Neumann KH120 - tweeter issue or normal unit difference?

[Lilith]

Hi,

I bought a pair of new KH120 last week and did some measurements today. First I measured them at the position where they will go to, but after finding a strange issue with the tweeter I measured both from the same location and reproduced it. As you can see in the plot from REW with 1/24 smoothing below the right speaker in red has a pronounced +3 dB peak at around 15.5 kHz which is not present for the left speaker. I did several measurments and also slightly (< 1cm) moved the mic. The mic was orientated towards the tweeter in a distance of ~10 cm.

Is this within the tolerance or is there something seriously wrong?

 

[mkt]

Any issues in the distortion measurements? I had a used speaker issue and the distortion was a clear sign.

 

[Lilith]

Any issues in the distortion measurements? I had a used speaker issue and the distortion was a clear sign.

Not sure. The second harmonic does not reach into that range. There's something in the noise floor, which is then noise or hiss from the speaker itself? The peak at 21 kHz seems to be normal.

 

[Lilith]

The rising slope comes from the mic. With mic calibration applied it looks like this:

Right red, Orange left

 

[ernestcarl]

Well, actually, the differences are too small, IMO, to bother exchanging/returning.

Some of that difference can be easily compensated for with fine EQ, but even then, I doubt it will be all that audible.

With the distortion, I would check with much higher level sweeps. Again, differences are normal, but as long as there are no relatively large deviations, I would not worry.

 

[Lilith]

Well, actually, the differences are too small, IMO, to bother exchanging/returning.

Some of that difference can be easily compensated for with fine EQ, but even then, I doubt it will be all that audible.

With the distortion, I would check with much higher level sweeps. Again, differences are normal, but as long as there are no relatively large deviations, I would not worry.

Thanks, I looked at other measurements and that peak at 15.5 kHz is quite common for KH120 and KH80. I only hear until ~13 kHz, so for me it's not audible. Just want to make sure that they are fine.

 

[test1223]

The tolerance of Neumann speakers is usually very tight and better than most others brands.

I would quadruple check your measurements. Placing the speaker or microphone a few cm left or right can change the result with this near field measurement. It is more easy to measure it at about 1m by leaving the microphone at the same place and moving the speakers only at exactly the same position.

Can you please show the whole frequency range plot of both speakers with microphone compensation on and zoomed in spl axis?

A small transparent plastic foil acts as a diffuser in this frequency range of the cone resonance frequency of the metal tweeter. This area can be an issues in terms of mounting and manufacturing tolerances.

Even the small 0.8dB spl deviation in this not so important frequency band seems to be either a not proper measurement from your side or bad luck and two samples which are on the upper and lower side of the tolerance band.

 

[Lilith]

The tolerance of Neumann speakers is usually very tight and better than most others brands.

I would quadruple check your measurements. Placing the speaker or microphone a few cm left or right can change the result with this near field measurement. It is more easy to measure it at about 1m by leaving the microphone at the same place and moving the speakers only at exactly the same position.

Can you please show the whole frequency range plot of both speakers with microphone compensation on and zoomed in spl axis?

A small transparent plastic foil acts as a diffuser in this frequency range of the cone resonance frequency of the metal tweeter. This area can be an issues in terms of mounting and manufacturing tolerances.

Even the small 0.8dB spl deviation in this not so important frequency band seems to be either a not proper measurement from your side or bad luck and two samples which are on the upper and lower side of the tolerance band.

Are you sure that the tolerance of 1 dB also applies to 15kHz. Neumann gives a tolerenace of 0.97 dB between 100 Hz-10kHz and my pair lies in that range. There are no specs given above that range.

I reproduced that 15.5 kHz at almost the same position and on the stands at different positions. I also moved the mic slightly and only in one measurement the peak was a bit less pronounced. I didn't figure out why. So, at 15.5 kHz the deviation is ~2.5 dB

 

[test1223]

I don't know if and what tolerance Neumann guarantees. But from my experience the monitors are very good in this regard. Therefore the assumption that even this small divination is usually smaller.

The directivity in the highest frequencies is very complex due to the break up of the cone and the plastic pieces in front of it. Therefore even small movements of the microphone or some unexpected reflections of the microphone itself or microphone arm and such can cause a divination especially in the very near field of 10cm. If my understanding is correct and you moved the microphone and not the speaker I am sceptical if your results are that accurate.

 

[Lilith]

I don't know if and what tolerance Neumann guarantees. But from my experience the monitors are very good in this regard. Therefore the assumption that even this small divination is usually smaller.

The directivity in the highest frequencies is very complex due to the break up of the cone and the plastic pieces in front of it. Therefore even small movements of the microphone or some unexpected reflections of the microphone itself or microphone arm and such can cause a divination especially in the very near field of 10cm. If my understanding is correct and you moved the microphone and not the speaker I am sceptical if your results are that accurate.

I changed slightly the mic position between the sweeps to exclude such effects. The monitors were place at the same position and the mic as well. Yes, I found some measurements here and there was a better agreement between two units also above 10 kHz.

 

[Lilith]

I don't know if and what tolerance Neumann guarantees. But from my experience the monitors are very good in this regard. Therefore the assumption that even this small divination is usually smaller.

