quick XO question

[ultrachrome]

I'm feeling my way around a 2-way passive crossover and am getting peaks and valleys in the crossover region.

What's this indicative of?

I've used Bagby's passive crossover spreadsheet and played with the xyz coordinates of the drivers but that doesn't seem to be the problem.

[Ron E]

caused by phase wrapping from time of flight, you need to remove time of flight - not sure how it's done with sound easy.

[ultrachrome]

I thought I had been doing this but I'll go back and recheck.

Thanks!

[phase_accurate]

I've used Bagby's passive crossover spreadsheet and played with the xyz coordinates of the drivers but that doesn't seem to be the problem.

How do you come to the conclusion that it doesn't have an influence ?

What you see here is some sort of lobing. I assume that this software assumes the drivers being point-sources. Am I right with my guess ?

Regards

Charles

[ultrachrome]

I didn't know the cause of the problem at the time so playing with xyz values on the order of 10's of mm did not have an appreciable effect.

The ripples shifted their relative positions but remained within the crossover region.

Maybe if I had added a meter to the z position, the ripples would have been largely eliminated.

I carefully went back over the SPL measurement steps last night and I no longer have ripples.

[john k...]

Originally posted by ultrachrome
I'm feeling my way around a 2-way passive crossover and am getting peaks and valleys in the crossover region.

What's this indicative of?

I've used Bagby's passive crossover spreadsheet and played with the xyz coordinates of the drivers but that doesn't seem to be the problem.

By the way. That 8.2 ohm resistor in series with the 6.28uf cap at the amp side of the woofer filter is not a good design.

[ultrachrome]

I'll keep that in mind. That was a straw man to see if I could get SoundEasy to show a summed response.

Still have a ways to go before I feel confident enough to order the parts.

[ultrachrome]

Here's an attempt at a 3rd order butterworth targeted at 2500Hz.

I'm stuck at this point and need to get a way from the keyboard for a while.

I did the following to get to this point:

• optimized active high pass section (F0, F4, & F9) for 325Hz 24dB/oct target
• optimized passive low pass for 2500Hz 18dB/oct target
• optimized passive high pass for 2500Hz 18dB/oct target
• optimized F15, LP shelving -6dB/oct, to help flatten the tweeter response.

I've run out of tricks, not that I had any to start with. Attached are individual and summed with both tweeter polarities.

One problem I've run into is that if I try to use the optimize function on the T component which is my active 2nd order high pass, SoundEasy locks up. I'm preferring to ignore this for the time being.

[ultrachrome]

Getting closer. Now I get a nice null when I reverse the polarity.

Spent some time away from this and went skiing yesterday. Didn't think about it all day. But before hitting the sack I had a little bit of inspiration.

I tried to use the optimizer in a more focused manner. For example, the parallel trap was optimized separately from the HP 2nd order LR. Optimizing each function separately seems to be the trick.

Can anyone characterize what I have so far? I'm a little elevated above 6k.

Playing around this morning, I can definitely do a better job of taming that region. I understand that it's give and take but I'm not sure what shape to aim for.

Next on my list is to go back over the active sections. I might move the parallel trap to active and I want to see if I can combine the first and third active elements.