• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Neumann KH120 II Monitor Review

Rate this monitor speaker:

  • 1. Poor (headless panther)

    Votes: 1 0.3%
  • 2. Not terrible (postman panther)

    Votes: 3 0.8%
  • 3. Fine (happy panther)

    Votes: 34 8.6%
  • 4. Great (golfing panther)

    Votes: 357 90.4%

  • Total voters
    395

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,803
I did some measurements to confirm the resonance at low levels. Just a quick setup, speaker at listening position, mic at 50cm, level calibrated at 1m distance. No gating - so there is all the room influence included.

My measurements look different. Not the steep resonance at 330Hz but still, there is a leftover at 400-500Hz at low levels. When going even lower in level (70dBSpl and less) THD sinks in the noise floor (of the measurement mic).
View attachment 298823

There is also a little hump in the THD % measurement, but by far not what amirm measured.
View attachment 298824

THD at 80dBSpl from H2 and H3:
View attachment 298825
I first thought about noise artefacts but it probably isn't. Not sure what's going on.


@amirm - can you repeat your 80dBSpl measurement, probably with a different speaker? What does Neumann say to your data? And can you show a logarithmic THD scale - with these low distortion speakers the linear scale lacks information. And humans don't work linear - we are logarithmic people :p

So I can not confirm the spike at 330Hz from these measurements but there is probably some small effect going on here. But at 0,2% THD I would not overestimate this effect.
Look at Neumann’s webpage. They show Distortion for 80dB, 85, 90 and 95dB. You can compare to Amir’s data
 

IamJF

Senior Member
Joined
Apr 27, 2023
Messages
453
Likes
564
Location
Austria
I also checked port resonance - it behaves extremely well.
port fr.PNG


But it also shows why I don't build reflex 2 ways - even in tis design port noises are only 10-15dB under the signal from the midrange driver.
port fr norm.PNG
 

IamJF

Senior Member
Joined
Apr 27, 2023
Messages
453
Likes
564
Location
Austria
Look at Neumann’s webpage. They show Distortion for 80dB, 85, 90 and 95dB. You can compare to Amir’s data
Here compareable to Neumanns measurements, I use an Earthworks M50.
THD Ratio in dB.png

My speakers are standing close to a wall for these measurements, this helps low frequencies. That's why it shows lower THD as Neumanns measurements.

Compare the response of the tweeter at 95dBSpl (>1,5kHz) - I would say Neumanns measurements look legit. :cool: (and yes, they use a logarithmic THD scale :p)
 

Pritaudio

Active Member
Joined
Nov 7, 2020
Messages
205
Likes
77
If the dsp allows almost ALL the bass to play through the kh750 and the rest of the spectrum on the kh120ii, it should theoretically better than just the kh150- acting as a three way.
but I doubt that it is optimised for this (allowing higher spl without distortion).
maybe the midwoofer on the kh120ii doesn’t go loud enough as the limiting factor here.
 

HQY

Active Member
Joined
Jan 16, 2023
Messages
117
Likes
145
Location
Canada
for use in a small small room and with listining distance no more than 2 meters, i would prefer 120ii with 750 rather than 150 without sub. 150 with subs could be better but cost way more than 120ii with sub.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,155
Likes
231,597
Location
Seattle Area
It would be nice if @amirm could do freq. resp. measurements vs level. @napilopez used to do that and it nicely showed effect of dynamic eq for speakers with limited bass output.
That is trivial to do with Audio Precision but surprisingly difficult with Klippel. Audio Precision is a closed loop measurement system so it is easy to setup sweeps like this. Klippel is open loop and has no method to sweep levels. You have to run independent measurements at different levels and then analyze them yourself. It can do some automated things like find the power that produces X distortion but every such module costs serious dollars!

