开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
  1. DahlquistSpeakers
  2. Topics

DQ-10 phase issue found and corrected

 

I bought a pair of DQ-10s at a local thrift shop about a year ago, and had been thoroughly enjoying them for at least a month when I discovered that one of them had a phase issue. Here's what happened...

I was reading another audio forum and came upon the subject of using test records/CDs to evaluate stereo systems and or speakers. Out of curiosity, I found The Best of Chesky Classics & Jazz Audiophile Test Disc Volume 3 on Tidal and gave it a listen. To my surprise, Track 28 - Natural Stereo Imaging, and Track 29 - Artificial Stereo Imaging indicated there was a problem with my right speaker. The imaging would not extend past the right edge of the speaker. But with the left speaker, the imaging extended well past the left edge. In other words, the soundstage from my DQ-10s was being cut off drastically on the right side.?

After several hours of trying to figure out what the heck was going on, I discovered that the tweeter and super tweeter on the right speaker were wired out of phase. It was an easy fix, and now my DQ-10s sound the way they should... a soundstage that extends nearly 90 degrees to the left and to the right.?

Funny thing is, I was under the impression that my speakers were factory mirror imaged (SN 47247 and 47295), and the wiring appeared to be untouched. Anyway, I was clueless that there was a problem until I listened to that test CD.?

Disclaimer: I have no affiliation with Chesky recordings...
 
Edited

When you say the right side tweeter and super tweeter were out of phase, ?what were you using as a reference -the left side? Or a schematic?
The behavior you have observed, after reversing, apparently is correct per the original designer.
The dome mid, tweeter, and super tweeter are all out of phase with respect to the cone mid and the woofer. This gives the speakers their spacious imaging feel . I?found this out by reading through all Jon Dahlquist’s interviews and also talking to Carl Marchisotto who was Jon’s right hand man. He confirmed the phase shown in Rev9 schematic is correct. Unfortunately, ?I have lost the interview link where Jon described the crosstalk cancellation. If anybody has it I would appreciate a copy.


The gradual 180 degrees phase shift , combined with the laterally offset ?drivers causes crosstalk cancellation ( reduces leakage between channels). The implementation strategy is similar to ?Polk SDAs without using a second set of drivers. This is an advanced design given its age. Mirror imaging the speakers and carefully aiming them so a point midway between ?the ?mid cone and mid dome points to the corresponding ear maximizes the effect. Images will appear outside the lateral extent of the speakers.
 
Edited

Hi Charlie,
?
First, using the left speaker as a guide, I verified the wiring according to the crossover schematic. Then I carefully compared everything to the right speaker. I could see how the tweeter and super tweeter had been rotated 180 degrees making the?+ and - terminals reversed. I swapped the wires and whoa Nellie! That did the trick.?
?
The?test tracks I referred to in my OP feature a drum being tapped. It starts out dead center, then moves left, then far left, then right, then far right. Before my fix, when the drum was supposed to go far right, it actually sounded like it was moving slightly to the left. After the fix, the drum moved to the far right, as was intended.?
?
I'm absolutely positive that they were phased incorrectly, and are now correct. There is absolutely no doubt in my mind, and the test tracks prove it as far as I'm concerned.
?
I'll post a photos tomorrow showing the left and right speaker wiring.
?
Bill

On Fri, Dec 22, 2023 at 7:21?PM Charlie Conger <ctconger@...> wrote:

When you say the right side tweeter and super tweeter were out of phase, ?what were you using as a reference -the left side? Or a schematic?
The behavior you have observed, apparently is correct per the original designer.
The dome mid, tweeter, and super tweeter are all out of phase with respect to the cone mid and the woofer. This gives the speakers their spacious imaging feel . I?found this out by reading through all Jon Dahlquist’s interviews and also talking to Carl Marchisotto who was Jon’s right hand man. He confirmed the phase shown in Rev9 schematic is correct. Unfortunately, ?I have lost the interview link where Jon described the crosstalk cancellation. If anybody has it I would appreciate a copy.


The gradual 180 degrees phase shift , combined with the laterally offset ?drivers causes crosstalk cancellation ( reduces leakage between channels). The implementation strategy is similar to ?Polk SDAs without using a second set of drivers. This is an advanced design given its age. Mirror imaging the speakers and carefully aiming them so a point midway between ?the ?mid cone and mid dome points to the corresponding ear maximizes the effect. Images will appear outside the lateral extent of the speakers.


