Subwoofer variable phase control

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I think it is something like this that is used (found it from the net somewhere)

***********

Phase shifter circuit
(From Richard Karlquist)
Here is a well known op-amp phase shifter. I am surprised no one has posted it yet, so I guess I will have to. The circuit will have 90 degrees phase shift at:
FREQ = 1/(2*PI*R3*C)

At low frequencies it has 0 degrees of phase shift and at high frequencies it has 180 degrees of phase shift. By making R3 a pot, you can vary the phase at a given frequency from nearly 0 to nearly 180 degrees. Since you want to work at 1 MHz., you will need to use a high frequency op amp, like the Burr-Brown OPA620.

An externally hosted image should be here but it was not working when we last tested it.
 
The one and only
Joined 2001
Paid Member
Maybe you would like to try a circuit that looks
like this:
 

Attachments

  • phase.png
    phase.png
    1.7 KB · Views: 959
Audiofreak if you are trying to keep the phase relationship of the woofer information constant. I think you realy do want delay, but you need to think about it in a different way.

If, for example you want to shift the phase of a 100 Hz tone by 180 degrees, delay it by 5ms, for 90 degrees 2.5ms.

In the high power RF world this type of circuit is known as an allpass filter. It is used to correct for group delay caused by bandpass filters in TV transmitters. This is very much the same problem just at a different frequency. I would guess that there are digital devices out there that can do this.
 
AudioFreak said:
I am looking for a schematic for a variable phase control as found on commercial subwoofer amps ...

Most diyer's seem to just use a 0/180 degree control but i would like a little more control than that say 15degree intervals or so...

Whats involved?


You can have a look here Audio Analogue unofficial webpage , under the section "further projects".

Cheers

Andrea
 
Danne said:
This is exactly what I want to build: a variable phase control. I've looked at Mr. Linkwitz "Delay circuit" but how do I make it variable?

allpass2.gif


I want to use it up to say 100 Hz.

/Daniel

Reverse R and C, and make R variable. Follow R with another resistor to ground so that R is never 0. Place another resistor equal to Ro ahead of C to balance the two out. This is what I have for a schematic for one of those plate amps that you can buy, that actually has variable phase (that's the time delay that you're looking for). The values are of course different, but the circuit is the same.

Ron
 
Funnily, I have debated a great deal about this subject in one our national forum. It seems that people have difficulty understanding the concept of constant group delay or constant phase shift vs. frequency. Time alignment (i.e. delay) is really what is required, not phase. But phase is invariably married with delay.

I once made plot with Matlab where all square wave frequency components (I don't remember anymore how many harmonics I included, and texts are in finnish (sorry), but numbers mean amount of phaseshift in degrees) were shifted same amount of degrees instead of constant time delay. As you might have guessed, that signal won't resemble square wave at all when phase is changed that way.

Device which shifts all frequencies 90 degrees is called a hilbert transformer.

With Best Regards,
Janne
 
C gets connected directly to the + terminal of the opamp. Rv gets connected between the + terminal and R1. The wiper of Rv gets connected to the connection between Rv and R1. Basically Rv and R1 are in series, and both are shunted between + terminal and ground.

TL072 will probably work, but why not use something better, it's only 1 opamp.

I don't have exact values with me (I'm at work) and the schematic I have is very difficult to read. C is 0.022uF, Rv is 100k, R1 I think is 988 ohm, the rest I can't read. I'd have to open up my amp and measure the values. Perhaps you have a simulator program? Or if you can wait, I might have time after the weekend to look at it. I should have the values for myself anyways.

Regards,
Ron
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.