I guess move around in your room until you are "aligned" with the horn cell walls to where you can't see the driver. Do you get the same effect as before? Does it result in poor center image localization, reduced apparent vocal intensity, with the now diffuse image "recessed"?
Emphasis, I think - i.e. on channel becomes dominate enabling you to localize better. (..makes me think of mid-side recording needing sum and difference for "stereo".)
I suspect you could also try altering channel balance (..lowering the spl of one speaker), of course doing that should result in "image pull" to that loudspeaker where simply flipping polarity in one should not. If that at least partially restores center localization then you have your answer.
-well, at least now you have an answer to the original question (..if perhaps not this new one.)
Last edited:
Hello,
In the case of poynomial filters you may want to reverse the polarity of the HF loudspeaker (e.g. 2 ways case). If you don't either you can have a hole in the frequency response curve (e.g. 2nd order Butterworth or LR) or a bad impulse response (e.g. if you don't reverse the polarity of the HF driver using a Butterworth 3rd order, or a quasioptimal Le Cleac'h 3rd order crossover...).
This is not related to stereo listening but the midrange drop out may eventually has such origin.
Best regards from Paris, France
Jean-Michel Le Cléac'h
In the case of poynomial filters you may want to reverse the polarity of the HF loudspeaker (e.g. 2 ways case). If you don't either you can have a hole in the frequency response curve (e.g. 2nd order Butterworth or LR) or a bad impulse response (e.g. if you don't reverse the polarity of the HF driver using a Butterworth 3rd order, or a quasioptimal Le Cleac'h 3rd order crossover...).
This is not related to stereo listening but the midrange drop out may eventually has such origin.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Thanks Scott. I may play around with that a bit and let you know what I find. I can do a lot of level, polarity, delay right from the listening position via my player software (JRiver)
Anyhow, in practical terms, problem solved with toe in. Still a strange phenomenon that had me scratching my head for weeks. Thanks for all the suggestions and knowledge. This is an odd one.
I know that effect! Got is once with another system using Altec 811 horns. Woofers correct, one horn flipped polarity. It was wild. Trumpets out the window to the right, saxophones in the kitchen, voice behind me at the neighbor's house. A cool effect, but a definite muting of vocals. I was not getting that effect this time. Things sounded normal, except the loss of midrange in the center.
Acoustically the slopes are close to L-R 4th order. Electrically they are a bit arbitrary in order to achieve the acoustic slope I need. Woofer and mid impulses are both positive, as they should be this filter. The mystery was that flipping the polarity of just ONE horn made it better. The horns were then 180 degs out from each other - one with a positive impulse, the other with a negative impulse.
Anyhow, in practical terms, problem solved with toe in. Still a strange phenomenon that had me scratching my head for weeks. Thanks for all the suggestions and knowledge. This is an odd one.
Thanks Scott. I may play around with that a bit and let you know what I find. I can do a lot of level, polarity, delay right from the listening position via my player software (JRiver)
Anyhow, in practical terms, problem solved with toe in. Still a strange phenomenon that had me scratching my head for weeks. Thanks for all the suggestions and knowledge. This is an odd one.
I know that effect! Got is once with another system using Altec 811 horns. Woofers correct, one horn flipped polarity. It was wild. Trumpets out the window to the right, saxophones in the kitchen, voice behind me at the neighbor's house. A cool effect, but a definite muting of vocals. I was not getting that effect this time. Things sounded normal, except the loss of midrange in the center.
Acoustically the slopes are close to L-R 4th order. Electrically they are a bit arbitrary in order to achieve the acoustic slope I need. Woofer and mid impulses are both positive, as they should be this filter. The mystery was that flipping the polarity of just ONE horn made it better. The horns were then 180 degs out from each other - one with a positive impulse, the other with a negative impulse.
Anyhow, in practical terms, problem solved with toe in. Still a strange phenomenon that had me scratching my head for weeks. Thanks for all the suggestions and knowledge. This is an odd one.
Charliemb came by for a visit and system listen today. We had a great time and listened to a wide verity of tracks.
He wanted to get an idea of what a large driver and horn system like the one Lynn is building sounds like. I think I sold him on the idea.
Charlie was nice enough to bring a pair of Susan Parker Zeus mono-blocks that he built. Excellent amps! Smooth, strong, detailed. You can see them on the floor in the photo below.
Pretty different tonal balance from the little Dynakit ST-35 we started the session with.
Lynn's system will be similar enough to mine to make the comparison valid. Lynn's speakers have a good chance of being even better.
He wanted to get an idea of what a large driver and horn system like the one Lynn is building sounds like. I think I sold him on the idea.
Charlie was nice enough to bring a pair of Susan Parker Zeus mono-blocks that he built. Excellent amps! Smooth, strong, detailed. You can see them on the floor in the photo below.
Pretty different tonal balance from the little Dynakit ST-35 we started the session with.
Lynn's system will be similar enough to mine to make the comparison valid. Lynn's speakers have a good chance of being even better.
Attachments
I use Ultimate Equalizer from Bodzio Software to do it.OK wise guy, come set up my linear phase response for me.
The key is where you take the measurement. In my case, I felt that the best measurement was near field measurement, with a full range driver, I took measurement 5mm from the root of the cone, not even on center.
The reason for doing this is I wanted it to correct for nonlinear effects of the motor and amplifier, and avoid trying to remove diffraction from the driver, enclosure, etc. Different drivers will have different measurement locations depending on what you are trying to eliminate. Anything up to certain driver diaphram vibration modes are okay. For horns, I would probably consider measuring inside. I have made some experimental measurements with wave guides I made, and inside still seems the best location.
Trying to compensate for anything after radiation from the driver becomes very location specific, but the addition in the signal is universal, so it is best to take the driver measurement where diffraction content is at the lowest.