The directivity in the highest frequencies is very complex due to the break up of the cone and the plastic pieces in front of it. Therefore even small movements of the microphone or some unexpected reflections of the microphone itself or microphone arm and such can cause a divination especially in the very near field of 10cm. If my understanding is correct and you moved the microphone and not the speaker I am sceptical if your results are that accurate.

I called Neumann and they said it´s not possible to measure this correctly at home. 1 mm difference in the position can make a difference already. They are also measrued and testet before they go out and damage of the tweeter from transport is not possbile / very unlikely.

 

[Lilith]

All the measurements here show this dip at 15.5 kHz like my right speaker.

Neumann KH 120A - spinorama and misc measurements

Hello all, This will be a summary of (many :)) measurements I did with the Neumann KH 120A that I recently purchased. The pair were manufactured in 2013 and sport the older, IMO slightly industrial looking finish: A few notes on measurement methodology: The loudspeaker has been measured at...

www.audiosciencereview.com

The other one is more linear in that region and I see this in all measurements and it's also not very much dependent on the position of the mic. In practice it definitely won't play any role. Maybe my left speaker is an extremely good one

 

[Lilith]

Here are two units of the KH80 measured that also show a deviation at around 15 kHz if I understand correctly:

Neumann KH 80 DSP Speaker Measurements: Take Two

It is the same machine, but it wasn't the same speaker. Until the same speaker is measured with both machines it is hard not to speculate about inter-specimen differences. But it's not exactly the same machine. It's a different instance/installaction of a machine from the same product line...

www.audiosciencereview.com

 

[Lilith]

Here are two units of the KH80 measured that also show a deviation at around 15 kHz if I understand correctly:

Neumann KH 80 DSP Speaker Measurements: Take Two

It is the same machine, but it wasn't the same speaker. Until the same speaker is measured with both machines it is hard not to speculate about inter-specimen differences. But it's not exactly the same machine. It's a different instance/installaction of a machine from the same product line...

www.audiosciencereview.com

After reading the whole thread the speakers stay here. Seems that variances in that frequency range can occur and to be honest in practice it will not matter. Up to 10 kHz all is fine, so it's very unlikely that something is seriously wrong or even broken.

 

[Lilith]

The tolerance of Neumann speakers is usually very tight and better than most others brands.

I would quadruple check your measurements. Placing the speaker or microphone a few cm left or right can change the result with this near field measurement. It is more easy to measure it at about 1m by leaving the microphone at the same place and moving the speakers only at exactly the same position.

Can you please show the whole frequency range plot of both speakers with microphone compensation on and zoomed in spl axis?

A small transparent plastic foil acts as a diffuser in this frequency range of the cone resonance frequency of the metal tweeter. This area can be an issues in terms of mounting and manufacturing tolerances.

Even the small 0.8dB spl deviation in this not so important frequency band seems to be either a not proper measurement from your side or bad luck and two samples which are on the upper and lower side of the tolerance band.

Where did you see 0.8dB? Just went through the answers again

 

[test1223]

Approximately the deviation excluding the resonance peak. I would equalize this deviation (if the measurements are accurately with the same distance and angle and show up under different angles) and leave the resonance peak as it is, since you need a lot of measurement to get a decent average frequency response in the resonance area to equalize it correctly.

 

[Lilith]

Approximately the deviation excluding the resonance peak. I would equalize this deviation (if the measurements are accurately with the same distance and angle and show up under different angles) and leave the resonance peak as it is, since you need a lot of measurement to get a decent average frequency response in the resonance area to equalize it correctly.

Ok, I see. I used Drc-fir and there probably is some correction. Could also be that at my listening distance this difference is masked by other issues like comb filtering from my desk. Measurement above was very close to the tweeter.

 

[test1223]

Okay I don't know what exactly drc fir is doing. Bringing both speakers fr closer together is beneficial.

Place the speaker exactly at the same position (ideally microphone 1m away and free from first reflections) and don't move the microphone and repeat it with one or two different angles. If the measurements show the same deviation I would eq it by hand.

 

[HarmonicTHD]

After reading the whole thread the speakers stay here. Seems that variances in that frequency range can occur and to be honest in practice it will not matter. Up to 10 kHz all is fine, so it's very unlikely that something is seriously wrong or even broken.

These are the final KH80 measurements. There was some hiccup in the previous measurements and Amir needed to reinvestigate. Flat as a ruler (no 15kHz peak) and in very good alignment with the Neumann values.

Neumann KH80 DSP Monitor Measurements #3

Seems that if you want to create a riot in the streets of ASR, all you have to do is show a non-ruler flat response of Neumann KH80 DSP professional monitor (speaker). Not knowing that, that is precisely what I did in my first review of KH80 DSP: This was one of the earliest speakers I...

www.audiosciencereview.com

For the KH120 I am sure Neumann also published reliable values on their website. They even might state the tolerance. I would still suspect your measurements. But if you are 100% sure than get your speakers exchanged.

 

[wwenze]

I would try averaging, by using ARTA using frequency spectrum measurement with white noise and set averaging to linear, then slowly shift the mic over a 15 or 30 degree angle. That should smooth out any noding. And if both speakers measure the same this way, I won't be worried.