Um....What? I Need HELP

[ousooner2]

So for awhile now I've been wondering why my AVR is having to boost my top end so much.....and why I feel like I'm not getting any lower end content (40hz and under) in my room. I felt maybe I'm just expecting too much bass that low and just went with it, not to mention I don't get to listen or watch movies as much as I used to. So today I had some free time and wanted to use my Radioshack SPL meter to gather some data.


Test Disc/Tracks: Alan Parsons Soundcheck Disc -- Pink Noise 20hz to 20kHz (in 1/3 octave steps)




I'll start off by posting anything BUT my complete lower end numbers....b/c I'm still trying to figure out wtf is going on there. Yeah. It's that wonky. Lets just say I'm within 3-4dbs from 1kHz down to 50hz.......and then:


1kHz to 50hz: flat within about 3-4db variation (reference 75db at 1kHz and I'm getting 78db at 50hz)
40hz: 73db
31.5hz: 64db
20hz: 60db


Yeah....and that's AFTER Audyssey has done it's thing. Set to FLAT (MultEQ XT on my 1712). Using what I've found below I figured disabling MultEQ XT would have no effect on these numbers. Likely be the same pattern









So lets move on to whatever the heck is happening to my top end. I will not that while I was doing this using a reference tone at 1kHz @ 75dB I heard something odd from the left e55ti tweeter. I can't recall the frequency, but it was obviously around 2kHz+.....anyways, I heard an odd noise. This has happened before to me if anyone recalls this and I thought I had fixed it by just tightening the tweeter screws. I guess not. It makes a high pitched noise and when I press on the sides of the tweeter faceplate it goes away. It's ringing or something. It sounds like ringing. So no telling what that's about or if that could be the issue causing THIS problem below:




This is with Audyssey MultEQ DISABLED so it's just the e55's
1kHz: 75dB
1.25k: 71.5
1.6k: 66
2k: 59
2.5k: 57
3.15k: 65
4k: 78
5k: 78
6.3k: 53
8k: 44
10k: 41
12.5k: 42
16k: 43
20kHz: 43dB




Yes.....it goes from 75dB at 1kHz to 59dB at 2kHz at my listening position. And slightly lower by 2.5kHz! Then back up by 4kHz...and then DROPS OFF A TABLE at 6.3kHz!!! That's a 25dB swing from 5kHz to 6.3kHz I played the same list WITH AUDYSSEY SET TO FLAT and had slightly different dB's, but the EXACT same trend. Falling from 1.25kHz to 1.6kHz and ramping up again at 4kHz and falling again from 5kHz to 6.3kHz.


I thought maybe my SPL meter was getting messed up by listening from 12ish feet away at the main listening position so I did some measurements from about 1' away:


Pink Noise
5kHz: 81dB
6.3kHz: 68dB


Sine Waves
5kHz: 83dB
6.3kHz: 51dB






I have no idea what's going on..... I'm starting to wonder if my SPL meter is bad. But it takes readings normally on other things so I would think that idea is out no?

 

[ousooner2]

Could this be an AVR issue? Weird thing is that they don't sound like the SPL numbers suggest. So maybe the SPL meter is bad?

 

[ski2xblack]

If you used REW and a calibrated mic you would get a much better look at what's going on in your room, and have the ability to tweak in real-time, seeing the changes, from repositioning speakers, adding bass traps, absorbtion, diffustion, etc. Fair warning: the measurements you got from such gear would probably be just as wild as what your spl meter measurements indicate, so don't get too upset by what you've noticed so far. It's probably uglier than you imagine! All our rooms are, at least until we address those issues. Everyone wants a frequency response graph that looks like a Kansas landscape, but they generally more closely resemble the Rocky Mountains.

 

[ousooner2]

Yeah I've thought about getting an Omnimic or something cheap that works well and DL'ing REW. I don't have a laptop so that's a pain, but maybe I could find one cheap.

I'm amazed by the HUGE dips I'm getting though. And my bass (dual rw-12d's) drops off very very quickly after 40hz. This is in my main living room so I really can't just go placing room treatments, etc everywhere. Next house though.....bonus room/theater room!

 

[agarwalro]

Are you applying the correction offsets to your readings?

http://forum.audiogon.com/cgi-bin/fr.pl?htech&983682086&openflup&1&4

It will be a lot easier if you visualize the measurements for SPL vs. frequency in Excel.

 

[charmerci]

Yeah I've thought about getting an Omnimic or something cheap that works well and DL'ing REW. I don't have a laptop so that's a pain, but maybe I could find one cheap.

I'm amazed by the HUGE dips I'm getting though. And my bass (dual rw-12d's) drops off very very quickly after 40hz. This is in my main living room so I really can't just go placing room treatments, etc everywhere. Next house though.....bonus room/theater room!

Forget measuring much of anything under 200hz in a standard room with any accuracy - at least as far as the speaker output is concerned. A 40 hz sound wavelength is 28 feet long. (80>14 feet.) Anything below 200 is more about your room than your speakers.

 

[ousooner2]

True to some extent, but that doesn't explain the huge dips from 1kHz and up ya know

 

[charmerci]

How reflective or "dead" is your room? Maybe certain corners/walls are cancelling/reinforcing reflections. Are you taking measurements 1 meter in front of your speakers? I'm thinking all of this is simply your room.

 

[ousooner2]

How can I tell those things? I can take pictures if that'd be helpful

Measurements are from the LP about 11-12' away. Setup is 11' speaker distance apart and slightly longer than that to the LP. In the original post my measurements are from the main listening position, but I posted some from about 1' away as I wanted to see if it was a room function or possibly my SPL meter wasn't reading correctly at that distance.