For now you do have the frequency response and three different levels in the measurements. I will see if I can quantify the difference in them tonight.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,155
Likes
231,597
Location
Seattle Area
Is that definitely a resonance at 334Hz? It doesn't correspond to the resonances visible in the decay graph or driver components graph, and the absolute level of distortion doesn't grow at that point with SPL, except at 96 dBSPL where it's dwarfed by broader distortion that is actually higher on either side.
In the waterfall, resolution drops at lower frequencies so hard to see it there. At higher SPLs, it gets dwarfed by the non-linear bass distortion.
 

jhaider

Major Contributor
Forum Donor
Joined
Jun 5, 2016
Messages
2,781
Likes
4,439
I find the 708p output to be adequate but not end game.

One has to wonder how big your space is, how loudly you listen, or if you're judging output by perception (i.e. onset of distortion) rather than SPL. Speakers with excess clean headroom never "sound loud" in use. Just more expansive. You often don't realize how loud they are playing at full tilt until you hear a noise from a different source (e.g. a doorbell) or try to talk. Then the relative difference helps you calibrate perceptual loudness.This isn't the thread for that discussion, but see here for a comparison of the dynamic compression of 708P at high levels compared to several other speakers:

At any rate, I think it's reasonable to infer that from the data we have that, terms of clean headroom, the rank order of these four speakers ls:

708i/P >>>>>>> KH 150 > 705i/P >> KH 120 II, with KH 150 ranked above 705 due to higher headroom above 300Hz or so.

But if you don't need that much headroom - and many don't! - KH 120 II looks like a really superb choice.
 

375HP2482

Active Member
Joined
Jan 13, 2020
Messages
151
Likes
140
While we're on the subject of asking Amir to do still more tests, I'm interested in subjecting speakers to the multitone tests he uses with amplifiers. Sans the lowest octave, of course, or other restrictions implied by the manufacturer's specifications.
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
19,844
Likes
19,897
Location
Canada
While we're on the subject of asking Amir to do still more tests, I'm interested in subjecting speakers to the multitone tests he uses with amplifiers. Sans the lowest octave, of course, or other restrictions implied by the manufacturer's specifications.
What is that going to do? They get tone bursts in the measurement process already and that is all in the FFT I think.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,155
Likes
231,597
Location
Seattle Area
While we're on the subject of asking Amir to do still more tests, I'm interested in subjecting speakers to the multitone tests he uses with amplifiers. Sans the lowest octave, of course, or other restrictions implied by the manufacturer's specifications.
I started by running such multitone tests. I could not get them to make sense or be useful:
index.php


Since every speaker has different frequency response, the magnitude of spikes is all over the place. Noise also pollutes the measurements as you see from its rise in low frequencies.

I just don't like to run measurements that I can't make sense out of or back.
 

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
3,303
Likes
3,827
I started by running such multitone tests. I could not get them to make sense or be useful:
index.php


Since every speaker has different frequency response, the magnitude of spikes is all over the place. Noise also pollutes the measurements as you see from its rise in low frequencies.

I just don't like to run measurements that I can't make sense out of or back.
Indeed, hard to make a lot of sense with all that "pollution", but perhaps, and I'm asking not saying, shows how relatively unimportant are measurements of electronics in the grand scheme of things if that's what you get at the output? Assuming that a measurement microphone is as, or more accurate than the human ear, which is not necessarily a given.
 

thecheapseats

Addicted to Fun and Learning
Forum Donor
Joined
Feb 22, 2023
Messages
727
Likes
775
Location
Los Angeles refugee
This is a review, detailed measurements and listening tests/EQ of Neumann KH120 MKII DSP active monitor (speaker). It is on kind loan from the company and costs US $999 each.
View attachment 298552
I love the form factor for desktop use. It is not too deep or unwieldly as some larger configurations can be:
View attachment 298553

Back panel shows the settings as used for testing:
View attachment 298554

One nit: while having power and XLR inputs vertically inserted allows for compact placement against the wall and such, it is a bit of a pain to get the cables to go in and stay properly. It is a one time nuisance fortunately. Here are the high level specs:
  • High-precision drivers, Mathematically Modelled Dispersion (MMD) waveguide
  • Linear frequency response 44 Hz to 21 kHz (±3 dB)
  • Linear phase response 120 Hz to 16 kHz (±45°) with latency <2.6 ms
  • DSP engine allows for room adaptive calibration via Neumann’s MA 1
  • Razor-sharp imaging due to extremely low tolerances (±0.5 dB)
  • May be combined with other KH line speakers in immersive/surround setups

Neumann KH120 II Speaker Measurements
As usual we start with our anechoic frequency response generated with Klippel Near-field Scanner:
View attachment 298555
I expect precision from Neumann but I am always amazed how incredibly flat on-axis response is from their speakers. I mean this is the level of flatness we get out of electronics, not something electromechanical! Despite having just a 5.5 inch woofer, the speaker dares to go nearly flat to 50 Hz! Directivity is excellent other than the woofer slightly beaming before tweeter takes over.

