Upper Frequencies - Home Theater Forum and Systems - HomeTheaterShack.com

Old 04-22-13, 07:57 PM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Upper Frequencies

Hi can someone please comment on these graphs and questions.

I have noticed that the upper frequencies are lacking and wondering which calibration file to use.

The 90% file graphs don't look as bad.

1. With the upper frequencies especially 3khz > between the Left & Right and Both graphs the Both graph drops off before 10khz. Why does this happen when the Left & Right seperate doesn't have as much drop off?

2. Which calibration file (0 vs 90 degrees)/mic position(horizontal vs vertical) should i use when there is a difference?

Green trace = Left
Purple = Right
Red = Both

Purple = Left
Green = Right
Tan = Both

Last edited by Phillips; 04-22-13 at 08:06 PM.
Phillips is offline

Old 04-22-13, 09:24 PM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: Upper Frequencies

Quote:
Phillips wrote: View Post
1. With the upper frequencies especially 3khz > between the Left & Right and Both graphs the Both graph drops off before 10khz. Why does this happen when the Left & Right seperate doesn't have as much drop off?
This happens when the mic is not exactly equidistant from both the speakers. The phase difference at the high frequencies reduces the combined SPL as the frequency rises. An 8.6 mm distance difference is a 180° phase difference at 20 kHz (a null). Your distance difference was probably nearer half that, i.e, 4.3 mm or 90° at 20 kHz as the combined SPL level was near the individual levels.

You can prove this to yourself if you like. With loopback on, adjust the mic position slightly until the FL and FR IR peaks is exactly aligned. Then try the measurements again. The SPL reinforcement will now occur all the way to 20 kHz. You can also do this experiment by manually offsetting the individual IR peaks and then using A + B math to add them together to see the impact.

Quote:
2. Which calibration file (0 vs 90 degrees)/mic position(horizontal vs vertical) should i use when there is a difference?
It theoretically wouldn't make a difference if you had a a free field condition.

For measuring high frequencies in a room though, it is best to use the 0° cal file with mic pointed toward the speakers.

The mic is more sensitive to 0° signals for the high frequencies so if you point it up at 90° and use the 90° cal file the mic will emphasize the ceiling reflections a little more possibly making the 90° HF readings appear higher that the 0° readings. The 0° on axis approach deemphasizes the floor and ceiling reflections a little.

Also a 20° error pointing a mic on axis has almost no affect on the reading while a 20° error on a 90° mic setup will provide significantly more SPL error at high frequencies. Check out the typical polar plots of a mic. The story is shown there.
jtalden is offline
Old 04-22-13, 10:28 PM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Re: Upper Frequencies

Quote:
You can prove this to yourself if you like. With loopback on, adjust the mic position slightly until the FL and FR IR peaks is exactly aligned. Then try the measurements again. The SPL reinforcement will now occur all the way to 20 kHz. You can also do this experiment by manually offsetting the individual IR peaks and then using A + B math to add them together to see the impact.
Thank you

I use a USB mic.

Can i still use the manually offsetting the individual IR peaks?

The head is 20cm wide so how do i fix this?

Quote:
It theoretically wouldn't make a difference if you had a a free field condition.

For measuring high frequencies in a room though, it is best to use the 0° cal file with mic pointed toward the speakers.

The mic is more sensitive to 0° signals for the high frequencies so if you point it up at 90° and use the 90° cal file the mic will emphasize the ceiling reflections a little more possibly making the 90° HF readings appear higher that the 0° readings. The 0° on axis approach deemphasizes the floor and ceiling reflections a little.

Also a 20° error pointing a mic on axis has almost no affect on the reading while a 20° error on a 90° mic setup will provide significantly more SPL error at high frequencies. Check out the typical polar plots of a mic. The story is shown there.
Doesn't the difference in files compensate for this?

Thanks again
Phillips is offline
Old 04-22-13, 11:34 PM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Re: Upper Frequencies

Have done a bit of reading and it appears:

0 degree file for measuring speakers directly and subwoofers.
90% degree file for measuring rooms/speakers with the mic pointed 75-80% slightly pointed towards the speaker, so not quite 90%.

Bit confused
Phillips is offline
Old 04-23-13, 07:07 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: Upper Frequencies

Quote:
Phillips wrote: View Post
I use a USB mic.

Can i still use the manually offsetting the individual IR peaks?
You cannot align the IR peaks experimentally by moving the mic unless you have loopback capability.

We can still manually move the IR peaks using the offset box in the Impulse controls. With the individual IRs exactly aligned the HF is reinforced. If we find the exact very small IR offset that you actually measured at then the A + B math combination will result in the same trace that you physically measured.

