Ugghh...help

[mamm7215]

I finally got this working and did an analysis of my home studio room. I have bass traps floor-ceiling in all 4 corners and 2 5x4' panels on each long end wall and wall/ceiling traps above those and a 5x4' cloud above the mixing desk. I used another eq tool and it reported a big null at about 90hz. REW confirms this. Is it the room or my speakers or mic??? What can I do about a null like this, it's a critical bass frequency?
Tia.
M.

[JohnM]

If you have a sub in the system adjusting the sub's phase may help. If not, the dip is likely to be due to the position of you or your speakers or both, try measuring in different locations to see what effect changing the listening position has and try moving the speakers to see if there is a position that gives a smoother response.

[brucek]

Let me make a few points about REW first.

It would be wise to use a vertical scale of 45dB-105dB for your graphs. This allows an reasonable swing of +/-30dB (total 60dB) around a 75dB target. Your graph uses a total of ~190dB. This makes the presentation quite kind to your results.
Use a vertical axis of 45dB-105dB.

The horizontal scale shows an extension at the low end to 2Hz. You can see the response of your speakers extending into the noise at about 20Hz. Use this as you bottom figure of your horizontal axis. You've measured to 200Hz, so make that your top figure.
Use horizontal axis of 20Hz-200Hz.

When using waterfal it's most reveiling to use the same axis as the response graphs and also to use a LOG view rather than the LINear view. This way you can use the slider to dial down to a single slice that matches the response graph and then slowly move to 30 and watch the signal degrade out.

Is it the room or my speakers or mic???

Not enough information. You need to tell us what type of mic you're using and whether you are using the correct calibration file.
Is this a single monitor at the listening position at the console?
If you know your mic is correct, then you can do a near-field measure of a monitor outside or at least in the middle of a room where there are no boundaries. This will (attempt) to reveal the response of the speaker itself without the room influence.
Then when you know it's not your speakers, take measures at different spots in the room. You may have to move the console a bit to reduce the dip..

brucek

[mamm7215]

Ok, I've done a bunch more tests, it seems that the null follows mic placement, ie if I move the mic back from the listening position in 6" increments, then the null follows at higer positions. Ex: Listening pos, -6in, -12in, -18in the null goes from 92hz, 104hz, 108hz, 112hz

[mamm7215]

Interesting...after a bunch of measurements, I've found that between the current listening position and 2 other positions, the last being standing up 3 feet back of listening, I can get a composite accurate graph if overlayed with each other. Would it be ok to use 3-4 listening positions to get a better idea of overall response? There's not a lot of room for any more treatment in here but I can live with the comporomise.

[brucek]

Actually, we just had a member here a week or so ago that had a small studio and some of the same positional problems you're experiencing. I think you would find it interesting to read this thread completely through. Positioning ended up being the best fix. I know it's a bit more difficult when it's a studio, but sometimes it's needed.

I think unlike a home theater where there are concerns for listening positions of other individuals, wouldn't a studio attempt to have a single sweet spot for mixing?

brucek

[mamm7215]

Absolutely, but it's not a big room so moving around a bit won't kill me
I'll read that thread. The other problem is I'm using Yorkville YSM1P's with no sub. They're good for mid and high freq but the lows are weak on them, even though they're rated down to 45hz.

[mamm7215]

BTW, how do I set the graph to read 20-200hz? I can't get it exact using the magnifying glass in the corner?

[mamm7215]

Another update...I just redid the test using another pre-amp. This with the same mic (Apex 460 modded with peluso capsule, cinemag xfrmr and changed out output cap). A bit different response so there is obviously some user error here as well?? or original pre's characteristics (ART MPA GOLD) came out in the plot...

[mamm7215]

Nevermind, figured the graph out...

[mamm7215]

Alright, after all this (and thanks for all your help), here's what my last graphs look like, I don't think this is too bad, I do also check mixes with spectral analyzers as well to confirm what I think my ears are hearing, I wouldn't mind trying this with a sub but I'll experiment with more placement as per the other thread like this to see what I can do with speaker and listening placement as well.

[brucek]

BTW, how do I set the graph to read 20-200hz? I can't get it exact using the magnifying glass in the corner?

Just click the Graph Limits icon in the top right hand corner of REW. Once you set the limits, you can retrieve those stored limits any time you click the icon.

brucek