Unusual Impulse Response Reading

[saiteiman]

Hello,

I've been getting consistent IR readings with unusual "spikes" within -20db and extends down to 100ms that should only occur in an extreme reflective room according to the acoustician. However, the room tested is treated with multiple acoustic panels in the front wall and corners in combination with area rug and drape/curtain. I've tried with multiple different mic placements and REW settings with no avail. I'm wondering whether this could be caused by my test method or defective mic. Thanks.

Equipment Chain:
minDSP UMIK-1 USB Mic --> MacBook Pro --> DAC --> Amp --> Speakeer

Room size:
19'Wx14'Lx8'H
Stereo setup w/Mic pointed to front wall

 

[John Mulcahy]

Are you measuring a single speaker? How is the mic supported? The mic support and furniture near the mic also contribute reflections, best putting the mic on the end of a boom. For more commentary better posting the mdat file and some info on placement of mic and speakers in the room.

 

[saiteiman]

I measured with 2 speakers. The miniDSP comes in short tri-pod stand. It's placed on top of the sofa (listening position) and prop up by stack of sofa pillows as I don't have a tall mic stand. There's no furniture obstructing sound path other than the sofa against the rear wall. I even cover the mic with a blanket and still getting spikes in -20db region. See attached files.

 

[John Mulcahy]

Standing the mic on anything is a recipe for frustration I'm afraid, you need to get a proper mic stand that lets you place the mic in free space where your head would be or rig something up that lets you support the mic with nothing around it. You could tape the stand to a broomstick and crouch off to the side to hold it for something very crude.

Focus on the individual speaker measurements. Your listening position is very close to the rear wall so there will be strong reflections from that. The right speaker is much worse than the left, with many reflections that are stronger than the direct sound, which is very unusual. The whole room seems very lively, with rising decay times above 1 kHz. Panels will have little effect unless they cover a significant proportion of the room's surfaces. You can probably get advice on that in the room acoustics and treatments forum.

 

[saiteiman]

John, thanks for your advise. The IR looks better now with mic on boom and recorded with listening position, especially in the middle of room. However, still recorded with spikes within -20db. I'm wondering if this lively room is caused by 1/2" sound dampening sheet rocks constructed for this room or omni-direction speakers? (Although the acoustician pointed out this high numbers of spikes within -20db would be incoherent as to be unlistenable) On the contrary, I don't feel this room is "unlistenable" based on the IR readings. I thought covering the mic with a blanket would shift the graph down in the IR, no? Jimmy

 

[John Mulcahy]

The IR graph is normalised to the peak, so everything probably moved together. The measurements do all drop off very sharply at the top end, some at 10k and some at 12k. That is pretty unusual.

 

[saiteiman]

Is it normal to see recordings (graph) on the IR tab when speakers are in mute? See attached files.

 

[John Mulcahy]

That's background noise, muting the speakers doesn't turn off the mic.