Two Boxsim projects, which one would you build?

[Kaspari]

Originally Posted by cT equals piD

Most of the time, it isn't too difficult to get DRIVER IMPEDANCE in the cross-over region fairly flat.

That's what I've learned too. But I calculated two LCR circuits to compensate the tweeter resonance and both didn't have much, or any effect on the tweeter impendance peak at crossover point in the Boxsim simulation. Maybe the software is just lacking here. When I look in the Boxsim projects database my design has a very moderate impendance swings compared to most other projects I see. Pretty much everything I see in the boxsim projects database has swings almost twice as big or even bigger.

And yes, I was indeed referring to the impendance the amp sees. I know a proper soldid state won't have trouble coping. But still the rising impendance will have some effect on the crossover won't it?

[mondogenerator]

I wasnt even aware that there was a project database in boxsim. Ive never used it.

[cT equals piD]

Originally Posted by Kaspari

But still the rising impendance will have some effect on the crossover won't it?

I don't think that rising impedance of the speaker system would necessarily adversely affect crossing over.

Assuming that Boxsim is a sophisticated simulation program, which I would think that it is, if the simulation shows flat frequency response, then you are done. Flat frequency response is the end goal of any speaker system design and everything else doesn't matter if the simulation shows a fairly flat amplitude versus frequency characteristic.

If for example the simulated impedance of the woofer at the cross-over frequency is twice the impedance that the low-pass filter was designed to, then in the simulated frequency response you would expect to see a hump in the vicinity of the cross-over frequency.

My two cents,
Pete

[Kaspari]

Originally Posted by cT equals piD

I don't think that rising impedance of the speaker system would necessarily adversely affect crossing over.

Assuming that Boxsim is a sophisticated simulation program, which I would think that it is, if the simulation shows flat frequency response, then you are done. Flat frequency response is the end goal of any speaker system design and everything else doesn't matter if the simulation shows a fairly flat amplitude versus frequency characteristic.

If for example the simulated impedance of the woofer at the cross-over frequency is twice the impedance that the low-pass filter was designed to, then in the simulated frequency response you would expect to see a hump in the vicinity of the cross-over frequency.

My two cents,
Pete

Thank you for your two cents, I can use them

• I do have to ad to this that phase response is not something to be overlooked either, but you're right in the end the freq response should be your main guideline.
• Also there is flat and flat. I don't want a big driver adding to the high octaves eventhough the end plot is reasonably flat. You'll still hear the breakup of the bigger driver eventhough the freq response plot is flat.

I have one problem though, in the electrical response plot the voltage on the woofer rises again after the crossover point. It annoys the hell out of me and so far I haven't been able to fix it, well not without adding quite a few extra parts to the XO. It has very little influence on the freq response plot and it only rises to -25dB, but I just want it flat after the crossover point. I feel if I don't fix it it points towards a flaw in the crossover I don't want there. I don't know what or how though.

Help would be greatly appreciated

[cT equals piD]

Originally Posted by Kaspari

Thank you for your two cents, I can use them

• I do have to ad to this that phase response is not something to be overlooked either, but you're right in the end the freq response should be your main guideline.
• Also there is flat and flat. I don't want a big driver adding to the high octaves eventhough the end plot is reasonably flat. You'll still hear the breakup of the bigger driver eventhough the freq response plot is flat.

I have one problem though, in the electrical response plot the voltage on the woofer rises again after the crossover point. It annoys the hell out of me and so far I haven't been able to fix it, well not without adding quite a few extra parts to the XO. It has very little influence on the freq response plot and it only rises to -25dB, but I just want it flat after the crossover point. I feel if I don't fix it it points towards a flaw in the crossover I don't want there. I don't know what or how though.

Help would be greatly appreciated

I was oversimplifying somewhat, that's true. Dickason in his Loudspeaker Cookbook, if I remember correctly, quotes some figure of phase angle of the impedance of a speaker system that can be "troubling" for an amplifier. Unfortunately, that is all he has to say about it, he doesn't elaborate as to what "troubling" would entail.

Also some amplifiers aren't tolerant of a load impedance well below the nominal load impedance that the amp is designed for.

The obvious suggestion to eliminate rising voltage drop across the woofer is a zobel network on the woofer.

Regards,
Pete