Please help!!!!!

[joko3838]

Hello everyone,

I tried moving the bass traps around and monkeyed with placement of my desk, as well as changed up the amp to make sure that it wasn't coloring the signal, nothing changed in the lower frequencies. They were still 20 - 30 db lowder than the upper frequencies.

So, I almost gave up until I decided to use a different microphone and here are the results.

Dynamic Omni-Directional (Blue line)
Dynamic Cardioid (Pink Line)
ECM800 (Green Line)

(I was going to post some links to pictures but the site won't let me.)
I have a thread open on Ethan Winer's forum if you want to look at the pictures.

From what I can tell, the ECM is low frequency sensitive below 300 Hz and the Omni Directional is Hi Frequency sensitive about 300 Hz.

What do you make of this? Is there something wrong with the mics?

Like I said, I re-arranged the room completely and I'm still getting louder low frequency readings using the ECM8000.

Now, I must mention that I'm using the ECM8000 with a regular mixer and not the recommended XENYX802 preamp. I had to buy a phantom power box to allow the mic to work. But non the less, both the phantom box and the microphone are from Behringer (PS400 Phantom box and the ECM8000 mic.

Any thoughts, anyone???????

[thewire]

Do you have a spectral decay of before moving the traps still? The older links to those do not seem to work now.

[brucek]

Any thoughts, anyone???????

Were you using appropiate calibration files for the three microphones?

Did you insert the mic preamp in the soundcard loopback when you were creating the soundcard calibration file to account for its response?

BTW - we have a Post Padding thread where new members can go an fire off five quick posts so that they can then post graphs etc.

brucek

[joko3838]

Hi Thewire, I don't have those because I change went out and bought an ECM8000 so I didn't think that I needed them.

Sorry.

[joko3838]

Were you using appropiate calibration files for the three microphones? I got the cal file for the ECM8000 mic from this site. Not sure where to get cal files for the other mics?

Did you insert the mic preamp in the soundcard loopback when you were creating the soundcard calibration file to account for its response? No, I looped just the sound card and didn't include the outboard mixer. I'll try this tonight.

Thank you.

[joko3838]

Hi and thank you for your prompt responses.

I calibrated the sound card with the mixer in the loop. I've taken readings again and you can find the images in attached.

Please let me know what you think.

Thanks.

[brucek]

Please let me know what you think.

First you'll have to tell us what you're doing and what you're trying to accomplish.

Note, when you measure a subwoofer, use a measure range from 0-200Hz.
Use a vertical graph axis of 45dB to 105dB and a horizontal graph axis from 15Hz to 200Hz. Use the Graph Limits button to set this. Use a target level of 75dBSPL.
Always use stereo mode with the crossover engaged in your receiver with all soundfield and effects turned off.

Waterfall plots use the same graph limits and must be changed to LOG rather than LIN mode (using the Freq Axis button).

brucek

[joko3838]

Hi Brucek,

- I have a home studio/mixing room that I'm trying to control my bass and high frequencies. When I read the graphs, it looks like my lower frequencies are much louder than my upper frequencies and the crossover seems to be at 300 Hz. The difference between each side seems to be between 20 - 30 db. I don't know what is causing this decrease in levels between the upper and lower frequencies. I would think that I need to correct this in order to get accurate mixes.

• Dimensions of the room are 12' x 12.5'.
• I'm using a computer with an M-Audio sound card and Room EQ.
• An ECM8000 Microphone with an outboard phantom power supply from Behringer.
• I have an outboard mixer.
• I'm using an flat amplifier without any controls and my speakers are JBL4410a's.
• No subwoofer.
• I have a total of 20 bass traps in my studio made with Rockwool 5" and Corning 703
  4".
• I have 5" x 7' x 2' bass traps in the side corners. x 4
• I have 4" x 32" triangle bass traps in the top tri-corners. x 4
• I have 4" x 6' x 2' bass traps in the side corners where the wall meets the ceiling. x 4
• I have 4" x 4' x 2' bass traps on each side of me for early reflections. x 2
• I have 4" x 4' x 2' bass traps in front of me and two behind me. x 4
• I have 4" x 4' x 2' bass traps directly above my listening position. x 2
• My speakers are approximately 2" away from the front bass traps and 4' a part.
• My listening possition is approximately 38% from the front of the room. 5' from the front of the room and 7' from the back of the room. I also have 6' from each side of the room. My microphone position for testing is at the listening position, triangular to the speakers. 4 ' away from the center of the speakers.
• My ceiling is 8' tall.
• I have hardwood flooring with carpet covering most of it.