?

?
 

开云体育

I agree and have heard about that 180 degree swap before. To be clear I totally believe you and agree.
This is the way Jon intended the speaker to be.?

Charlie Conger

On Dec 22, 2023, at 10:09 PM, wmsimpson125@... wrote:

?

[Edited Message Follows]

Hi Charlie,
?
First, using the left speaker as a guide, I verified the wiring according to the crossover schematic. Then I carefully compared everything to the right speaker. I could see how the tweeter and super tweeter had been rotated 180 degrees making the?+ and - terminals reversed. I swapped the wires and whoa Nellie! That did the trick.?
?
The?test tracks I referred to in my OP feature a drum being tapped. It starts out dead center, then moves left, then far left, then right, then far right. Before my fix, when the drum was supposed to go far right, it actually sounded like it was moving slightly to the left. After the fix, the drum moved to the far right, as was intended.?
?
I'm absolutely positive that they were phased incorrectly, and are now correct. There is absolutely no doubt in my mind, and the test tracks prove it as far as I'm concerned.
?
I'll post a photos tomorrow showing the left and right speaker wiring.
?
Bill

On Fri, Dec 22, 2023 at 7:21?PM Charlie Conger <ctconger@...> wrote:

When you say the right side tweeter and super tweeter were out of phase, ?what were you using as a reference -the left side? Or a schematic?
The behavior you have observed, apparently is correct per the original designer.
The dome mid, tweeter, and super tweeter are all out of phase with respect to the cone mid and the woofer. This gives the speakers their spacious imaging feel . I?found this out by reading through all Jon Dahlquist’s interviews and also talking to Carl Marchisotto who was Jon’s right hand man. He confirmed the phase shown in Rev9 schematic is correct. Unfortunately, ?I have lost the interview link where Jon described the crosstalk cancellation. If anybody has it I would appreciate a copy.


The gradual 180 degrees phase shift , combined with the laterally offset ?drivers causes crosstalk cancellation ( reduces leakage between channels). The implementation strategy is similar to ?Polk SDAs without using a second set of drivers. This is an advanced design given its age. Mirror imaging the speakers and carefully aiming them so a point midway between ?the ?mid cone and mid dome points to the corresponding ear maximizes the effect. Images will appear outside the lateral extent of the speakers.


?

?
 

From what I understand the woofer and tweeter are in phase while the mid-woofer, midrange and supertweeter are out of phase. This unfortunately not only alters the imaging but also the timbre of the instruments. In my DQ-10s (I have two pairs) I connected all the transducers in phase and currently listen to them with an equalization that compensates for the alterations in the frequency response module generated by the modified connections. In my opinion, the result is very satisfactory, not only as regards imaging but also impulsive sounds.

Franco

Il 22/12/2023 22:28 CET wmsimpson125@... ha scritto:
?
?
I bought a pair of DQ-10s at a local thrift shop about a year ago, and had been thoroughly enjoying them for at least a month when I discovered that one of them had a phase issue. Here's what happened...

I was reading another audio forum and came upon the subject of using test records/CDs to evaluate stereo systems and or speakers. Out of curiosity, I found The Best of Chesky Classics & Jazz Audiophile Test Disc Volume 3 on Tidal and gave it a listen. To my surprise, Track 28 - Natural Stereo Imaging, and Track 29 - Artificial Stereo Imaging indicated there was a problem with my right speaker. The imaging would not extend past the right edge of the speaker. But with the left speaker, the imaging extended well past the left edge. In other words, the soundstage from my DQ-10s was being cut off drastically on the right side.?

After several hours of trying to figure out what the heck was going on, I discovered that the tweeter and super tweeter on the right speaker were wired out of phase. It was an easy fix, and now my DQ-10s sound the way they should... a soundstage that extends nearly 90 degrees to the left and to the right.?

Funny thing is, I was under the impression that my speakers were factory mirror imaged (SN 47247 and 47295), and the wiring appeared to be untouched. Anyway, I was clueless that there was a problem until I listened to that test CD.?

Disclaimer: I have no affiliation with Chesky recordings...




