Re: 2nd order crossover networks

Posted by andyr (M) on December 11, 2004 at 23:34:49

In Reply to: Re: 2nd order crossover networks posted by Jon Risch on December 7, 2004 at 20:10:07:

Hi Jon,

I am currently wrestling with a problem that is associated with crossover phase, so I would really appreciate it if you could answer some simple (for you, I'm sure!) questions.

I have Maggie IIIa speakers; these have a 2nd order crossover for mid-panel to ribbon.

The problem that I am "wrestling with" is:

* because it's a 2nd order crossover, the mid-panel connections are reversed. This is because a 2nd order crossover reverses phase by 180 degrees, so reversing one driver supposedly puts the two drivers back in phase.

So, does this mean that the two drivers ARE really "in phase" as much as one driver would be, if it spanned the frequency range covered by the mid-panel & ribbon, without a crossover??

My reason for asking this question is that I recently added a "polarity reversal" circuit to my phono stage. I know it works because I've seen a 1kHz tone from a test LP "flip" on my CRO when I flick the switch. However, I can't hear much (any!) difference between "normal" and "inverted" when I play a record! :-(

The reason for this could be that I just have tin ears, so I'm stuffed, but it could also be that because one driver is inverted, when I flick the "polarity invert" switch, all I'm doing is exchanging one inverted driver-setup for its reverse??

Perhaps only if the two drivers were connected with the same polarity would I get any benefit from my "polarity inversion" switch?? However, if the inverted connection of one driver restores an "in phase" connection to a 2nd order crossover, does this mean the resulting effect is the two drivers really ARE in phase??

That's the issue I'm wrestling with!!

Your post mentions the theoretical suckout with a 12db crossover when you DON'T invert one driver ... which can be compensated for by increasing the overlap between the two drivers. However, as you say, amplitude summing is not the be-all and end-all!

It seems to me if I followed your suggestion, I might end up with a frequency response which was flat (ie. no suckout) but which had the 2 drivers reversed in phase due to the 12dB crossover ... ie. this is definitely worse than the standard theory (which is to have drivers inverted).

Can you comment?

Regards,

Andy

This post is made possible by the generous support of people like you and our sponsors:

Follow Ups Full Thread

Topic - 2nd order crossover networks - bunni 10:24:05 12/07/04 (19)

Re: 2nd order crossover networks - pkell44 16:33:44 12/09/04 (2)

Re: 2nd order crossover networks - Jim85IROC 09:25:53 12/14/04 (1)

Not quite... - Commuteman 11:40:08 12/14/04 (0)

Re: 2nd order crossover networks - Jon Risch 20:10:07 12/07/04 (2)

Re: 2nd order crossover networks - andyr 23:34:49 12/11/04 (1)

Re: 2nd order crossover networks - Jon Risch 22:05:42 12/12/04 (0)

Can you say "suckout"? - Commuteman 11:34:28 12/07/04 (12)

Re: Can you say "suckout"? - andyr 18:01:48 12/07/04 (6)

See JR's post.. - Commuteman 07:04:58 12/08/04 (5)

Re: See JR's post.. - andyr 12:19:23 12/08/04 (4)

Here's a URL for ya.... - Commuteman 15:18:02 12/08/04 (1)

Thanks, Peter, for all your help. (nt) - andyr 18:31:41 12/08/04 (0)

No, you don't reverse the phase - Commuteman 14:30:58 12/08/04 (0)

Good question - Commuteman 14:24:12 12/08/04 (0)

Wrong math. - SysInfo 17:19:20 12/07/04 (4)

Re: Wrong math. - Ken Perkins 19:33:27 12/08/04 (1)

Just go and measure. - SysInfo 15:59:22 12/09/04 (0)

You're a funny guy..... - Commuteman 09:51:47 12/08/04 (1)

Re: You're a funny guy..... - SysInfo 15:54:03 12/09/04 (0)

Follow Ups

Re: 2nd order crossover networks - andyr 23:34:49 12/11/04 (1)

Re: 2nd order crossover networks - Jon Risch 22:05:42 12/12/04 (0)

You can not post to an archived thread.