Erin's Audio Corner gets a Klippel NFS!

[Mnyb]

How strong is the servo ? Play nice with the robot .

There is a reason that there are cages around industrial robots and light sensors if you should intrude anyway.

It's expensive unfortunate and embarrassing if you crash the speaker or the klippel , but your not hurt .
Be careful so you don't get whacked in the eye or something ?

 

[hardisj]

First SPIN data set is complete. This is the Kef R3. If you compare it to Amir's results you can see the trends follow very well.

Some differences I know of with my test are:
1) I know the shadow flare was pushed all the way in. I imagine Amir's was, too. I only mention this because the shadow flare has been proven to be impactful on the FR. And it explains the dip ~900Hz in my measurement. Keep in mind every other 3rd party result shows a strong dip ~1.2kHz other than Amir's. IOW, the flare effects the response and if you want to read about it further, I posted my findings on that here:
https://www.audiosciencereview.com/forum/index.php?threads/kef-r3-speaker-review.12021/post-672823

2) I also have a different mic setup. I don't know if Amir was using the cage back then. Or if he was using the boom/mic configuration he posted previously. No matter which way, though, we know there is some difference in how I set mine up vs his setup. This is not a knock against his!!!! No drama needed here. Just pointing out a difference.

3) In the menu for the setup there is an option to set the Frequency Resolution. It is not response resolution (the typical 1/3-octave, 1/24-octave, etc). It is simply the number of data points per each octave. The default is 10 as you can see below. It is possible @amirm has his set up with a higher number and thus there are more little 'squigglies' in his response. I don't think either are wrong. Just pointing this out as a potential difference for those who want to dive in to the weeds.

So, at any rate, here is the SPIN data from my measurement. I adjusted the scale so it would match what Amir had in his review. Overall, they match quite well. The differences I explained above are likely the differences you see in our results (i.e., the extra squiggles in Amir's data, the dip ~900Hz in my data, and potential HF differences due to the mic boom. I don't expect them to match perfectly because no speaker is the exact same anyway; even if the NFS would provide the exact same results on the exact same unit.

 

[tktran303]

Wow. KEF R3? Buy with confidence (for smaller rooms)

 

[NTK]

3) In the menu for the setup there is an option to set the Frequency Resolution. It is not response resolution (the typical 1/3-octave, 1/24-octave, etc). It is simply the number of data points per each octave. The default is 10 as you can see below. It is possible @amirm has his set up with a higher number and thus there are more little 'squigglies' in his response. I don't think either are wrong. Just pointing this out as a potential difference for those who want to dive in to the weeds.

Would like to point out that CTA-2034-A asks for 20 points per octave or more. Amir's measurement data files usually have ~200 frequency points, which correspond to 20 points/octave (i.e. 10 octaves from 20-20 kHz).

 

[amirm]

3) In the menu for the setup there is an option to set the Frequency Resolution. It is not response resolution (the typical 1/3-octave, 1/24-octave, etc). It is simply the number of data points per each octave.

I use 20 points/octave per Harman papers:

I switched to 10 points for a short while and it indeed produces the smoother graphs you show here. But there were some complaints so I switched back to 20 points. It has been that for many months.

 

[hardisj]

Would like to point out that CTA-2034-A asks for 20 points per octave or more. Amir's measurement data files usually have ~200 frequency points, which correspond to 20 points/octave (i.e. 10 octaves from 20-20 kHz).

View attachment 112800

Noted.

 

[hardisj]

Round two is currently underway with the updated 1/20th octave resolution.

PS: don’t judge me for the wires on the floor. I’ve got those wire cover things on the way and I still haven’t decided where I want to have the computer set up.

PPS: It is currently snowing outside, here in north Alabama. And I am sitting in my 70 degree garage testing a speaker. Man, this is awesome.

 

[richard12511]

Round two is currently underway with the updated 1/20th octave resolution.


View attachment 113010



PS: don’t judge me for the wires on the floor. I’ve got those wire cover things on the way and I still haven’t decided where I want to have the computer set up.

PPS: It is currently snowing outside, here in north Alabama. And I am sitting in my 70 degree garage testing a speaker. Man, this is awesome.

I'm guessing y'all aren't on power cycling like we are here in Texas. We're 30mins on, 30mins off here, all day yesterday, all day today, and who knows when it will stop. Power companies don't have enough power to heat all of our homes all the time. Also, no water. Makes me realize how much my life relies on power, internet, and running water. Also makes me appreciate not being born in 1690.

 

[Doodski]

I'm guessing y'all aren't on power cycling like we are here in Texas. We're 30mins on, 30mins off here, all day yesterday, all day today, and who knows when it will stop. Power companies don't have enough power to heat all of our homes all the time. Also, no water. Makes me realize how much my life relies on power, internet, and running water. Also makes me appreciate not being born in 1690.

