• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). 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!

Eltax Monitor III MK2 / measurements and review

Nuyes

Member
Forum Donor
Reviewer
Joined
Jun 8, 2022
Messages
87
Likes
1,113
Location
South Korea
00.jpeg


This is a measurement and review of the Elax Monitor III MK2.

Here we go.






I measure and review speakers using DIY turntable, Earthworks M30 mic, and Klippel Distortion Analyzer 2.


First, the impedance response.

01.png02.png


It was measured at 0.1V and 2.83V, respectively.



Next, SPL/FR and near-field measurements from the front.
(I follow the manual of the Klippel document when measuring ported speakers.)



1/12 Smoothing.

03.png
04.png




There is something unusual about the nearfield measurement data.
No filter was found on the woofer unit at the crossover point between woofer and twitter.





Next, the directivity.
I use DIY turntable to measure orientation, and I use 5ms window, so the data is meaningful after 200Hz.


Horizontal Directivity.

05.png
06.png






Vertical Directivity
07.png
08.png







Beamwidth
09.png
10.png






Polar plot

11.png
12.png




Next, a nonlinear distortion measurement.

I use CHD (Characteristic Harmonic Distortion) to display THD-relative because it is impossible to measure an odorless room.

It calculates the proportion of the harmonic distortion component based on the average sound pressure in a predetermined frequency band (possibly within the critical distance).
Using this method, the In-room measurement can also obtain relative data that is not affected by the fundamental's dip and peaks.



13.png
14.png
15.png
16.png






Next, the multi-tone test.

I've originally included a multi-tone test in all my reviews.
And this is even more disadvantageous for a single unit with a wide lower bound.

So, from this time on, we decided to do additional measurements that limit the range to '80Hz - 20khz'.
(I was inspired by Erin's Audio Corner.)


17.png
18.png
19.png
20.png




Strong multi-tone distortion was found near about 5.5khz where the dip occurred.

And overall, the level of multi-tone distortion is high.
(approximately -35 dB)





The following is the multi-tone distortion for each sound pressure level.

21.png


The 5.5 khz multi-tone distortion noted earlier has less movement than the increase in sound pressure levels.

Something strange is happening.




Next, the compression test.
(I used a multi-tone, not a sine sweep.)

22.png
23.png




I am careful to make subjective evaluations of speakers other than data.
The speaker is currently on sale at a Korean store for about $92 a.m.
(I think it's a joint purchase.)


What do you think?
Please feel free to share your opinions.
 

Curvature

Senior Member
Joined
May 20, 2022
Messages
326
Likes
375
the compression test
Thank you for posting your measurements here. I like the detail and variety of data.

Could you please provide a few reasons or speculations about why compression tests show both positive (expansion) and negative (compression) deviations from flat? I do not understand what mechanisms cause expansion.
 

test1223

Senior Member
Joined
Jan 10, 2020
Messages
309
Likes
319
I do not understand what mechanisms cause expansion
The internal friction of materials can have a similar behavior as static vs. kinetic friction, where a higher starting energy is needed.

The spider of a woofer can have a degressive force vs. extension part in its curve.

The electro magnetic motor system can be of center or behave in another unwanted none linear way.
 
OP
Nuyes

Nuyes

Member
Forum Donor
Reviewer
Joined
Jun 8, 2022
Messages
87
Likes
1,113
Location
South Korea
Thank you for posting your measurements here. I like the detail and variety of data.

Could you please provide a few reasons or speculations about why compression tests show both positive (expansion) and negative (compression) deviations from flat? I do not understand what mechanisms cause expansion.
Positive direction is compression.
For example, if you increase the output by 10 dB and actually increase by only 9 dB, 1 dB of data is displayed in a positive.
 

Curvature

Senior Member
Joined
May 20, 2022
Messages
326
Likes
375
Positive direction is compression.
For example, if you increase the output by 10 dB and actually increase by only 9 dB, 1 dB of data is displayed in a positive.
So the chart is the opposite of Erin's here for example: https://www.erinsaudiocorner.com/loudspeakers/klipsch_404cii/
The internal friction of materials can have a similar behavior as static vs. kinetic friction, where a higher starting energy is needed.

The spider of a woofer can have a degressive force vs. extension part in its curve.

The electro magnetic motor system can be of center or behave in another unwanted none linear way.
I have trouble understanding the explanation. Can you give some examples of specific issues leading to expansion? Perhaps using measured results?

