This should be interesting [DEQX by two]

[Mr. Widget]

Once these correction filters are applied the speaker produces a virtually amplitude and phase perfect response, at least at the mic position.

Yes, and therein lies the rub.

I and other friends have found that with large horn systems it hasn't been possible to get a really good set of correction filters due to room interaction. I have chosen not to use any correction filters and I do all of my corrections manually... with a little help from my pal CLIO. The results are quite impressive. When used properly the DEQX PDC2.6 is a great tool... but as Steve said it is a bear... especially at first.


Widget

[boputnam]

Hi, Steve...

Thanks - that's what I thought, and figured it must be so.

But then, those response/correction curves of Ken's have some extremely hi Q anomalies. I wonder what was going on - never seen that sort of response...

[Ken Pachkowsky]

But then, those response/correction curves of Ken's have some extremely hi Q anomalies. I wonder what was going on - never seen that sort of response...

Bo

The curves I posted are from the EQ window not the Filter window. There is a huge difference. The correction filter curves are pretty flat and smoothed out. I will post an example later on.

Ken

[Ken Pachkowsky]

In your opinion, have I moved the boundry line on this reading to the proper spot to get the best analysis?

Ken

[boputnam]

The curves I posted are from the EQ window not the Filter window. There is a huge difference. The correction filter curves are pretty flat and smoothed out. I will post an example later on.

Yea, I would expect that - that is what DEQX is 'sposed to do. But I remain transfixed on the response curve - I've never seen something like that. Curious...

[Ken Pachkowsky]

Bo

After applying correction filters this is the frequency response.

[Ken Pachkowsky]

What, your not going to call me on that

Ken

[Mr. Widget]

Yea, I would expect that - that is what DEQX is 'sposed to do. But I remain transfixed on the response curve - I've never seen something like that. Curious...

What's bugging you Bo? Is it the dip I circled? That is likely a cancellation due to the distance of the mic to speaker to floor ratio... it is not all that uncommon to see something like that. I would guess that moving the mic up and down or back and forth will move it/remove it...


Widget

[Mr. Widget]

In your opinion, have I moved the boundry line on this reading to the proper spot to get the best analysis?

Yes, but it is unfortunate that your first reflection is so soon... you really won't be getting much useful data below the midrange.

This is a graphical example of what I was talking about when I said that you won't be able to create a good correction filter. If you took the speaker outside and pointed it up in the air with the mic suspended over it, you would get a much better correction filter.

I realize that DEQX is giving you a ruler flat curve, but that is BS... it is simply subtracting the correction from the source... of course it will be flat... since the real curve isn't what you are measuring, the results are??? If you did this procedure with a small speaker elevated off the floor and a good distance from any objects you would be able to really get it to work... I don't think I am making this any clearer am I... if you read, Joe D'Appolito's book on measurement techniques it will make sense... he is way better at describing what is going on.


Widget

[Steve Schell]

Widget, I agree that your dip around 160Hz. is probably a cancellation from the floor bounce. If the additional distance to the floor and back to the mic is about 3.5 feet it would be 180 degrees out of phase at 160Hz. I have seen this many times when making LMS measurements.

I also agree that making in room ungated measurements below 300Hz. or so is pretty much useless, as the room modes take over. I spoke with DEQX founder Kim Ryrie and one of his engineers about this, and they agreed with your suggestion of the outdoor measurement. They recommended placing our horns pointing up at 45 degrees and the mic 18" out from each horn on axis to make the measurements.

One problem that we've had is in setting the window on the impulse measurement. It overlays the plots of the midrange/high frequency horn and midbass horn, the latter having a 6 1/2' longer path length. There are so many closely timed reflections in the bass plot that it is difficult to know where to set the right border. I suppose if I hauled the horns outside then this might not be so much of a problem.

[Shane Shuster]

This is a graphical example of what I was talking about when I said that you won't be able to create a good correction filter. If you took the speaker outside and pointed it up in the air with the mic suspended over it, you would get a much better correction filter.
Widget

But then it wouldn't be good when you placed the speaker back in the room? Or does a DEQX not work like a TACT?

[Steve Schell]

Shane, I am not familiar with the TACT, but the room is going to screw with the speaker's response no matter what you do. Mostly large peaks and valleys in the bass are created by the room modes, and vary by location in the room. The best that can be done is to have the DEQX (or whatever unit) measure the speaker in an environment that is free of the room problems. The room will still introduce problems when the speakers are brought back in, but at least the room will not influence the measurements.

Having said all that, I've wondered what the results would be if all DEQX measurements were taken with the mic in the listening chair. Perfect amplitude, phase and arrival time in that spot might be great, although the next seat over could be awful.

[boputnam]

I also agree that making in room ungated measurements below 300Hz. or so is pretty much useless, as the room modes take over.

Bingo.

[johnaec]

After applying correction filters this is the frequency response.

Are you saying that after correction the frequency response of the speaker via calibrated mic is actually that flat?? I truly can't believe that.

What, your not going to call me on that

Or are you pulling our leg?

Edit:

I realize that DEQX is giving you a ruler flat curve, but that is BS... it is simply subtracting the correction from the source... of course it will be flat... since the real curve isn't what you are measuring, the results are???

Oh - I see...now I understand - that's the theoretical response.

John

[Ian Mackenzie]

Ken,

Have you been able focus on getting DEQX to emulate exactly what the Westlake active crossover does as far as real voltage drives are concerned?

There appears to be a notch at 4.5Khertz and at 6.5khertz? They maybe crossover point related.