> simple stepup transformer and bridge rectifier not work ?
It can.
If the audio signal is BIG (loudspeaker amp) and his moving-coil gizmo is small (meter, pen recorder), and no strict requirements on rise/decay (or coil mass can give suitable action), then it can work.
VU meters work just this way. And though meters were old, it was about 1938 before they could make a moving coil meter with dynamics useful for monitoring audio, and then not cheap. True VU meters are out of production, VU-marked meters are less sensitive and dynamic, but opamps are so cheap we can do the buffering and timing in 19 cent chips and 13 cent caps.
It can.
If the audio signal is BIG (loudspeaker amp) and his moving-coil gizmo is small (meter, pen recorder), and no strict requirements on rise/decay (or coil mass can give suitable action), then it can work.
VU meters work just this way. And though meters were old, it was about 1938 before they could make a moving coil meter with dynamics useful for monitoring audio, and then not cheap. True VU meters are out of production, VU-marked meters are less sensitive and dynamic, but opamps are so cheap we can do the buffering and timing in 19 cent chips and 13 cent caps.
OK, so the project is progressing and I've decided to alter my original mechanism for something off-the-shelf that will likely be A LOT more precise than what I originally had in mind. I really don't mean to over complicate this, but I really do need it to do what I have described. It isn't enough for it to be a rectified signal. It NEEDS to be as perfect a representation of the original audio as possible. Here is what I will be driving.

It's a galvanometer from a laser scanner and it's driver board. It can accurately reproduce 20,000 points per second, or the equivalent to 20kbps audio. Certainly not enough for full 20-20khz audio, but more than plenty for my needs which are for 100hz and below. Instead of a mirror on the end, I have another mechanism that will translate the rotational movement into linear movement that will move a lens.
Like I have described before, The lens NEEDS to come to rest at the end of it's travel NOT the middle, like a speaker cone does. The driver board for the galvo can easily handle DC input without any problems at all (unlike a speaker coil) and regularly does in it's normal operation.
So back to the problem at hand. I need to translate an audio signal from a half positive, half negative signal to one that is all positive, and comes to rest at zero when there is no signal. This translation needs to maintain as much fidelity to the original signal as possible.
One of the things I have considered trying as a proof of concept for this project is to use audacity to manually alter a signal. I'm not a programmer and so would have a pretty hard time of it, but I think it's possible.
I would really rather avoid this if possible because it means it can't be used without these custom files. I'd like to have it take live audio signals either from music or just from the surrounding environment.
I really think the peak detector is the answer I'm looking for. Or some other kind of amplitude follower. Combining this signal with the original audio would produce what I'm looking for.

It's a galvanometer from a laser scanner and it's driver board. It can accurately reproduce 20,000 points per second, or the equivalent to 20kbps audio. Certainly not enough for full 20-20khz audio, but more than plenty for my needs which are for 100hz and below. Instead of a mirror on the end, I have another mechanism that will translate the rotational movement into linear movement that will move a lens.
Like I have described before, The lens NEEDS to come to rest at the end of it's travel NOT the middle, like a speaker cone does. The driver board for the galvo can easily handle DC input without any problems at all (unlike a speaker coil) and regularly does in it's normal operation.
So back to the problem at hand. I need to translate an audio signal from a half positive, half negative signal to one that is all positive, and comes to rest at zero when there is no signal. This translation needs to maintain as much fidelity to the original signal as possible.
One of the things I have considered trying as a proof of concept for this project is to use audacity to manually alter a signal. I'm not a programmer and so would have a pretty hard time of it, but I think it's possible.
I would really rather avoid this if possible because it means it can't be used without these custom files. I'd like to have it take live audio signals either from music or just from the surrounding environment.
I really think the peak detector is the answer I'm looking for. Or some other kind of amplitude follower. Combining this signal with the original audio would produce what I'm looking for.
AHH!! I found something else that looks promising if it can be implemented in hardware
Envelope detector - Wikipedia
Check the "Audio" section
And this is basically what some of you have suggested. It's just a rectified and filtered signal, but if I then use this signal to DC bias the original signal, it will push the signal positive but still allow it to return to zero.
Envelope detector - Wikipedia
Check the "Audio" section
And this is basically what some of you have suggested. It's just a rectified and filtered signal, but if I then use this signal to DC bias the original signal, it will push the signal positive but still allow it to return to zero.
Last edited:
> I found something else ... Envelope detector
I apologize for not spelling it out.
> if it can be implemented in hardware
It is a dead-standard audio building block. Most audio meters, most audio limiters.
Yes, this:
More elaboration wanted for low-level precision. Standard stuff.
...a rectifier. And a low-pass filter so it shows the "envelope"
I apologize for not spelling it out.
> if it can be implemented in hardware
It is a dead-standard audio building block. Most audio meters, most audio limiters.
Yes, this:

More elaboration wanted for low-level precision. Standard stuff.
Right. and the circuit I posted a few days ago is exactly that. A more elaborate version with more precision. Now it's time for prototyping 🙂
We suggested that earlier, but he has since clarified he wants the true waveform undistorted, just offset to the V- peak is at zero. The simple detector rectifies the signal thus cutting off the negative half of the wave form.
> the true waveform undistorted, just offset to the V- peak is at zero.
Except for a constant-amplitude tone, this implies distortion.
> The simple detector ... cutting off the negative half of the wave form.
Full-wave can be done.
Except for a constant-amplitude tone, this implies distortion.
> The simple detector ... cutting off the negative half of the wave form.
Full-wave can be done.
But that is what he wants, he wants to move the signal offset so zero volts is the negative peak.
For discussion's sake, lets say that I have a traditional speaker motor with voice coil, permanent magnet and spider, but no cone. Instead of a cone, lets say there's a very light weight control mechanism that moves a small lens which then changes the focal point of an optics mechanism. For the effect to work , the lens needs to return to either side of it's focal range, but NOT in the center as the cone would if it was a traditional speaker. When there is no audio, the lens needs to come to rest at displacement zero and then when the music plays it should only have positive displacement. So, I need to convert the +/- audio signal into an all positive signal that is still as accurate (at frequencies of about 100hz and below) to the original signal as possible. It would in effect sound exactly the same if played on a normal speaker, except that the speaker would only be able to use half of it's excursion and the audio amp powering it would only use NPN or PNP transistors instead of both. I hope that helps give a better idea of what I'm going for. Like I was saying it's a bit of a strange experiment lol
I understand that what you want is the original audio, but 'riding' on a DC voltage that is higher than the peak value of the signal. I also believe you want to be able to independently control the level of the audio as well as the DC.
You need a simple summing opamp that sums the (variable) signal and a variable DC. Trivial. Google 'summing amplifier'.
Jan
- Status
- Not open for further replies.
- Home
- General Interest
- Everything Else
- convert AC amplitude into modulated DC signal