Originally Posted by JerryLove
The crossover does not "create different resistance". I understand Ohm's Law, KCL, KVL, Maxwell's equations, etc. I don't need finger puppets, thanks; I'll suffer through with my various EE degrees and decades of experience.
Chest thump much?
Ok. Are the LF and HF paths on a shared crossover run in serial or parallel.
Remind me again: If I run two electrical paths with the same resistance in serial, is the resistance the same as each individually or does it double? How about in parallel? The same or half?
It's good to hear that clipping is caused by the input signal though. Now I know I can run 50 speakers off one amp without clipping; afterall, the input voltage is constant. Yay!
The other guy started it, with statements that showed a lack of understanding of basic principles. Chest thumping? No, I was trying to make clear that I understood what was going on. Should have known better, on the 'net presenting credentials inevitably raises hackles and results in derision. Seems strange, but that's the way it is...
You are missing the point of the crossover; it crosses over from one driver to another at a particular frequency. Except right at and near the crossover frequency the two drivers (for a two-way) are never driven by the same signal at the same time. The LF signals go to woofer, which for sake of argument is 8 ohms. The HF signals go the tweeter, again assume 8 ohms. If it were a perfect crossover and the drivers ideal, there would be no bump and every frequency would see 8 ohms. Put another way, if you put in a low tone, it sees only the woofer, which is 8 ohms. Put in a HF and it sees only the tweeter, again 8 ohms. Put in a tone at the crossover frequency, and it sees both, but the crossover splits the energy so you can in fact maintain 8 ohms.
In the real world all sorts of things happen to the impedance and crossovers are usually designed for best sound, not the flattest impedance, but hopefully you get the idea. The two drivers are not really in parallel as you are implying (I assume you know they are not in series).
As for the input signal comment, I do not know if you do not understand what I was trying to say or merely being insulting. Assuming the former, of course eventually current demands will drag the amp down, often by dropping the voltage rails. They are not independent in a real amplifier. However, most audio amplifiers are designed to be near ideal voltage sources, with very low output impedance and high (ideally infinite) current output. Increasing the current will cause the power supply (voltage rails) to sag, and of course there is voltage drop across the output devices that does increase with current. However, barring your 50-speaker example, for reasonable loads the amplifiers clip when the input signal reaches a level that the output signal can no longer follow and the amp becomes nonlinear. When most people speak of clipping, the picture they have in mind is of a signal that is clipped (chopped off) at the top and bottom of the voltage waveform, what you might see on an oscilloscope. And, for most audio amplifiers driving most speakers, that is a reasonable picture.
When you passively bi-amp the same input signal goes to both amplifiers. The output power may be significantly different, and indeed typically the HF signal requires less output power. However, the voltage are roughly the same at the outputs of the LF and HF amps since their inputs are the same; the HF amp is still putting out the LF signal since there is no crossover before the amp. A large LF signal that clips the LF amp, will also clip the HF amp because it is also outputting the large LF signal. Yes, there is essentially no LF current and thus power is low, but the voltage headroom is still compromised by the LF signal.
Example: If there is a two-tone input such that at the output we have a 10 V peak HF signal riding on a 100 V peak LF signal, just two tones, then the peak output voltage is 110 V. If we have two identical amplfiers with 99 V +/- rails, then passive bi-amping that drives both amps with the same signal, clips both amplifiers. It does not matter that there is essentially no power at LF from the HF amp, it runs out of voltage headroom and clips. Just like it does not matter that the extra HF signal on the LF amp requires essentially no HF power (or current); if the amp had +/-100 V rails, that extra 10 V HF signal would still cause the LF amp to clip. Does that help?
Under significant power demands, there will be more drop across the output devices and in the power supply (now assuming non-ideal supply), but that is usually not a large factor, and I said earlier I was in fact neglecting that extra drop. I do not think it is significant in most cases; losing a few Volts of headroom is not a large change in power.
HTH - Don