Excessive Variation in Measurement While Calibrating Soundcard

trondeez

Registered
Thread Starter
Joined
Aug 13, 2019
Messages
4
Hello...

I am getting some strange results while trying to calibrate my soundcard in REW. I have calibrated this soundcard in the past with success; But it has been awhile and I just moved into a new studio. I want to make sure the card is working properly and my new measurements will be as accurate as possible. I used REW to measure and guide the acoustic treatment of my last studio with fantastic results. As I started to do this for the new studio, the strange calibration results started. I have attached an mdat of one of the wonky measurements as well as the good original calibration mdat.

In looking for an answer to my problem, I found some threads pointing to the sample rates for ins vs outs getting switched somehow. This looks like a good candidate for what is happening in my situation as well.

I have tried to remedy the problem through the following:

Uninstalled / reinstalled ASIO drivers and the soundcard software.
Unistalled / reinstalled REW to the latest version. (I was previously using version 5.19 now using 5.20 beta 19. Got same results in both versions. Tried all the below listed attempts to fix in both versions.)
Made sure "listen to this device" in Win recording sound properties is unchecked.
Made sure monitoring input is off. (It is a knob on the hardware.)
Made sure Windows shared mode sample rate properties were the same for recording and playback. Also tried different settings and then matched in REW. (44100, 48000,16bit and 24 bit etc.)
Tried calibrating using the Java drivers in REW.
Tried all combos of outputs and inputs for calibration.
Tried different cables.
Tried line and instrument inputs.
Tried switching soundcard to mono.
Closed all programs and disabled WiFi while calibrating.

Setup:
Win 10 Home 64bit, Asus Strix ROG laptop, i7-7700HQ 2.8GHz, 32GB Ram, NVIDIA GE Force 1050 ti, SSD for operating and programs, HDD for storage.
Soundcard is Behringer U-Phoria UMC 404HD


Any help is greatly appreciated.



Thanks
 

Attachments

  • REW Soundcard Calibration.mdat
    3.6 MB · Views: 17
  • Soundcard Wonky.mdat
    6.1 MB · Views: 12

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
It doesn't look like it is suffering from monitoring/feedback, it is just very noisy. Is it OK when used for playback? Power supply OK?

The windows settings don't matter when you use ASIO. Use a TRS - TRS cable from output to input. With the cable connected, open the REW signal generator and select a 1 kHz sine wave test signal at -12 dBFS. Don't start playing it yet. Open the RTA, set it for Spectrum, 64k FFT, No averaging, Blackman-Harris 7 window. Start the RTA, set the axis to dBFS and make sure the vertical span goes from 10 dBFS at the top and whatever you need to see the bottom of the plot, hopefully around -140 dBFS. Capture a screenshot of it with the generator off. Start the generator, leaving the RTA running, and capture another screenshot. Post those here, might give some clues.

As an unrelated aside, looks like you have the Analysis preference for IR truncation set to "Keep full IR". Best setting that to "Truncate after 1.7 s" to avoid unnecessarily large measurement files.
 

trondeez

Registered
Thread Starter
Joined
Aug 13, 2019
Messages
4
Thanks for quick response John.

I tried XLR to XLR and TRS to TRS. For the attached RTA, I used TRS. I had just switched the IR truncation because I saw someone recommend this for the same problem in another forum.

Think you are correct about the noise but it is not affecting playback. No hum / ground loops. (I would think we would see a bump around 50-60Hz in the RTA if there was electric noise?)

Please see the attached RTA screenshots. I also attached a better calibration where I have turned down the input (6db) from exactly matching the output so the soundcard doesnt show clipping. This helps but the noise is still there.

When you mentioned it looks like noise, I tried a bunch of things to see if something is causing it. I switched power supply, I went direct into the wall, I tried unplugging from wall (Soundcard can get USB power as well), I tried unplugging laptop, I switched USB cords, I switched USB ports...

The first time I switched USB ports I went from a USB 3.0 to a 2.0 port and voila, got a flat result. However, when I unplugged the soundcard from the wall, I got the noise back. I then plugged the soundcard back into the wall and the noise was still there? I then tried various combos of the 4 USB ports with different cords and plugging into the wall and out and unplugging the laptop etc... But have not been able to get a flat result since the first time I switched ports. The noise is still there.
 

Attachments

  • RTA Gen.jpg
    RTA Gen.jpg
    171.4 KB · Views: 13
  • RTA No Gen.jpg
    RTA No Gen.jpg
    171.9 KB · Views: 10
  • Soundcard Cal Better.mdat
    1.6 MB · Views: 8

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
Although that cal when you switched ports is cleaner, with the noise floor about 80 dB down rather than 40, it is still a lot noisier than your older one. It may be that the interface has developed a fault, perhaps there is a poor internal connection somewhere. You could open it up and check that all the connectors are properly seated, might help.
 

trondeez

Registered
Thread Starter
Joined
Aug 13, 2019
Messages
4
Thanks John. Appreciate the help. Really love REW. Great program and model. Cant say enough about how much the program has helped my mixing / mastering through acoustical knowledge. If you want to "hear" music, you absolutley need REW and some acoustic treatment.

I have an Asurion plan on the soundcard so I just filed the claim and they are sending a new one. :)

And let that be a lesson in protection plans folks... They work, I have claimed many things over the years and have never had a problem with refund or replacement. Dont believe the negative. Insurance does work.
 
Top Bottom