Front ported speakers always worry me as far as amount of box/port resonances they can project. But in capable hands of Neumann engineers, this is a non-issue:
View attachment 298556

Our modeling of early reflections is for far field listening but they still provide good insight as to how good the off-axis response is for near-field studio monitors:

View attachment 298557
You get excellent response other than unavoidable vertical dip which can be managed as noted. Perceptually it is not very important though.

Predicted in-room response has the same caveat but nevertheless, is almost textbook perfect:
View attachment 298558

Due to its smaller size, I decided to go down to 80 dBSPL and measure distortion:
View attachment 298559
There is a resonance at 334 Hz or so. This is seen as a tiny blip in on-axis response as well. It is well below my threshold though so not an issue. And at any rate, is overwhelmed at higher SPLs I normally show:

View attachment 298560

This is amazing level of performance. Look ag the gap between measured distortion and our 50 dB target. It is massive. At 96 dBSPL we do get more distortion but it is where the speaker response drops rapidly anyway:
View attachment 298561
As noted, the warning light came on in red although I didn't detect any sudden rise in distortion. Still, this is probably a good limit for the performance of this speaker.

Horizontal beamwidth/directivity shows excellence in design:
View attachment 298562

View attachment 298563

2-way non-coaxial speakers show the classic dip off-axis so best to stick to reference axis (upper ring of the woofer):
View attachment 298564

Waterfall shows the slight resonance I detected with other tests:
View attachment 298565

But look at how clean the back of the chart is.

I don't normally comment on step response but I had to make an exception on how idealized it is here:
View attachment 298566

Neumann KH120 MKII Listening Tests
Going into this review, I expected to focus on tonality as I always do with my female reference/test tracks. But what I immediately noticed was the warmth of the bass out of this little speaker! Even on a track that doesn't show case this, there was nicely present low frequency response. And it wasn't just the ears that detected it. Low notes were accompanied by cool puffs of air landing on my nose! This was at 1+ meter/4 feet which again, is impressive for such a small speaker.

Tonality was of course was right on the money and a joy to experience.

An issue I often I have with active speakers is lack of power. Not here. Despite only testing one speaker, I could get to very impressive levels with no hint of distortion above sub-bass (see below). As noted there was impressive bass that at times started to even resonate into the floor of my room! It is hard to rationalize your eye seeing a small speaker and your body saying, "man that is impressive bass!"

Throwing my torture track for sub-bass response did result in distorted bass. I took a shot at dialing out the extreme lows with a simple high-pass filter and it nicely did the trick:

View attachment 298567

At 35 Hz corner frequency, there was just a hint of distortion in sub-bass but no impact on tonality. If I increased that to 40 Hz, all distortion would vanish but there was tiny, tiny reduction in bass. Given that this is a DSP speaker, I imagine you could just dial this into the speaker.

Conclusions
Neumann once again shows that it takes its commitment to neutrality in professional monitor to the extreme. Once again, we see essentially ruler flat on-axis response and near perfect off-axis. Importantly, distortion is kept to very low level while pumping out a ton of bass relative to the size of the speaker. Combine this with a very compact form factor and you have one heck of a monitor -- for professional or consumer use. I have no need for another set of speakers but keep thinking I should try to buy this from Neumann! :)

It is my pleasure to recommend the Neumann KH120 II. You owe it to yourself to get a reference quality monitor like this and hear what your source material is supposed to sound like!

