Home Theater Forum and Systems banner

1 - 10 of 10 Posts

·
Registered
Joined
·
868 Posts
Discussion Starter · #1 ·
So I was one of the winners of the miniDSP UMIK microphone giveaway :yay2: Thanks HTS.

I got to playing around with it last night. The microphone installed correctly (as far as I can tell) on my old laptop running Microsoft XP SP2, and REW recognized it. I have used REW in the past a bit, though I'm not an expert by any means.

I was looking at the RTA meter and noticed that at many frequencies it was showing a negative value for SPL. That really surprised me. Since 0 dB is defined as the threshold of audibility, I suppose that negative values for dB SPL are not unreasonable. However, I did not expect to find such values in my room. It was quiet, but this is not a sound-proofed room by any means. It made me wonder if perhaps there is something amiss, something I did wrong.

I didn't calibrate for absolute SPL measurements, but it's my understanding that REW does this automatically from the calibration file for the UMIK mic (which I did provide). The overall SPL from REW's SPL meter was about what I've measured in the past for this room when silent (about 36 - 37 dB).

Any insight anyone can provide about my negative SPL readings on the RTA would be helpful. Thanks.
 

·
Premium Member
Joined
·
1,562 Posts
The big number in the SPL meter should always be positive.
The small number below over the bar scale is marked "dB FS" (Full Scale). That shows how much headroom is available before the meter is overloaded.
 

·
Registered
Joined
·
868 Posts
Discussion Starter · #5 ·
Yes the SPL meter always reads positive. There is always substantial low-frequency energy in the room, which is low enough in both frequency and level to not be noticeable, but the mic and meter pick it up so there's no way that overall SPL is going to be negative. It is the frequencies in the top few octaves as shown on the RTA graph that are dipping below 0 dB.
 

·
Registered
Joined
·
868 Posts
Discussion Starter · #7 ·
Thanks will do.

Can an mdat capture a non-static measurement? In the case of frequency response or waterfall plots, etc, obviously not a problem. For a dynamic (constantly changing) measurement like RTA, would it just be taking a snapshot?
 

·
Premium Member
Joined
·
6,711 Posts
It wouldn't be too difficult to see negative values on the RTA at any particular frequency in such a quiet room, especially if you are using a high RTA resolution such as 1/48 octave. The trace shows the sound within each octave fraction, so if you look at the value at 1 kHz (say) on the 1/48 octave setting you are reading the SPL just for the energy that falls within about a 20 Hz span around 1 kHz, which will naturally be a lot lower than the overall figure for the audio band as shown in the SPL meter.
 

·
Registered
Joined
·
868 Posts
Discussion Starter · #10 ·
Okay I installed REW on a different computer, changed resolution and turned on averaging. That smoothed things out to where the data makes sense. Not sure if the original computer was a problem or not but I will check another time.

I also compared a frequency response measurement to a professional measurement system with a $1000 microphone at work and the measurements were pretty much on top of each other. Looks like everything is well with this system.
 
1 - 10 of 10 Posts
Top