Use a vertical graph axis of 45dB to 105dB and a horizontal graph axis from 15Hz to 200Hz. Use the Graph Limits button to set this. Use a target level of 75dBSPL.
Always use stereo mode with the crossover engaged in your receiver with all soundfield and effects turned off.

Waterfall plots use the same graph limits and must be changed to LOG rather than LIN mode (using the Freq Axis button).

I'll take some readings and change the settings as you requested and post the pictures today.

Thank you.

[joko3838]

Here are the graphs as per your request. I didn't have to change any axis because they were already set int he Graphs Limit button to the settings that you mentioned.

Thanks.

[brucek]

I see no difference in those plots. The axis are still incorrect.

Look at your vertical axis - it is from 20dB to 140dB - change it to 45dB - 105dB. The horizontal should go (in your case) from 15Hz - 4KHz.

Enter the values into the Graph Limits button popup and apply.

We don't require spectral or scope. just waterfall (log, not lin) and response with the correct axis.

brucek

[joko3838]

I hope these are better. I'm trying to understand what you want.

Thanks.

[Wayne A. Pflughaupt]

Top one looks good. Change the axis of the waterfall to the same as the other - i.e. 45-105.

Regards,
Wayne

[brucek]

I hope these are better. I'm trying to understand what you want.

Getting better.

Since you are testing a sub plus mains in pursuit of checking room treatments (I think), then set the Measure screen testing range that you'll sweep from 0Hz -~500Hz and set your graphs horizontal scale to 15Hz-500Hz (for a sub only it would be 15hz-200Hz).

Also be sure you're testing at 75dBSPL target level.

And as Wayne said, set your waterfall to the same axis as the response graph. It also looks like you have a lot of ringing and as such I would change the time range in the waterfall screen from 300ms to 600ms.

Here's what they should look like.






brucek

[joko3838]

Thanks to both of you for your prompt responses.

I think I'm starting to get what you want. A little dense at times.

Just so that you know, I don't have a subwoofer. I only have 2 studio monitors JBL 4410a's.

THanks.

[Wayne A. Pflughaupt]

You might want to check your SPL calibration. 104 dB - wow, are you trying to blow up those speakers?

Regards,
Wayne

[brucek]

I think I'm starting to get what you want

You've changed the response chart to LIN from LOG. Click the Freq Axis button to display LOG.

See how the spacing on your response graph is equally spaced? It looks nice, but we don't hear that way. It has to be LOGarithmically spaced. Look again at my graph I posted above. Clicking the Freq Axis button switches between LOG and LIN. Use LOG.

Also you're still having trouble with your horizontal axis. It should be set to 15Hz to 500Hz. You have your response graph set to 2Hz-500Hz and you have your waterfall from 20Hz-500Hz. Set them both to 15Hz - 500Hz and both of them LOG.

brucek

[joko3838]

That's weard. My SPC1000 meter says that it's reading between 73 and 76 db's. Perhaps it's my mic sensitivity or trim on the mixer that I might be running a bit hot. I'll try to adjust it and see if I can get the reading down. It's not very loud on my speakers at all.

[brucek]

Perhaps it's my mic sensitivity or trim on the mixer that I might be running a bit hot. I'll try to adjust it and see if I can get the reading down.

Hopefully you've loaded the correct calibration file for your microphone into REW?

Hot? No, you must have a mic mixer absolutely flat. Most people include the mixer in the loopback when they create their soundcard cal file.

Hopefully, you've calibrated your soundcard and created a soundcard file that is loaded into REW?

There's not much reason taking measurements unless you have flat calibrated all the equipment you're using, then when you take a measurement, the only variable is the speakers and the room.

brucek

[joko3838]

Here are my latest readings with the levels brought down. I'm going to try a different mixer that should be completely flat and re-calibrate my sound card with the mixer to see what I get.

The cal file that I'm using is as follows:

newecm.cal 09-09-2008 2:31 PM

Hope it's right.

[brucek]

The cal file that I'm using is as follows:

newecm.cal

But you can only use that cal file if you are using a Behringer ECM8000 microphone (and no other).

Again, please use a horizontal scale that is the same for both waterfall and response graphs of 15Hz-500Hz.

brucek

[joko3838]

The microphone that I'm using is a Behringer ECM8000.

I'll check out my other mixer, recalibrate and try it again.

Thanks.

[joko3838]

Okay,

I changed the mixer to another one and I'm getting the same results even though I calibrated my mixer with the sound card M-Audio 2496. Both the spectrum and the waterfall look almost the same with little to no change as with the other mixer.