In @Nuyes measurements above the 5.5khz resonance stands out in every measurement and leads to expansion. What specific issue causes it? Should we understand high Q resonances to be additionally problematic because on top of all the other recognizable issues they present they also playback nonlinearily in an expansion direction? I apologize for poor English here.

For compression, thermal issues seem to be the most common reason as I understand it.
 
OP
Nuyes

Nuyes

Member
Forum Donor
Reviewer
Joined
Jun 8, 2022
Messages
87
Likes
1,113
Location
South Korea
So the chart is the opposite of Erin's here for example: https://www.erinsaudiocorner.com/loudspeakers/klipsch_404cii/

I have trouble understanding the explanation. Can you give some examples of specific issues leading to expansion? Perhaps using measured results?

In @Nuyes measurements above the 5.5khz resonance stands out in every measurement and leads to expansion. What specific issue causes it? Should we understand high Q resonances to be additionally problematic because on top of all the other recognizable issues they present they also playback nonlinearily in an expansion direction? I apologize for poor English here.

For compression, thermal issues seem to be the most common reason as I understand it.
Temperature is definitely one of the major contributors to compression.
But I did the compression measurement for 4 seconds, which will not be enough to increase the heat of the driver.

Due to limitations such as the temperature of the voice coil and the physical properties of the spider and edge, the speaker has power limitations.

And this compression test will be the first to reveal in this compression test.

But in addition, let's take a specific resonance as you mentioned earlier.

Resonance, at a lower sound pressure level, produces a dip by interrupting (or canceling) the output sufficiently, but as the sound pressure level increases, the effect decreases and the dip may be filled.
(This is also found in multi-tone distortion data.)
 
OP
Nuyes

Nuyes

Member
Forum Donor
Reviewer
Joined
Jun 8, 2022
Messages
87
Likes
1,113
Location
South Korea

Curvature

Senior Member
Joined
May 20, 2022
Messages
326
Likes
375
I think Erin uses Sine-sweep and provides plots using software other than Klippel dB-Lab.
This is because Erin's compression data is somewhat different from his other Klippel data in the font and legend, and Klippel indicates cupfreshing in a positive direction.
Thank you again. I appreciate your explanation above as well.

Is your compression data anechoic?

It seems that speaker nonlinearity also includes frequency response vs. level. Perhaps this is an underappreciated area by the community. It suggests that directivity may vary somewhat with level.
 
OP
Nuyes

Nuyes

Member
Forum Donor
Reviewer
Joined
Jun 8, 2022
Messages
87
Likes
1,113
Location
South Korea
Thank you again. I appreciate your explanation above as well.

Is your compression data anechoic?

It seems that speaker nonlinearity also includes frequency response vs. level. Perhaps this is an underappreciated area by the community. It suggests that directivity may vary somewhat with level.
The compression test was conducted in a space with reverberation.

BD66F7F1-25C6-406B-8F6E-E6DC3F7C0504.jpeg

(The measured volume of the space is 37.65 m^3, and RT60 from 100hz remains below 200 ms.)

However, the effects of space increase linearly with respect to the sound pressure level (up to the wall property limit)

Therefore, if there is no change in the position of the speakers and microphones, comparisons can be made relatively simple.

Of course, this cannot be free from the influence of powerful dip depending on the measurement position of the speaker and microphone, so I limit the frequency display range (50hz-20khz)


I agree with what you said.
The speakers affect FR/SPL in any way beyond the 'Linearity' area.

Therefore, FR/SPL data must be presented with the measured SPL
 

test1223

Senior Member
Joined
Jan 10, 2020
Messages
309
Likes
319
Temperature is definitely one of the major contributors to compression.
But I did the compression measurement for 4 seconds, which will not be enough to increase the heat of the driver.

Due to limitations such as the temperature of the voice coil and the physical properties of the spider and edge, the speaker has power limitations.

And this compression test will be the first to reveal in this compression test.
Yes temperature issues which cause compression in the electro magnetic motor are more important in the higher frequencies (low excursion for the motor). These issues can only be fully shown with longer test, since they need a head up time. And therefore the compression in itself has a very long signal dependency which is not often looked at from the signal processing perspective and more typical present in control theory.

The electro magnetic motor and a mostly progressive spider cause most of the compression in the lower frequencies, since higher excursion of the cone is needed.
 
Top Bottom