As I'm designing my next speaker system, I'm taking into account more of the aspects of speaker design that I simply did not know (we build where we are, right?)... One of the things that my first system lacked was planning around impedance and phase. I tried to design the crossover with the knowledge I had, and this is the resulting phase and impedance (according to XSim).
One of the things pointed out to me was that the impedance of ~2.3 was lower than optimal load for amps. Nothing was said about the phase, but I know it is directly related to impedance just from seeing them both move if I put an L-Pad in place, for example.
So I hunted down a graph of one of the Perlisten s7t so I could have a model of a truly high-end speaker to learn from (rights and gratitude to stereophile: https://www.stereophile.com/content/perlisten-s7t-loudspeaker-measurements
https://www.stereophile.com/images/1121Perl7fig1.jpg ) I see the impedance dips to 3 ohms, are dips better than a constant low load? A constant load is better for the amp, right? This impedance varies quite a bit (relatively to mine, anyway)... but I guess that isn't bad? A constant low load is worse than a varying load? How much load variance is acceptable/optimal?
And then phase... Trying to stay close to 0 is easier for the amp, right? But how much variance there should be allowed/optimal?
On my new design, I can get either of the following (because speakers are really all about compromise, aren't they):
This fairly flat impedance and phase, at a slightly lower SPL and frequency response range:
Or this "worse" (?) impedance and phase, but with a slightly higher SPL and better frequency response range:
If both are acceptable, I'd like to go with the second option since it has better SPL and FRD.... Thoughts? I've done some digging but can't find a general rule of thumb for what makes an "optimal" system with these particular questions. Are there articles about this already that I missed?
Thanks, all!
Jess
One of the things pointed out to me was that the impedance of ~2.3 was lower than optimal load for amps. Nothing was said about the phase, but I know it is directly related to impedance just from seeing them both move if I put an L-Pad in place, for example.
So I hunted down a graph of one of the Perlisten s7t so I could have a model of a truly high-end speaker to learn from (rights and gratitude to stereophile: https://www.stereophile.com/content/perlisten-s7t-loudspeaker-measurements
https://www.stereophile.com/images/1121Perl7fig1.jpg ) I see the impedance dips to 3 ohms, are dips better than a constant low load? A constant load is better for the amp, right? This impedance varies quite a bit (relatively to mine, anyway)... but I guess that isn't bad? A constant low load is worse than a varying load? How much load variance is acceptable/optimal?
And then phase... Trying to stay close to 0 is easier for the amp, right? But how much variance there should be allowed/optimal?
On my new design, I can get either of the following (because speakers are really all about compromise, aren't they):
This fairly flat impedance and phase, at a slightly lower SPL and frequency response range:
Or this "worse" (?) impedance and phase, but with a slightly higher SPL and better frequency response range:
If both are acceptable, I'd like to go with the second option since it has better SPL and FRD.... Thoughts? I've done some digging but can't find a general rule of thumb for what makes an "optimal" system with these particular questions. Are there articles about this already that I missed?
Thanks, all!
Jess
A typical audio amplifier, if designed well, has been made to work properly despite impedance variations and despite the impedance level (down to the rated impedance). There are also some amplifiers where it can make a difference due to a different design philosophy, without that being a bad thing.. but you need to take care with the load.
Thanks, Allen.
So as long as the overall load is sufficient (above 4 ohms overall....), and the dips of low load don't coincide with high phase (absolute value), then I'm okay?
Jess
Impedance magnitude variations go hand in hand with impedance phase variations. If you make one flat, the other will also be flat.
The impedance phase variations I see above are not likely to be a problem.
The impedance phase variations I see above are not likely to be a problem.