Mid range out of phase?

Dedicated to those large boxes at one end of the room
Post Reply
David
User
Posts: 123
Joined: Wed Jan 05, 2011 3:12 pm

#1 Mid range out of phase?

Post by David »

Whoops I should have posted this here......

My new speakers from Wilmslow Audio use 12ins bass, ATC mid ranges and Vifa trebles. Wilmslow state the following:
'I can confirm that the wiring diagram is correct and that the midrange is out of phase. This produces a much smoother response from the ATC midrange and is our standard detail when designing crossovers using this unit.


It sounds counter intuitive to connect a mid range unit out of phase. What do you think guys?
User avatar
The Stratmangler
Shed dweller
Posts: 2890
Joined: Fri Aug 24, 2007 1:50 pm
Location: Rossendale, Lancashire

#2

Post by The Stratmangler »

You've picked the wrong night for posting - many will be in attendance at the pre Owston seminar.
They're probably on the fourth or fifth pint by now :lol:

The speaker is a three way, and if the crossovers are 2nd order Wilmslow Audio's statement would be correct.
Chris :happy3:
David
User
Posts: 123
Joined: Wed Jan 05, 2011 3:12 pm

#3

Post by David »

Thanks-reassuring!
User avatar
rowuk
Old Hand
Posts: 453
Joined: Sun Jun 01, 2014 2:50 pm
Location: Germany

#4 Re: Mid range out of phase?

Post by rowuk »

David wrote:Whoops I should have posted this here......

My new speakers from Wilmslow Audio use 12ins bass, ATC mid ranges and Vifa trebles. Wilmslow state the following:
'I can confirm that the wiring diagram is correct and that the midrange is out of phase. This produces a much smoother response from the ATC midrange and is our standard detail when designing crossovers using this unit.


It sounds counter intuitive to connect a mid range unit out of phase. What do you think guys?
I disagree with the concept of out of phase to smooth out frequency response. We do not "hear" frequency response. Our brain is very powerful at filling in holes. That is why audio can be enjoyable at all. We do hear the artifacts that screw the frequency response up! The electrical phase is not the acoustical phase. Connect the midrange in phase. You will notice the difference.

The sound improves even more when the drivers are acoustically aligned.
chris661
Shed dweller
Posts: 2556
Joined: Mon Feb 22, 2010 7:29 am
Location: Sheffield

#5

Post by chris661 »

The above is blatant misinformation, well-intentioned or not.

The crossover circuitry may result in a phase shift of 180 degrees relative to the bass driver around the crossover frequency.
If the midrange driver was connected with positive terminal to positve output of the crossover, the midrange driver would be out of phase with the bass driver.
By reversing the terminals, the midrange driver is now in phase with the bass driver again.

Both drivers will still move towards you during an impulse, but one of them has to be wired backwards to achieve this.

PS - yes, we do hear frequency response. Get a graphic EQ and start pulling frequencies out. Hear the difference?
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#6

Post by Paul Barker »

Thank you Chris, very well presented correction.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
rowuk
Old Hand
Posts: 453
Joined: Sun Jun 01, 2014 2:50 pm
Location: Germany

#7

Post by rowuk »

chris661 wrote:The above is blatant misinformation, well-intentioned or not.

The crossover circuitry may result in a phase shift of 180 degrees relative to the bass driver around the crossover frequency.
If the midrange driver was connected with positive terminal to positve output of the crossover, the midrange driver would be out of phase with the bass driver.
By reversing the terminals, the midrange driver is now in phase with the bass driver again.

Both drivers will still move towards you during an impulse, but one of them has to be wired backwards to achieve this.

PS - yes, we do hear frequency response. Get a graphic EQ and start pulling frequencies out. Hear the difference?
Hi Chris,

let me explain. The phase shift is only at the crossover point and applies to those frequencies to the left and right of the xover frequency. The rest of the range, any interaction between the rest of the crossover and the response cannot be dealt with without seeing the whole circuit diagram. When we reverse the phase of the midrange we screw up 3-4 octaves not around the crossover to fix what is only perhaps a broken ONE octave around the crossover frequency. In my opinion a VERY bad exchange. The supposed suckout due to electrical phase can easily be solved by moving the HP on the low frequency driver up or LP on the midrange down to "flood" the crossover area with more signal.

Your example with the equalizer may be academically correct, but not what I meant and not for the reason that you mentioned. Our hearing has a great ability to fill in holes or correct for exaggerations. If it didn't, MP3 (reduction of information) or Lowther speakers (smooth frequency response challenged) or small listening rooms (big dips and peaks) for instance would not work. Because of this gift of great hearing ability, many audiophiles in fact do live with very large deviations from "flat", but still haven't lost any information as long as the phase response remains more or less intact.

Our aural sense of perception is based on phase. When we hear sounds, there is a set time delay across the face that our ears and brains can piece together. We know from where the sounds come - even without seeing them. The first time that we hear a new sound, we cannot localise it. Our brain has to learn how to process the phase information first.

Recording engineers also have to keep early reflections under control to preserve many aspects relating to imaging and "color" of the sound.

Many audiophiles have a similar early (out of phase) reflection frequency response suckout in their playback. This comes from a major floor reflection between the speakers and listening position. For speakers about a metre off the floor and 3 metres away from the listening seat, this suckout is at around 900Hz. If we had true flat response (perhaps a thick carpet on the floor to prevent the reflection) but set our equalizer to simulate the response anomaly, we could learn to ignore the frequency response error after a short while. Once made aware of a phase induced issue, it NEVER goes away because too many things other than frequency response are also simultaneously screwed up.

My only point is that when we get it right, our brains have much less work to do and that means that the quality of our listening improves greatly.

Improper phase can have various effects on the sound. For static, held out notes, perhaps nothing is even noticable. The "attacks" of the violin bow, the tonguing of a wind instrument, percussive attack from drums all profit by getting it right however. The space around the instruments, often thought to be a low frequency artifact, is broad band and determined during the first 10ms of sound information fed to our brain. That is why on a well set up playback, a piano or symphony orchestra can sound very "big" but a chamber group or solo instrument sounds intimate.

I do suggest experimenting and extensive reading as there is so much good information on our hearing and wave propagation completely ignored by the audio community. I am not trying to start a flame war. I have extensively experimented with analog and digital crossovers and have come up with quite a bit that challenges the marketing machine for commercial offerings.

There is a great deal more depth in the discussion of proper frequency response, phase response, crossover design. In this particular case, it seems strange to me that one octave "perhaps" is more important than three octaves "for sure".
Whenever I feel blue, I start breathing again.
Post Reply