 

When it comes to matters like these, it would be great if members could snap photos (maybe before and after a mod, or just to clarify what component they are referring to) and post them. Much easier for some of us to follow along and benefit from the discussion.
 

The out-of-phase connection of Woofer-mid, midrange and superTW can already be seen on the crossover diagram. However, since the latter is not usual, I wanted to confirm by simulating the crossover with the Simetrix/Simplis Elements software (see simulated diagram).
The phase values calculated with the simulation are visible in the double graph (overall crossover phase at the top, individual speaker phases at the bottom). In the graph below you can see that around 20 Hz there are three curves that are close to the 0 degree value. The red one is the woofer phase, the green one is the woofer-mid phase and the purple one is the midrange phase. As you can see, the green curve (woofer-mid) takes on the value 180° (out of phase) at around 1500 Hz and remains around this value (+/-30°) in the 500-7000 Hz range. The purple curve (midrange) takes on the value 180° (out of phase) at around 2900 Hz and remains around this value (+/-30°) in the 1700-4000 Hz range. The blue curve (superTW) tends to the value 180° ( out of phase) at high frequencies.
Instead, the phase of the woofer (red curve) tends to 0 degrees (in phase) at low frequencies and that of the tweeter (second green curve) tends to 0 degrees (in phase) at high frequencies.
In conclusion, woofer and tweeter are in phase, while woofer-mid, midrange and superTW are out of phase.
Merry Christmas to all!
Franco

Crossover.jpg
Dahlquist DQ-10 crossover simulation by Elements.pdf
Dahlquist DQ-10 crossover simulation by Elements.pdf
Dahlquist DQ-10 crossover phase.pdf
Dahlquist DQ-10 crossover phase.pdf

 
Edited

Here are a few pictures to (hopefully) help illustrate the phase issue I encountered with my DQ-10s. The first photo is the rear of the Left Speaker. This is the untouched factory wiring as far as I can tell. Note that the tweeter (upper left side of the photo) has the + and - terminals on the inward side (or middle) of the speaker.?



So when the speakers get mirror imaged, the tweeter on the Right Speaker gets moved from one side of the array to the other, and in doing so, it has to be rotated 180 degrees so that the terminals are in the center of the array. Like this:

The Left Speaker Tweeter:


Right Speaker Tweeter after rotating:


But notice how the the + and - terminals are now reversed. Yet on MY right speaker, the connections were exactly the same as shown on the Left Speaker, meaning they were no longer phased the same as each other. To correct the phasing, and make them identical I had to rewire the tweeter and super tweeter.?

My Right Speaker... now that it is connected properly. All drivers are now wired according to the crossover schematic



Perhaps the wiring in my right DQ-10 was just an anomaly due to the fact that the serial numbers are not consecutive. And had I not listened to the test CD, I would probably never have known there was an issue. After all, they sounded fine the way they were. I just didn't know what I was missing!

Now that the wiring has been fixed on the right speaker, I know that I'm hearing the soundstage as it was intended?to be heard by the recording engineer.
 

Guida - Great analysis!!
 

Merry Christmas and Happy New Year.? Thanks for the analysis.? May your ESR remain low and your surrounds remain flexible.? Cheers!

On Sat, Dec 23, 2023 at 12:28?PM SpongeSound <wm.edmiston@...> wrote:
Guida - Great analysis!!

 

Did you all have a good Christmas? I hope so. For completeness I am sending you, in addition to the phase graphs that I already sent you a few days ago, also the amplitude graph. The latter is quite in agreement with the amplitude measurements made in the seventies on the connections of the transducers by an Italian magazine called "Audiovisione" (see attachment). You can find the curves relating to these measurements in figure no. 10 on page 58 of the magazine. Both my simulation and the measurements are influenced by the attenuations that the crossover performs (also due to the different sensitivities) on some transducers.
In the same review you can also see that the measurements in the time domain confirmed the correct alignment of the emission centers of the transducers (photos 10 to 13) but not the absence of phase shifts (photos 3 to 5, in which you can see that the square wave is significantly deformed by phase shifts).
Have a good day everyone
Franco

Dahlquist DQ-10 crossover_magnitude_R9_R12.pdf
Dahlquist DQ-10 crossover_magnitude_R9_R12.pdf
dahlquist-dq-10_Audiovisione.pdf
dahlquist-dq-10_Audiovisione.pdf

1 - 11 of 11