Umik-1 and J-BoxII giving different curve above 4kHz

[Phil Pope]

Hi

not sure why I am getting different measurements. Umik-1 USB mic is set up with mic calibration file, headphone output goes to mixer and then to power amp. J-box II was set up according to instructions and mic calibration file loaded, aux output connected to mixer and power amp, REW output voltage increased until curves overlaid each other. Checked the mics were both pointing directly at the centre of the horn. Speaker being measured it a large MEH. Mic position is 30cm from the front of horn. Red trace is the Umik-1, Blue is the J-Box. Why do the measurements diverge at high frequency?

 

[Wayne A. Pflughaupt]

For others checking this thread, the OP is not talking about the stand-alone karaoke system with the same name as the speaker calibration kit in question.

https://www.thanksbuyer.com/j-box-ii-audio-measurement-second-generation-speaker-frequency-response-curve-measurement-kit-support-for-justmls-rew-86413

Regards,
Wayne

 

[Phil Pope]

just tried it again at a lower drive level. less of a difference but still there.

 

[Wayne A. Pflughaupt]

Is the SPL scale accurate?

Regards,
Wayne

 

[Phil Pope]

Is the SPL scale accurate?

Regards,
Wayne

No

 

[John Mulcahy]

Probably best attaching an mdat file with the two measurements, that will make it easier to compare them and their respective cal file responses.

 

[Phil Pope]

here is mdat file

 

[John Mulcahy]

Main contribution looks to be the UMIK's cal file.

 

[Phil Pope]

thanks for looking at this. So the UMik cal file is incorrect? How can you tell it is the Umik cal file rather than the JBox cal file?

 

[John Mulcahy]

There's no knowing which is correct without a known reference, I was just showing where the difference is.

 

[Phil Pope]

thanks. the J-Box measurements seem to keep the same shape regardless of drive level whereas the Umik roll-off seems to be all over the place - which suggests there is something wrong with my Umik set up I guess. And the J-Box measurements are more in line with what I was expecting from published response of the driver (faital HF1460) and crossover model.

 

[ddude003]

I wonder if the two measurements would converge if the measurements were at 75db? I have read in a few threads around the internet that suggest the UMIK-1 is fairly accurate at that level..

 

[Phil Pope]

I changed the internal gain switch on my UMIK to do high volume measurements of subwoofers. amended the cal file to change the gain. could this have made it less accurate at high frequency?

 

[ddude003]

That is an interesting revelation... Since you have modded the hardware UMIK and modded the cal file, "to do high volume measurements of subwoofers", I couldn't really comment on the accuracy at high frequencies... Wouldn't you only be interested in the typical 20Hz to 200Hz range of subwoofers?