Zero phase-shift

[gareth]

Are there any designs that have zero phase-shift from input to output that are simple enough to be built by someone who has only built a headphone amp up until now?

Thanks
Gareth

[GRollins]

To the extent that all real world devices have some amount of capacitance and inductance, the literal answer is that no such thing exists. In the real world, your concern is that all frequencies are delayed by the same amount. There are any number of competent designs that meet that criterion. Can you be more specific as to what you're looking for?

Grey

[gareth]

Thanks grollins,
What am I looking for? I'll start by saying why I was looking for zero shift. I thought (probably wrong) that by keeping both V and I in phase from input to output then (with good design) sound quality would remain high, and also power wastage would remain low.

I say this based on the fact that with normal ac electric principles (I am an electrician) that being in-phase (0degrees) reduces power factor (lead/lag). Also, I think I read that HF travels with the voltage while LF travels with the current (I could have totally mis-interpeted what I read though).

What sound would I like? As best as is reasonably possible without creating too many headaches (with the design). Probably what everyone is looking for.

I hope I have explained myself properly here.

Gareth

[myhrrhleine]

Originally posted by gareth
Thanks grollins,
What am I looking for? I'll start by saying why I was looking for zero shift. I thought (probably wrong) that by keeping both V and I in phase from input to output then (with good design) sound quality would remain high, and also power wastage would remain low.

I say this based on the fact that with normal ac electric principles (I am an electrician) that being in-phase (0degrees) reduces power factor (lead/lag). Also, I think I read that HF travels with the voltage while LF travels with the current (I could have totally mis-interpeted what I read though).

What sound would I like? As best as is reasonably possible without creating too many headaches (with the design). Probably what everyone is looking for.

I hope I have explained myself properly here.

Gareth

The phaseshift in that regard is negligible.

[gareth]

So i guess I am looking down the wrong road here then?

Gareth

[GRollins]

Voltage and current don't separately carry low and high frequencies. If you want to talk about phase shift, you're going to need to look at the reactive components of a given circuit and the bandwidth. There will be phase shift at the ends of the bandwidth, but you can safely assume that all the frequencies in the middle will arrive at the same time.
This can be made a lot more complicated if you want, but I'm assuming you want the sound bite version...

Grey

[gareth]

Originally posted by GRollins
Voltage and current don't separately carry low and high frequencies. If you want to talk about phase shift, you're going to need to look at the reactive components of a given circuit and the bandwidth. There will be phase shift at the ends of the bandwidth, but you can safely assume that all the frequencies in the middle will arrive at the same time.
This can be made a lot more complicated if you want, but I'm assuming you want the sound bite version...

Grey

OK, so I got a little confused or misunderstood what I read there then and to be quite honest I found it a little hard to get my head around (HF and LF carried on V and I).
So is what I am looking for a non-starter? I understand the complex rectances that take place due to the audio spectrum.
I guess I wil start looking around again then.

Gareth

[Conrad Hoffman]

In a nutshell, if you build something with much wider bandwidth than you need, it will have near zero phase shift in the area you care about. This assumes no crazy all-pass networks in the circuit, but that's probably a safe assumption. Build a power amp that's -3 dB at 3 Hz and -3 dB at 100 kHz, and the phase shift between 20 and 20000 will be very small. Reasonable phase shift isn't generally considered audible, but IMO it's still a desirable thing to keep it to a minimum.

[GRollins]

I'm still not clear how you came to believe that this is a major problem in audio gear. Depending on how you want to look at things, your average dynamic loudspeaker (the kind with a voice coil and magnet) is regarded as a current-driven transducer; the idea that voltage is involved is just an inconvenient side effect. In fact, you can make an excellent argument for using current source amplifiers (as opposed to voltage source amps) to run midranges and tweeters. There's an even stronger case to be made for using them for woofers, except for the annoying fact that woofers need a low impedance source driving them to keep the cone from running away. That, in turn, can be overcome by using a current source amplifier, then using a servo system to hold the cone back...but that's getting rather farther down the road than you'll be wanting to go at the moment.
Trust me, this is tantamount to fretting over the color of paint on a race car when discussing how fast it will go.

Grey

[Leolabs]

Unless got things such as 'zero capacitance,zero inductance',otherwise there is no such thing as 'zero phase shift'.