I've attached a copy of my mic and sound card calibration. The mic calibration is in black and the sound card is in yellow/orange.

Can you tell if I've successfully calibrated my card?

Thanks.

[brucek]

Can you tell if I've successfully calibrated my card?

Yep, it's good.

OK, I am definitely confused. You have a set of standard JBL studio monitors that have only a 10" woofer and a specification of 45Hz-20kHz (+/- 2dB).

Yet you have response that starts to rise below about 300Hz and increases about +30dB as the frequency lowers to an extension beyond 20Hz.

Something is not right. You're doing something wrong. Those speakers should be flat and start to drop off around 45-50Hz - not rise by 30dB starting at 300Hz. It's as if the drivers are bad in the speakers. Where is your microphone located when you measure?

Have you tried a single monitor in the middle of the room, away from reflecting surfaces, and taken a near-field measurement to remove the room effects..

brucek

[joko3838]

I agree about being confused...... the microphone is approximately 4 feet between the two monitors. In a triangular fashion. It's considered the listening position. The monitors are also 4 feet a part.

Perhaps I might have it hooked up wrong.

What I did was put a Y connector from the sound card right out channel to my amplifier left and right in. So that I get sound from both speakers. Then I ran a single wire from my sound card left out to my sound card left in as a loop back (the manual says that it is not necessary but I did it anyway). Then I took an out source from the mixer into the right in on the sound card.

I'll try one speaker tomorrow in the middle of the room and take a reading. Maybe I'll test both to see if one is reading different from the other. (Play it safe)

How far should the microphone be from the speaker when taking te reading?

It's amazing how something so simple can become so complecated when I put my hands on it.

One thing that I may have forgotten to mention to you is the following:

I decided to use three different microphones and here are the results.

Dynamic Omni-Directional (Blue line)
Dynamic Cardioid (Pink Line)
ECM8000 (Green Line)

http://boxstr.com/files/3523124_zu6fl/measured.jpg

From what I can tell, the ECM is low frequency sensitive and the Omni Directional is Hi Frequency sensitive.

What do you make of this? Is there something wrong with the mics? If I put two of the graphs together from the different mic's, I think it looks pretty good and even across with a dip at 120Hz. I'm just grasping here.

Also, I'm using a home system amplifier called B&K ST-202 in it's rawest form. There are no nobs to play with or turn on it at all, therefore I cannot turn up the bass or volume except from on my windows volume controller.
Do you think it's the AMP colouring the sound?

Thanks again to all that have helped. I'm really sorry if I'm causing a bunch of headaches and causing you to pull your hair out....

[brucek]

Then I ran a single wire from my sound card left out to my sound card left in as a loop back

Remove this cable. It will only cause noise.
There are two methods of compensating for soundcard response characteristics.
The first method (highly recommended) is to create a soundcard file that is loaded in REW to compensate for the soundcard. This file is created once (using a temporary loopback cable on the right channel which is later removed). The only cables when taking measurements are the right channel cables. The left channel is left open.
The second method is to use a permanent cable on the left channel and check the box in the settings screen called 'Use Left Channel as Calibration Reference'. No soundcard cal file is made or loaded. REW compensates for the soundcard by using the left channel loopback during each measurement. Extra noise is created using this process.

Use the first method and remove the left channel cable. Also note that to check your soundcard calibration and to ensure that everything is working correctly, you can simply connect your loopback cable on the right channel (with the soundcard cal file loaded and the meter cal cleared) and do a measure. The response should be a perfectly flat line response since the soundcard cal file is applied inversely to the soundcards response characteristics. (Of course if you included a mixer in the soundcard file creation, you would need the mixer in the loop for the test. Be sure that test passes.

How far should the microphone be from the speaker when taking te reading?

Well, depends on the speaker, but a couple feet to try and be equal distance from all drivers.

From what I can tell, the ECM is low frequency sensitive and the Omni Directional is Hi Frequency sensitive

Well sure, and that's why we need to use a microphone calibration file in REW to render each microphone flat. Hopefully, you have the file loaded. That's what all this calibration stuff is about. We render the soundcard and mixer and microphone flat so the only thing we're measuring is the speaker and the room......

Although, I have to say after looking at the ECM8000 response compared to the other mics, that it doesn't look normal (where the other two responses look like typical mains). The calibration file really only compensates below 20hz and above about 10KHz - other than that, it's relatively flat. There is a rather elaborate circuit board of electronics in the ECM. I suppose you may have a fault. I suppose the only solution is to get hold of another ECM8000 (or get a calibration file for one of the other mics).

brucek