Here it's been -28C (-19F) to -34C (-29F )nights and daytime highs are about -26C and we have had very steady electricity and gas heat. Yes, I do appreciate that very very much. > I live in central cowboy country Canada where we are compared to Texas in thinking and economy from petroleum resources.

 

[Krusty09]

Round two is currently underway with the updated 1/20th octave resolution.


View attachment 113010



PS: don’t judge me for the wires on the floor. I’ve got those wire cover things on the way and I still haven’t decided where I want to have the computer set up.

PPS: It is currently snowing outside, here in north Alabama. And I am sitting in my 70 degree garage testing a speaker. Man, this is awesome.

C'mon man. Clean up that cable mess. Get some gaffer tape and lay down the law. You have a beautiful space let's spiffy it up. We need to see some cable management.

Anyway congrats on the purchase and hard word.

Cheers

 

[Steve Dallas]

I'm guessing y'all aren't on power cycling like we are here in Texas. We're 30mins on, 30mins off here, all day yesterday, all day today, and who knows when it will stop. Power companies don't have enough power to heat all of our homes all the time. Also, no water. Makes me realize how much my life relies on power, internet, and running water. Also makes me appreciate not being born in 1690.

Hang in there!

Also in Texas and feeling very fortunate with temps down to -6F overnight. Our power blinked yesterday, but has otherwise stayed on, and the re-circulation pump has kept the pipes from freezing. We lost internet for about 3 hours this morning, but it has been solid since. The septic system's sprinklers are frozen, so we have to be careful with water usage until I get that figured out. Pouring hot salt water on the sprinkler heads did not fix it.

To summarize: the worst that has happened so far is that Dirac could not authenticate this morning, and I had to listen to too much bass. Tragic! (Also, we can't shower for a few days...)

 

[napilopez]

I'm guessing y'all aren't on power cycling like we are here in Texas. We're 30mins on, 30mins off here, all day yesterday, all day today, and who knows when it will stop. Power companies don't have enough power to heat all of our homes all the time. Also, no water. Makes me realize how much my life relies on power, internet, and running water. Also makes me appreciate not being born in 1690.

I partly grew up in the Dominican Republic and that was basically daily life into the 2000's! Although not quite such short intervals, but when the lights went out is when we'd bring out the dominoes!

 

[hardisj]

New Kef R3 measurements are complete. 1/20th octave resolution to match the standard (or, at least, the minimum it requires) and Amir's data.

Already ran the HD tests. Running the max SPL test now.

I'll make a more official post once the testing is officially done. I am still in the "make sure it makes sense" phase of the NFS. But so far, so good.

 

[amirm]

Your vertical scale is 60 dB but should be 50.

 

[hardisj]

Your vertical scale is 60 dB but should be 50.

I understand. I was trying to match yours, though, because I know people are going to compare them and complain the scales aren't the same. I'm in a lose-lose here. But for my actual review I will adjust the scale to be 50dB as all my others have been.

 

[amirm]

Oh, never mind then.

 

[amirm]

Your graph continues to be very smooth though, lacking resolution. What is the sweep period? If that is too short, then the starting resolution is too low to achieve 20 points/octave.

 

[hardisj]

I don't recall. I'm back inside. Computer is closed up in the garage. If you wanna tell me specifically what to look for I can post back tomorrow.

One difference might be your mic setup. I don't know when you took the cage off for your measurements; maybe you had the cage on for yours? I don't really want to dig that far back but it's just a thought since we already know that has an impact of about ±2dB above 5kHz and some other effects below that frequency. The response greatly smoothed out once I went with the straight-arm boom discussed earlier (and referenced again below to keep from jumping back and forth). In fact, upon second look, that's where I am seeing the greatest difference between our measurements wrt resolution. Mine is more smooth above 2kHz with less deviation. Below that, however, we both show the same sort of resolution peak/dips. So, maybe the cage is the difference here.

Edit: I searched because I didn't want you to have to bother with it and I found you mentioned using the cage in a few posts dating back to the summer in this thread.

On the other hand, there is a bit of jagginess in the midfrequencies that is caused by the microphone "cage" that I am using which doesn't show up as much in Harman's measurements. So I think we can expect some numerical differences between their data and ours.

I see where you did mention removing it in a post back in November, though.

Other refinements have been removal of the cage around the microphone. Been taking chances with my expensive microphone this way. You have to be very confident of the measurement parameters to avoid collisions with the speaker! I have had a few heart pumping close calls (not for the sake of the speaker but the mic). Anyone new with NFS should not attempt this. It is like removing the safety net below a trapeze artist.

You tested the Kef R3 in March so So it seems the cage was on for the R3 test. I'm gonna bet that is the driver for the MF/HF differences in our results. Certainly seems plausible given the below graphic.

 

[MZKM]

But for my actual review I will adjust the scale to be 50dB as all my others have been.

Don’t forget the error with the Early Reflections curve, that needs to be manually computed.

 

[jhaider]

You clearly need a Neumann KH 80 to test because that speaker is woefully under studied.