---------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
great review Amir - thanks... good to know they are still making great monitors... first pair of k+h/neumanns I purchased was forty-two years ago last month - a fine lineage...
 

beefkabob

Major Contributor
Forum Donor
Joined
Apr 18, 2019
Messages
1,619
Likes
2,044
One has to wonder how big your space is, how loudly you listen, or if you're judging output by perception (i.e. onset of distortion) rather than SPL. Speakers with excess clean headroom never "sound loud" in use. Just more expansive. You often don't realize how loud they are playing at full tilt until you hear a noise from a different source (e.g. a doorbell) or try to talk. Then the relative difference helps you calibrate perceptual loudness.This isn't the thread for that discussion, but see here for a comparison of the dynamic compression of 708P at high levels compared to several other speakers:

At any rate, I think it's reasonable to infer that from the data we have that, terms of clean headroom, the rank order of these four speakers ls:

708i/P >>>>>>> KH 150 > 705i/P >> KH 120 II, with KH 150 ranked above 705 due to higher headroom above 300Hz or so.

But if you don't need that much headroom - and many don't! - KH 120 II looks like a really superb choice.
I don't always listen loud, but when I do... The 708p start to fall apart as the volume gets near -0. The 705p will happily clip on bass at high volumes and don't even get close to as loud. Every once in a while, I wish I had the M2 or equivalent.

I love my 708p. I adore the 705p. If I were buying new right now, I'm not sure what I'd get. Certainly nothing is better than the 708p by enough of a margin to justify replacing them unless I went with the M2 or equivalent. These Alfred E. monitors are so fantastic but... the SPL?
 

375HP2482

Active Member
Joined
Jan 13, 2020
Messages
151
Likes
140
I started by running such multitone tests. I could not get them to make sense or be useful:
index.php


Since every speaker has different frequency response, the magnitude of spikes is all over the place. Noise also pollutes the measurements as you see from its rise in low frequencies.

I just don't like to run measurements that I can't make sense out of or back.
I'm trying to make sense of it myself.

Do the notations "-10 dB" mean that the 70 dB peaks in the 100-200 Hz region are actually 80 dB?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,155
Likes
231,597
Location
Seattle Area
Do the notations "-10 dB" mean that the 70 dB peaks in the 100-200 Hz region are actually 80 dB?
Unlike my Klippel measurements, these are not level calibrated. So you just want to use the highest peak as a reference and then look to see how low the distortion is relative to that.
 

hmt

Senior Member
Joined
Apr 11, 2020
Messages
384
Likes
462
If you have a sub and a little EQ, the JBL Stage A-130 will fill the bill as well:

Neumann KH 120 II
SCORE: 6.7
SCORE w/ sub: 8.5


JBL Stage A-130
EQ Score: 6.29
with sub: 8.54


The Neumann score should also indicate "EQ Score" as it likely uses internal EQ and steep DSP crossover.
Not really since the FR of the A-130 indicates resonances. This is one reason why the score is calculated like that. The KH 120II does not show resonances and as can be seen with the KH120 it is already very flat without EQ.
 

IamJF

Senior Member
Joined
Apr 27, 2023
Messages
453
Likes
564
Location
Austria
I started by running such multitone tests. I could not get them to make sense or be useful:

Since every speaker has different frequency response, the magnitude of spikes is all over the place. Noise also pollutes the measurements as you see from its rise in low frequencies.

I just don't like to run measurements that I can't make sense out of or back.
That's EXACTLY why I also don't do these tests. It makes no sense to do measurements where the results are unclear. When you do a lot of acoustic measurements you know and experience that - more often as you like. (I threw away a day of absorber measurements - the results simply didn't show the absorber but the room.)
And IMD scales with THD - there is a reason why Purify have BOTH very low ;)

@amirm you could do a linearisation with AP. Save the frequency response, invert and use as EQ. Better from 80/100Hz upwards to not kill speakers. Long test time to get the noise down, but you would need a very low reflection environment. Hmmmm - I actually could try that in my room .... but first back to payed work ...
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,155
Likes
231,597
Location
Seattle Area
@amirm you could do a linearisation with AP.
I don't use AP for any speaker measurements. I did use it for multitone and a few other tests early on but no more.
 
Top Bottom