Noob Question Regarding EQ Filters

[BBLV]

I'm brand new to REW and having an issue with REW not applying boost under my target house curve. I've read a bunch of guides and none of them address this issue, so I think I'm missing something obvious. One would think the max boost setting would do it, but 0db or 6db doesn't make a difference when I calculate filters. I have two large dips on my midbass response that I'm trying to address - hopefully someone can tell me what I'm doing wrong.

Attached is my averaged measurement for my front left and right speakers along with a screenshot of the EQ screen showing my settings, Harman house curve and the lack of filters addressing those two dips under the target curve.

 

[DanDan]

If you add two identical signals you will get +6dB. If you subtract them you get minus infinity or zero.
Nulls are in the latter category, boosting one signal also boosts it's inverse. Nada. You cannot Eq a Reflection Null.

 

[Wayne A. Pflughaupt]

Not all depressions in response are nulls, and those that aren’t nulls can be addressed with EQ. As Dan noted, the way to tell the difference is that a null will not respond to EQ.

The sharp dip to the far left (can’t name the frequency because you cut off the numbers) certainly looks like a null. But even if it isn’t, it’s beyond a fix. You can’t reasonably apply 15-20 dB of equalization.

BTW, look at your Harman Curve graph and notice the little camera icon on the top left corner. That will let you capture a graph to a .jpg. No reason to do screen shots.

Speaking of the Harman curve, there’s no guarantee you will find it to be satisfactory. Generally every room is different and needs its own house curve. If you happen to like it, great. If not, look for my article on the topic at Home Theater Shack Forum.

Regards,
Wayne A. Pflughaupt

 

[BBLV]

Thanks for the replies. I also realized that I may have not measured the front left and right separately.... I think I averaged the responses before creating the EQ filters. I'll give it another go and see if I see different results.

Of note is that I fully treated my screenwall behind the front speakers with 2" ductliner and bass traps in the corners, but have not yet treated the rest of the walls in the rooms. That'll likely show a noticeable improvement in the measurements.

 

[DanDan]

Front Wall Treatment often has no benefit at all. Clouds are always welcome though. https://realtraps.com/art_front-wall.htm

 

[not a wizard]

I am also new to this Forum. I appreciate feedback and comments. I read Wayne's article on House curve and the recommendations...definitely will try those to EQ my speakers to my room. In the article, Wayne talked about using normal listening volume/dB to measure. It makes sense. I also read that the UMIK-1 microphone has about 30 dB boost comparing to other mics. So, will this 30 dB from the UMIK affect my measurements at normal volume? Thanks!

 

[not a wizard]

It the large dip at 50 Hz a null. Please see my graphs. Thanks.

 

[DanDan]

Afaik the UMIK as supplied hits 0dBFS at 94dB SPL. Given the wild peaks we see in untreated rooms, and possible hearing damage, a sweep level of about 75dB is good.
Due to the nature of the process, REW can see or reach a good 15-20dB into the Noise Floor of Mic, Preamp, even the Room.

 

[John Mulcahy]

Max input volume for UMIK-1 with default internal gain switch settings and 0 dB input volume setting is 124 dB. Using max input volume (no reason to) applies 24 dB of gain so max is then about 100 dB.

 

[DanDan]

I stand corrected. My UMIK goes way back. I used it with FuzzMeasure, which if I remember correctly expected 0dBFS to be 94dB.

 

[not a wizard]

Thank you Dan and John for the confirmation. I will use apply the gain and cap the max.

 

[not a wizard]

I tested each speaker individually from 65-75 dB. I found distortions from 20 Hz to 60 Hz. It's not loud enough to have any room effects right? Please see the graphs.

 

[John Mulcahy]

Can't tell much from the distortion curve alone without seeing the noise curve and the measurement fundamental, but the sweep measurement method is very effective at removing harmonic distortion from the measurement result.

 

[not a wizard]

Sorry if I uploaded too many graphs. What else am I missing? Thanks!

 

[John Mulcahy]

Easiest is to attach the mdat file for the measurement.

 

[not a wizard]

Great, thanks!

 

[John Mulcahy]

The distortion is fine, the system has no output to speak of below 60 Hz so the "THD" there is basically just noise. Pretty uneven responses though.