
The noise falls below human hearing so you shouldn’t be able to hear this.
Could you upload a small recording with and without music ?
It could be possible that this littlebit USB module is defective.

@JackANDJude , @alexpikkert and other members of this forum, is it possible to check if you encounter the same issue ?
Greetings

Hi,
Strange indeed.
But when you save your recoded audio as a wav file, is it still there ?
The noise falls below human hearing so you shouldn’t be able to hear this.
Could you upload a small recording with and without music ?
It could be possible that this littlebit USB module is defective.

It does indeed persist when I save the recorded audio. I’m not able to hear it, but it does effect the potential dynamic range of whatever I record (unless I highpass it). I may not have a chance to post it until after the weekend, but I’ll upload an audio example soon.

Hi all,

Thanks for the thread! We would like to communicate that first of all, you’re not crazy. Sometimes there may be some noise created when using the USB I/O module. However, this can be minimized.

There are a few ways to replicate the noise you are seeing or hearing with the USB I/O module. Here are two cases:

=====================
Case 1:
Audio from your computer -> USB cable -> USB I/O bit -> synth speaker bit

It is normal for there to be noise at the output of the USB IO bit, but this noise is very small.
However, the amplitude of the noise is always constant. You cannot “turn down the volume” on this.

The best way to minimize the noise component is to ensure that the volume levels in the computer are high.

For example, on my PC, using a software sythesizer, I have three places I can adjust the volume:

  1. In my software synth
  2. In the sound settings of my computer (the soundcard driver)
  3. The volume knob on the synth speaker bit.

For this setup, I turn the volume as high as I can without clipping for #1 (the software synth),
then do the same for #2. After that, I usually don’t have to turn up the volume too high on the synth speaker.

This should minimize noise when injecting audio into the littleBits system.

=====================
Case 2:
Audio from littleBits -> USBIO bit -> USB cable -> computer software application

The first thing to note of importance here is that the USB IO module communicates BOTH DC and AC information…
Remembering that the littleBits system voltage ranges from 0 to 5 volts, this means that with this setup:

power bit -> USBIO bit -> computer

your computer is receiving FULL SCALE DC values.

Depending on the software you are using, this could pin the meters even though there is very little audio coming through. And again, depending on your software, an “audio level algorithm” could get confused by this, and possible turn up the gain so that you hear noise.

Instead, use this setup to test:

power bit -> oscillator -> USB I/O module -> computer

This should send a full scale square or sawtooth wave to your software to test.

++++++++++++++++++++++++++++
I tested case 2 on my system as follows:

power bit -> oscillator -> USB I/O module -> Audacity on my PC.

This gives me a full scale square wave into Audacity

power bit -> dimmer (or you can use mix too) -> USB I/O module -> Audacity on my PC.

This is me sending DC levels to Audacity. Audacity will record these, but its meters get confused: they range from 50% to 100%, with the minimum when I have 2.5 volts set at the input to the USB I/O module. This is because Audacity is not expecting DC values, so it computes the display improperly. But it still records it correctly.

There is a good reason why we send both DC and AC values through the USB I/O. With “regular” audio inputs, anything below 20Hz is removed, but control voltages, and modulation waveforms are very often lower than 20Hz. So this way, you can use the USB I/O module for both audio and control purposes.

1 Like

theodore_littlebits,

I appreciate the suggestions. They were very helpful in illustrating the concept of the constantly running DC signal. I went through each of the three suggestions and I think my findings are in line with what you’re describing.

  1. I hadn’t messed around much with sending things from my computer to the littlebits, but I was not surprised by the noise. This is totally workable for me. I understand the concept of raising the volume going into the littlebits to lower the overall noise floor.

  2. I had tested the first half of this before. In fact, I did a test with just the USB I/O bit with no power and the noise was constant. The second half of this example was very interesting to me. With the oscillator set to saw at a low frequency I can clearly see the level of the noise going up and down to almost nothing. With the square at a low frequency I could see an occasional change between the full onslaught of noise to a slightly reduced level. This didn’t make sense to me until I tried #3.

  3. As you said, when the dimmer is outputting approximately 2.5 volts right at 50% the low frequency noise is gone (other than the normal noise floor). This explains the square wave not going from zero noise to full noise as I had initially expected.

Additionally, when the dimmer was at 50% the timbre of the noise floor audibly changed and actually got louder around 9 kHz. It seems like the rest of the time the DC signal is cancelling out that part of the noise and it’s less audible.

I did a video of it (see below), but I think the software I used to record it actually added(/detected?) some extra noise.

^There’s a loud rhythmic pulsing sound that appears to follow the onscreen changes in the frequency spectrum which I can’t hear when I initially record it, but it must be something my DAW is smart enough to filter out. What I did was to I slowly turn the dimmer from 0% to 100%, then back to about 50% for a moment so you can hear the extra 9kHz noise (at around 16 seconds), and then finally back to 0%.

I also recorded it into Edison which is showing the waveform as it shifts over time:

I understand now that the DC signal is normal, but what I want to know is if there’s a way to cut it off completely after the driver reads whatever information it needs to glean from it (other than the workaround of filtering it once it hits my DAW).

Is there any other word on this?

Hi there,

At this time, there is not a way to cut this off completely. Sorry about that! The workarounds suggested are the best ways to mitigate the unwanted noise.

@theodore_littlebits,
That’s a shame, but thank you for helping me understand the issue. :smiley: