Decibels, gain, and perceived volume

[.Mike]

Hi,

I'm wondering if someone could provide me with, or direct me to, a good explanation on the relationship between gain, decibels, and perceived volume. I'm just having a hard time wrapping my head around it. I've searched and have read a decent amount on the subject, but it seems that a lot of the information I have read is conflicting or confusing.

My question is spurred by a tremolo I am working on that is based on the Tremulus Lune (Tonepad project page), but with a different LFO. My LED/LDR combo has an on-resistance of about 150-ohms and an off-resistance of about 1M-ohms. I used a 10k pot for the gain, and it provides a ridiculous boost-- way too much. I think I want to tone it down so the maximum available volume is about double the original volume.

I know I can lower overall gain if I reduce the value of the pot, or further limit the current going to the LED, or add series resistance with the LDR to set a minimum. I can calculate the gain of the inverting opamp stage, but I don't quite understand how that translates to perceived volume.

Thanks in advance,

Mike

[brett]

Hi
RE: gain, decibels, and perceived volume.
a FEW things that happen are:
perceived volume works on a non-linear scale.  The log function  is often used to linearise volts etc to a more linear scale.  Decibels are a log scale.
perceived volume varies with frequency.  Deep bass and highs are not perceived as clearly as mids at normal levels of loudness (hence the "loudness" control on older stereos boost bass and treble)
a sine wave seems to be perceived as much less loud than a square(ish) wave of the same power.  That's probably one reason why high gain and distortion "catch" the ear.
there are many other subtle effects too (like the perceived loudness and appeal of different harmonics)
cheers

[earthtonesaudio]

I'm just having a hard time wrapping my head around it.

Don't feel bad.  Taking the journey from voltage gain in an op-amp circuit, to total system power gain, to perceived physiological loudness in your brain is a bit like waking up, putting your shoes on, getting in your car and driving to Saturn.  The planet, not the dealership.

[.Mike]

Thanks for the responses guys.

I did some more reading last night, a ton of modeling, and a ton of math to try to get a better understanding of all of this. I then applied it to my circuit, and achieved some real world results that were in line with the modeling and research.

If I'm right, I ended up using 5.5X gain to come up with about 14.8dB of gain, or a bit more than a doubling of perceived volume. That's much better than the 34.8dB of gain I had before. Who needs that much boost for a tremolo.

If all goes well, I'll end up releasing a project file for what I'm working on in the next couple of weeks.

Thanks!



Mike