Neumann KH 80 DSP Speaker Measurements: Take Two

[amirm]

Then why the heck do you measure loudspeakers in the first place, if this difference is insignificant?

If you are having this much trouble understanding the simple argument I wrote, I don't know that i want to bother saying more. Read my post once again and Dr. Toole's book.

 

[gr-e]

The highlighted red graph is showing the "on-axis" as measured. The graphs above it are at -10 and -20 degree so they are more "correct."

Going 20 degrees above axis recommended by Neumann would be 25cm above the speaker @1m distance. If this gives the flattest response then something gone very wrong

 

[amirm]

Again, I don't know what Klippel's software allows. Mathematically, I don't think Amir needs to redo his test for this purpose. But I think the original test was also done at a too high SPL.

The explanation is given in Klippel software:

The configuration of the system specifically asks for Tweeter Axis, not reference axis. As above explains, the computation of soundfield becomes simpler if you star with tweeter as the reference axis. I think this is due to much more complex soundfield of the tweeter.

 

[amirm]

I forget who was talking about this but on balloon measurements, changing the distance does nothing. It only changes the SPL levels on the right.

 

[Thomas_A]

The explanation is given in Klippel software:
View attachment 49202

The configuration of the system specifically asks for Tweeter Axis, not reference axis. As above explains, the computation of soundfield becomes simpler if you star with tweeter as the reference axis. I think this is due to much more complex soundfield of the tweeter.

So that explains it. There second measurement does not say anything other than it is a flawed measurement and that is why the balloon points down

 

[gr-e]

The configuration of the system specifically asks for Tweeter Axis, not reference axis

When remeasuring the speaker did you set the tweeter axis at the top of the woofer?

 

[amirm]

When remeasuring the speaker did you set the tweeter axis at the top of the woofer?

For which one? The first one yes, I set it to half-way point which was roughly at the upper rim of the woofer. The second one is set to the center of the tweeter with me eyeballing it.

 

[amirm]

Going 20 degrees above axis recommended by Neumann would be 25cm above the speaker @1m distance. If this gives the flattest response then something gone very wrong

That is not what I said. What the outcome is that the tweeter axis as measured by NFS system produces flat response, not setting it to some "acoustic reference" mentioned by the manufacturer with no detail or back up as to what that term means.

 

[gr-e]

That is not what I said. What the outcome is that the tweeter axis as measured by NFS system produces flat response, not setting it to some "acoustic reference" mentioned by the manufacturer with no detail or back up as to what that term means.

Tweeter axis and "acoustic axis" mentioned by manufacturer are different things. Tweeter axis should always be perpendicular to the tweeter.
I guess when you changed it the klippel system expected the tweeter to be at the top of the woofer and that resulted in HF droop

 

[amirm]

So it is clear, the time and effort that it took to make these two measurements, cost us measuring two new speakers. The work did confirm that setting all measurements to the same axis, i.e. tweeter axis as the system asks for, is the right approach. Randomly moving that reference based on some line in a manual is not wise. It radically changes the response and generates meaningful differences in how one would think the speaker measures.

 

[amirm]

Tweeter axis and "acoustic axis" mentioned by manufacturer are different things. Tweeter axis should always be perpendicular to the tweeter.
I guess when you changed it the klippel system expected the tweeter to be at the top of the woofer and that resulted in HF droop

You need to say that to people complaining, not me.

 

[Blumlein 88]

So it is clear, the time and effort that it took to make these two measurements, cost us measuring two new speakers. Snip.....

In my opinion, in context of showing how the Klippel system can be used and trusted it was very much worth the repeat vs just adding two new speakers to the database.

 

[hardisj]

So it is clear, the time and effort that it took to make these two measurements, cost us measuring two new speakers. The work did confirm that setting all measurements to the same axis, i.e. tweeter axis as the system asks for, is the right approach. Randomly moving that reference based on some line in a manual is not wise. It radically changes the response and generates meaningful differences in how one would think the speaker measures.

But it gained you knowledge. And potentially instilled a bit more confidence in your methods here. Which is worth more than another couple measurements of some $200 bookshelf speakers. My $0.02.