"I thought maybe my SPL meter was getting messed up by listening from 12ish feet away at the main listening position so I did some measurements from about 1' away:


Pink Noise
5kHz: 81dB
6.3kHz: 68dB


Sine Waves
5kHz: 83dB
6.3kHz: 51dB"


It seems the same pattern is happening from 1' away so that's what made me think it wasn't the room acoustics playing the entire roll

 

[agarwalro]

Forget measuring much of anything under 200hz in a standard room with any accuracy - at least as far as the speaker output is concerned. A 40 hz sound wavelength is 28 feet long. (80>14 feet.) Anything below 200 is more about your room than your speakers.

To take it further, it is all room response when using pink noise and SPL meter technique...

Unless someone is doing near field gated measurements, with Omnimic or REW, etc., it is room response.

 

[ousooner2]

Wow. So I need something like Klipsch or something bright in this room lol

 

[agarwalro]

You have not responded to my question regarding applying the correction offset to you meter reading.

Are you measuring this in 2.0? That would be a meaningless measurement since it could be that one speaker is OK but the other is malfunctioning. This could explain why the SPL drops off. Remeasure the speakers one at a time.


Maybe the tweeter or a crossover component isvmalfunctioning. Did you give EmpTek your findings to solicit their opinion?

 

[ousooner2]

Sorry...I thought I posted the pic of my corrected/calibrated readings. I guess not though. Here they are corrected:





I'm going to test the speakers individually right now. Good idea. I'm trying to remember how to make a graph in Excel to see what it looks like lol

 

[ousooner2]

Testing EACH speaker INDIVIDUALLY

* Patterns were the same whether I had Audyssey on or off (no EQ). Both resulted in huge dips. The tests below were done with NO EQ


Center Channel from 1 meter
5kHz: 82dB
6.3kHz: 65dB
8kHz: 46dB


Left & Right (indiviually tested) from 1 meter
5kHz: 84dB
6.3kHz: 57dB
8kHz: 48dB

 

[ski2xblack]

You have not responded to my question regarding applying the correction offset to you meter reading.

Are you measuring this in 2.0? That would be a meaningless measurement since it could be that one speaker is OK but the other is malfunctioning. This could explain why the SPL drops off. Remeasure the speakers one at a time.


Maybe the tweeter or a crossover component isvmalfunctioning. Did you give EmpTek your findings to solicit their opinion?

For the reasons you mention I'm not sure his findings tell much, not enough to drag the manufacturer in. It seems more like a learning opportunity for Ousooner to help further assess the situation.

Ousooner, imagine a swimming pool the size of your room. Now you have a collection of rocks of various sizes representing different frequencies. You toss them all in the pool, and the ripples radiate away. Wavelengths that are longer than the pool's dimensions will "rock the pool," and this corresponds to the room's modal response, while those shorter than pool dimensions will propagate radially until they meet the sides of the pool, where they reflect. This eventually results in a complex mosaic of interference patterns of peaks and dips. Now imagine that interference pattern in three dimensions in your room as the sound waves interact. See where stereo tests can lead to some pretty bizarre readings? So if you must persist in your hand held meter measurements, make sure to do them right so you're not attributing them to the wrong root cause. If you haven't read them yet, there are some great primers in the AV Research tab on the front page that may help you make better sense of things and get the most out of this exercise. Toole's book is great, too.

It pains me to think of ousooner performing all these tedious individual measurements with a hand held meter. Aren't there any Audioholics in the neighborhood willing to provide their time, laptop, and REW expertise in exchange for beer?

[edit] I see your testing speakers individually now. Even using a single speaker, you will still be subject to the room's influence. You could compare close mic'd readings to those taken further away to see it (and those far readings are only specific to that exact spot and under those exact conditions).

 

[ousooner2]

Thinking about taking the Ratshack meter back and picking up a mic (omnimic?) to use with something like REW. Real time FR graphs would be nice. Seems about the same price.

I understand what you all are saying though. I understand cancellations, wave physics to an extent, etc. Obviously you're not going to have a flat response in any room..ever..but to be 40db down from 5k to 8k!!



Edit: Just thought about trying the rear speakers (emp e5b's) to see if they have a flatter response from the tweeter. So I mic'd them from about 6' away

5kHz: 82dB
6.3kHz: 57dB
8kHz: 47dB


Same thing. These and the previous measurements in post #14 aren't corrected, but it doesn't really matter as the correction for the Radioshack SPL meter isn't close to what it'd take to make this even close to flat

 

[ski2xblack]

The Omnimic is a specific Dayton product which includes a calibrated mic, the software, and a cd of test tones. It's very idiot proof, but kind of pricey. You can get a calibrated usb mic on it's own for $100, maybe a tad less, and download REW for free. It's a bit less intuitive to use than Omnimic, but just as thorough in it's suite of measurements. You still need a 'puter to run it on.

 

[ousooner2]

I'll have a look at calibrated mic's then. Maybe I can just run a long cable into the study and use my PC.

 

[agarwalro]

REW is a free download from Home Theater Shack. With this you should get the Mini DSP UMIK-1 from Cross Spectrum Labs. You'll still need the SPL meter to calibrate the REW SPL readings for absolute scale, but that is not a mandatory step.

 

[agarwalro]

I'll have a look at calibrated mic's then. Maybe I can just run a long cable into the study and use my PC.

Since this is a USB mic, you'll be limited to 15 ft. cable length. You could get a laptop into the room where the speakers are located.