DigiGrid IO Signal Characterization (Limiting)

SoundGrid Platforms and Technology
1

I’m measuring the electrical characteristics of my SoundGrid hardware channel IO, and would be interested in learning if anyone has gone down this road before.

Specifically, I’m generating tones that reach maximum output peak (digital max) and measuring their voltages on the outputs. I’m finding that digital -0.7dB is about the highest I can go without running into headroom issues.

This is accomplished by using the eMO Generator (sine wave) as an only plug-in on a channel in LV1, and monitoring the output on an oscilloscope.

I’m just getting started running experiments, so I haven’t reached any conclusions yet, but with what I’m seeing, I’m going to be limiting my signals to < -1dB for safety.

Any feedback/thoughts will be appreciated.

2

After further characterizing… patching a physical output to a physical input and watching the meters in the SG control panels…

-3.1dB seems to be the safe limit. Not just because LV1 is set to turn the meters red at -3.0db, but to actually have approximately 2.4dB of electrical headroom. I still don’t know why -0.7dB seems to be the analog limit for DG outputs before distortion artifacts. I suppose that might be better than hard digital clipping at 0dBFS, but that’s just a guess.

I’ve decided to enforce headroom using a two True Peak limiter approach.

  1. I use Waves WLM Plus on the main output (LR Channel in LV1) with True Peak enabled and set to -3.1dB. This enforces the analog headroom from LV1 through all of the amplification equipment. That circuitry has at least +4dB headroom, so complete safety there.

  2. In my DAW, I use another True Peak limiter set to -3.5dB. The DAW shows the gain reduction on the master channel, so I can instantly see if reduction is being applied, so I can back things off a bit if summing gets things out of control.

This approach keeps my analog realm distortion free, and provides a degree of safety when working with certain material, such as live synths. As long as I don’t see any gain reduction being applied, I’m in the clear.

I’d love to know if anyone else uses an approach like this, or if there’s a better way…

1 Like