Would a shifter that shifts +135 deg. over a decade range in the audio frequency spectrum be possible? The angle must be +/- 135 deg., not 90 deg. that I know can be done.

Thanks,
Pete

 

So what would you expect a square wave (that has been suitably band limited to your decade range) to look like after shifting every frequency 135°?

 

Would a shifter that shifts +135 deg. over a decade range in the audio frequency spectrum be possible? The angle must be +/- 135 deg., not 90 deg. that I know can be done.

In theory it could be done digitally for a single frequency, but if it is mix of frequencies, such as music or voice, then no.

 

Would a shifter that shifts +135 deg. over a decade range in the audio frequency spectrum be possible? The angle must be +/- 135 deg.

First thought involves a PLL.

Second thought involves a DDS signal generator. I cannot just shift any old signal 135 but if I make the signal, I can also make another signal that is shifted any amount you want. I have made signals generators that can make many signals that are based off the first signal but different in some way. Using DDS the frequency does not matter. The fastest I have is 500mhz and the slowest has a 1/f of hours.

I don't really understand what you are trying to do.

 

I'm sorry, I should have been more specific. Say over the range from 1 kHz to 10 kHz, I would want to shift the phase of each sine wave in that frequency range by 135 degrees. I'm not familiar with digital processing so I would prefer to do it by analog methods. The method that I'm familiar with is first and second phase shifters (or all pass filters) shifting in a way that the phase difference between the two outputs is equal to the needed shift angle over a range of frequency of the sine wave signal at input to the shifters. For a 90 degree shifter, the identical sine wave is at input to both shifters and the phase of the output of one of the shifters is 90 degrees with respect to that of the other one.

A method that I've seen is a first phase angle shifter that produces a 90 degree shift at 1 kHz and a second shifter that produces a 90 degree shift at 10 kHz. Apply a sine wave of a given frequency in the range1 kHz- 10 kHz to the inputs of both shifters and there is a phase difference between the two outputs equal to (about) 90 degreees.

 

What is the application for this phase-shifted signal?

 

What is the application for this phase-shifted signal?

That would be very complicated to answer and I don't think that it would contribute to being able to answer my question. Since you asked I would do if tor you, but it would unnecessarily make the discussion more complicated in my opinion.

 

If you can do 90, simply add in the right amount of 180 deg to get 135. In complex phasor terms, a 90 degree shifter multiplies by j, so multiplying by (-1 ± j) gives ±135 deg.

 

The method that I'm familiar with is first and second phase shifters (or all pass filters) shifting in a way that the phase difference between the two outputs is equal to the needed shift angle over a range of frequency of the sine wave signal at input to the shifters. For a 90 degree shifter, the identical sine wave is at input to both shifters and the phase of the output of one of the shifters is 90 degrees with respect to that of the other one.

That would give you two signals 90° apart relative to each other but not +/-45° with respect to the input signal. You can do 270° over a certain frequency range between the two outputs but that is not equivalent to +/-135° relative to the common input signal.

 

For every frequency to be shifted by 135°, that is 37.5% of its period. So every frequency would have to be delayed by a different period of time. And how many frequencies are there between 1kHz and 10kHz? Isn't the answer to that infinity? so the answer to the query is a clear "no".

 

And how many frequencies are there between 1kHz and 10kHz? Isn't the answer to that infinity? so the answer to the query is a clear "no".

Sorry, but this logic escapes me. How many voltages are there between 1 and 10 Volts? Infinity. Does that mean I cannot make a variable supply that outputs any voltage between 1 and 10 Volts?

 

That would give you two signals 90° apart relative to each other but not +/-45° with respect to the input signal. You can do 270° over a certain frequency range between the two outputs but that is not equivalent to +/-135° relative to the common input signal.

Yes, thanks I realize that. For my application it's acceptable for there to be a changing phase angle with frequency between that of the input signal and the two outputs. It's possible to make use of the fact that there is a constant phase angle difference between the outputs of the two shifters while frequency of the input signal is variable over a specified range.

 

For every frequency to be shifted by 135°, that is 37.5% of its period. So every frequency would have to be delayed by a different period of time. And how many frequencies are there between 1kHz and 10kHz? Isn't the answer to that infinity? so the answer to the query is a clear "no".

It can be done for a constant phase shift of 90 degrees. However my initial looking into it indicates that producing a constant phase shift of 135 degrees is problematic. Constant phase shifters were first developed in the 1950s.

 

Sorry, but this logic escapes me. How many voltages are there between 1 and 10 Volts? Infinity. Does that mean I cannot make a variable supply that outputs any voltage between 1 and 10 Volts?

Then you need to rethink your logic.
You really think there's a real analogy between generating a single supply voltage from another voltage, and generating the same phase-shift for any frequency in a range of frequencies?

 

That would be very complicated to answer and I don't think that it would contribute to being able to answer my question. Since you asked I would do if tor you, but it would unnecessarily make the discussion more complicated in my opinion.

And I believe it would be less complicated.
Sometimes the question being asked to solve a problem, could be done in a simpler manner if the complete problem is understood.

But it's you problem so if you want to be secretive about it, good luck finding a solution, since with the information we have, I see none.

 

Then you need to rethink your logic.
You really think there's a real analogy between generating a single supply voltage from another voltage, and generating the same phase-shift for any frequency in a range of frequencies?

I didn’t say there was no difference. Obviously, the latter is far more difficult. What I take issue with us saying that the fact that there are an infinite number of possible outputs does not mean it is impossible. In fact, that s a characteristic of all analog systems.

I am thinking a frequency to voltage converter, followed by a voltage controlled all-pass filter. Don’t know how to do that, but I doubt that it is impossible.

 

programmable delay line... measure input frequency and set desired delay.
perhaps LTC6994 or similar..

 

Sorry, but this logic escapes me. How many voltages are there between 1 and 10 Volts? Infinity. Does that mean I cannot make a variable supply that outputs any voltage between 1 and 10 Volts?

It would seem that you, @crutschow and I have very different interpretations of the TS's vague requirements.
To me "in the audio spectrum" implies an audio signal, i.e. a mix of frequencies. White noise as an representative example contains EVERY frequency, so phase shifting each one of them (delaying by θ/f) would therefore be impossible.

 

I am thinking a frequency to voltage converter, followed by a voltage controlled all-pass filter. Don’t know how to do that, but I doubt that it is impossible.

programmable delay line... measure input frequency and set desired delay.
perhaps LTC6994 or similar..

Don't see how either of those work when the signal is composite of frequencies, such as music.

 

i still see no purpose of this application or exact specs, just that 135deg shift is needed for frequencies in range 1kHz-10kHz
and if multiple frequencies can occur at the same time, split signal in frequency bands using filters, use individual delay for each band, then sum the signals back...
this can be done with discrete parts or using DSP (which TS does not want)