Umik2 falls at 14-15kHz with REW

[Robert Johansson]

Hello all!

Has anybody got a clue what it is that might be wrong if my measurements always drops too early? They are never straight up too 20kHz. I have tested with many different tweeters that i know is linear to above 20kHz but in REW with calibrated Umik2 and sample set at the same (48kHz but tried different) on both microphone and REW and D/A converter (optical out from macbook to Hegel D/A)
I have tested different amps and cables as well. Can it be a faulty calibration file? Many thanks for any help in advance and sorry if this question has come up before.

 

[John Mulcahy]

Have you placed the mic directly in front of the tweeter for a near field measurement?

 

[Robert Johansson]

Yes, i have tried that too. Still falls too early.

 

[thothsong]

You're using the 0-degree cal file for the near field measurement, mic pointed at the tweeter? What does that cal file look like, loaded into REW?

 

[Robert Johansson]

Yes, 0 degrees.
This measurement was nearfield, as close as i could to the tweeter.

The dip is caused by the midrange being out of phase with the tweeter.

 

[thothsong]

Go to the SPL & Phase graph for that measurement, select Mic/Meter Cal, and post a screenshot. Or post an mdat file instead.

 

[Robert Johansson]

Here it is.

 

[Robert Johansson]

Better resolution here. Same curve.

 

[thothsong]

FWIW, here's the miniDSP cal file for my UMIK-2, trying to match your scale. Mine's down 3 dB at 20 kHz.

Then again, I don't really know how much to trust it. Here's a comparison, for the same UMIK-2, between the miniDSP supplied cal file and the cal file done by Cross-Spectrum Labs:

 

[Robert Johansson]

Thanks for your replies, so you think the calibration file i use is faulty?
The one that miniDSP provides?

 

[thothsong]

I don't know what to trust, sorry Cal files are the weak link in the chain, from my perspective. But don't get too excited; at least with my UMIK-2, switching from the miniDSP cal file to the CSL cal file would pull the high frequency SPL down even further ...

 

[Robert Johansson]

Okay, something is obviously wrong though. I have tested multiple tweeters and ALL of them drops significantly around 15kHz. And i know they dont behave that way "for real"

 

[thothsong]

I don't know anything about Macs; if you look at the system sound input settings for the mic, is there any audio processing enabled (like Use Ambient Noise Reduction) that should be disabled?

 

[Robert Johansson]

I will look if i can find any settings that might affect the results, thank you for trying to help me.

 

[sam_adams]

The .cal file contains data that is the magnitude difference from a flat frequency response. It is applied inversely to REW measurements:

Make certain that you have not loaded the mic .cal file as a calibration file for your sound device. Make sure that any sound device .cal file was made with the same sample rate and bit depth that you are currently measuring with. Also, check to make absolutely certain that your sample rates and bit depths are the same for the sound device and the mic in AudioMIDISetup.app and in REW.

 

[teyhyrh4r]

if the cal file is falling it will actually rise the FR. it's not the cal.
I personaly have similar roloffs all the time, but I don't care for anything after 16Khz

 

[Robert Johansson]

I constructed a wider baffle and did more measurements, looks better now.
Near field it now looks okay.
Thanks.

 

[Rolledmyown]

Hi folks,
I find this an interesting discussion, as I was about to post a similar question. I'm doing some tweaks to my home-made speakers. A number of years ago I bought a pair of Scan Speak D2905/930000 tweeters, which were reputed to be pretty flat in their response well beyond 25K. When I installed them I found that my Behringer 8024 Ultracurve Pro couldn't perform an GEQ on those tweeters at 20K, at which point their output was dropping like a stone. I still have those tweeters and am now using REW to attempt PEQ on my system (no longer with the Behringer installed, I use a PC with JRiver MC31,and a Wadia di122 decoder as DAC and preamp). The output from the tweeter still falls like a stone, as can be seen below (this is a near field, on axis measurement, with only the tweeter connected. By 22K the ciurve is almost vertical). I'm using a UMIK usb microphone, with the proprietary calibration file (with my Behringer I used similar but different microphone from the same manufacturer). Initially I thought that maybe I had done something awful with my home-made crossovers (e.g., a bad impedance compensation or too much padding) but following the thread here, maybe all is well and what I'm also seeing is a measurement error and I don't have to get my soldering iron out!
I've included below the manufacturer's specs for the tweeter. I'm old enough that I can't hear well above 15K so I'm sure I'm not 'missing out' when it comes to the reproduction of my speakers, but I certainly want to know if I've made a mistake in installation of my tweeters. There is plenty of baffle surrounding the tweeter, at least 4" each side and above it.