 

[Pio2001]

View attachment 49202

The configuration of the system specifically asks for Tweeter Axis, not reference axis.

That explains all.
Especially the picture with the tweeter point in front of the woofer. From this point of view, if we are standing in the dashed line enclosing the speaker, we are going to see high frequencies travelling from top to bottom, and infer them to go on like that beyond that point. Which is exactly what the first balloon plot shows.

The confusion comes from the acoustic reference axis, which has nothing to do with the tweeter point.

 

[gr-e]

You need to say that to people complaining, not me.

The problem is, people (including me) thiought that you changed the reference axis i.e. the first measurement represents FR at point A and the second at point B.
But you actually changed the tweeter position in klippel configuration, which messed up the measurements. Am I correct?

 

[Blumlein 88]

Now this is a lazy question, and not a complaint. You pick a reference axis and point from which the Klippel measures 36 points of a vertical and 36 points of a horizontal circle. Each of those points being measured at two slightly different distances to allow subtracting out room effects.

Do I understand correctly the result can show the directional sound output for the entire speaker not being limited to only those measurement points?

If so then couldn't it after the fact change the reference axis without a re-measure? I'm not asking you to generate multiple reference points for each review, just wondering if others have that interest then they can do it using the measurement data you provide.

EDIT:I see from an earlier recent post, apparently the Klippel asks for tweeter position and not reference axis? Is this so, I apparently had the wrong idea of how it works. (probably need to go back and read the info on the Klippel).

 

[Sancus]

Personal note: keep this style of posting going and I will learn my lesson to never bother dealing with you all's objections. Is it too hard to ask to you to be polite and constructive?

FWIW I am happy you remeasured this, and now that I see the system measured two different samples and they're so similar I know that: a) The bass dip is almost certainly real and b) Neumann speakers are pretty consistent!

In real world use you are heavily EQing this frequency range either with the built-in EQ system(if you have an iPad... the biggest issue with this speaker, imo) or with an external one, so I don't think a ~2db bass dip is material. That said, if Neumann missed it due to their measurement technique, perhaps there's an opportunity here for them to fix it with a firmware update. I own a pair of KH80s and these reviews mostly just convince me that it's an excellent speaker as long as it's in your budget and fits your requirements(ie: size).

The tone of complaints on this forum continues to be an issue and it's important to remember nobody is entitled to explanations and re-testing for every little thing they think is wrong. Also that most of the people reading this content probably appreciate it and aren't posting negatively.

 

[DS23MAN]

That is just why every good measurement mic comes with calibration file (usually different for axial and balloon/90deg). Amir said that he didn't bother to use it...

A calibration file for free field isn't correct in nearfield measurements.

 

[Pio2001]

Now this is a lazy question, and not a complaint. You pick a reference axis and point from which the Klippel measures 36 points of a vertical and 36 points of a horizontal circle. Each of those points being measure at two slightly different distances to allow subtracting out room effects.

If I understand correctly, this is not exactly how it works.
The 36 points on two circles are the spinorama measurement in an anechoic room.
Klippel makes something like 500 measurements, everywhere around the speaker, in any room.
From this, it can generate the same curves as from a spinorama, and a lot more.

Do I understand correctly the result can show the directional sound output for the entire speaker not being limited to only those measurement points?

If so then couldn't it after the fact change the reference axis without a re-measure?

Yes, exactly. If you are 2 meters away from the speaker, and the reference axis is supposed to be 5 cm below the tweeter, then just ask the software for the response curve 1.5° down.

 

[ctrl]

Do I understand correctly the result can show the directional sound output for the entire speaker not being limited to only those measurement points?
If so then couldn't it after the fact change the reference axis without a re-measure?

Yes, exactly. If you are 2 meters away from the speaker, and the reference axis is supposed to be 5 cm below the tweeter, then just ask the software for the response curve 1.5° down.

Shouldn't it then also be possible to shift the coordinate system origin afterwards and thus use the manufacturer reference point for the evaluation?

Then the 20 kHz balloon soundfield shown by Amir, in the second measurement, would point upwards (which would be expected), but at the crossover frequency the radiation lobe should be on the axis (as the loudspeaker crossover was tuned to this reference point).