Quote:
The head is 20cm wide so how do i fix this?
?? I don't understand what you are asking here.

The width of the tip of the mic does impact the frequency and rate of the HF directional characteristics of the mic.

The width does not impact how far we can move it however, so we can still move it 4.3 mm if we want.

Quote:
Doesn't the difference in files compensate for this?
All measuring mics are somewhat directional at HF and changing the calibration file does not change this physical characteristic in any way. It just compensates for the SPL sensitivity in the specified direction.

Our omnidirectional mics are only omnidirectional below about 2 kHz and above that they are progressively become directional. The mic gives more voltage out (SPL) for an on axis vs an off angle HF signal of equal magnitude.

Calibration files will properly compensate the SPL for a signal from the designated direction only in a free field condition. A room is not free field as room boundaries allow reflections from different directions that are also measured by the mic. The calibration files cannot compensate for this as all rooms and setups are different.
jtalden is offline
Old 04-23-13, 07:47 AM
Elite Shackster

Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,368
My System
Re: Upper Frequencies

Quote:
Phillips wrote: View Post
Have done a bit of reading and it appears:

0 degree file for measuring speakers directly and subwoofers.
90% degree file for measuring rooms/speakers with the mic pointed 75-80% slightly pointed towards the speaker, so not quite 90%.

Bit confused
I remember Wayne recommending pointing the mic at the speaker (on axis) and then tilting it upward 20° for measurements of a speaker/room if you are interested in the HF response.

If you are measuring LF he notes that it does not make a difference (the mic is omnidirectional in the bass range).

If you use 0° and 90° calibration files the difference between 0° and 90° mic orientations measurements is minimized. In my room I still see a clear difference.

In the end, I am not sure if there is a practical impact between these two options. If we EQ the HF, we are still going to tilt our house curve to suit our taste given our specific setup. If we aren't going to EQ the HF, then there cannot be a practical impact. I just prefer to point the mic generally toward the speaker under test as the repeatability is improved and there is slightly less emphasis given to the room reflections.
jtalden is offline
Old 04-23-13, 08:56 PM
HTS Senior Moderator

Join Date: Apr 2006
Location: Katy, Texas
Posts: 8,811
Re: Upper Frequencies

Quote:
Phillips wrote: View Post
Hi can someone please comment on these graphs and questions.

I have noticed that the upper frequencies are lacking and wondering which calibration file to use.

The 90% file graphs don't look as bad.
You forgot to tell us what mic orientation you were using for these measurements.

That said, do any of the graphs represent what you’re actually hearing? IOW, do the highs sound “right,” or do they sound soft, like the first three graphs show?

Quote:
Phillips wrote: View Post
Have done a bit of reading and it appears:

90% degree file for measuring rooms/speakers with the mic pointed 75-80% slightly pointed towards the speaker, so not quite 90%.
I assume you got that from the Downloads page? It’s bad information, especially since our stock calibration file is 0-degree.

Regards,
Wayne A. Pflughaupt

Wayne A. Pflughaupt is offline
Old 04-24-13, 01:43 AM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Re: Upper Frequencies

Quote:
We can still manually move the IR peaks using the offset box in the Impulse controls. With the individual IRs exactly aligned the HF is reinforced. If we find the exact very small IR offset that you actually measured at then the A + B math combination will result in the same trace that you physically measured.
Ok will give it a try.

Quote:
?? I don't understand what you are asking here.
The difference in moving the mic might be 4.3mm but we measure in the middle (between our ears so to speak). We don't measure at each ear.

Thank you
Phillips is offline
Old 04-24-13, 01:49 AM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Re: Upper Frequencies

Quote:
jtalden wrote: View Post
I remember Wayne recommending pointing the mic at the speaker (on axis) and then tilting it upward 20° for measurements of a speaker/room if you are interested in the HF response.

If you are measuring LF he notes that it does not make a difference (the mic is omnidirectional in the bass range).

If you use 0° and 90° calibration files the difference between 0° and 90° mic orientations measurements is minimized. In my room I still see a clear difference.

In the end, I am not sure if there is a practical impact between these two options. If we EQ the HF, we are still going to tilt our house curve to suit our taste given our specific setup. If we aren't going to EQ the HF, then there cannot be a practical impact. I just prefer to point the mic generally toward the speaker under test as the repeatability is improved and there is slightly less emphasis given to the room reflections.

Actually got it from the site below, Googled it.

http://www.hifizine.com/2012/09/dayt...spectrum-labs/

Thanks
Phillips is offline
Old 04-24-13, 02:00 AM Thread Starter
Elite Shackster

Join Date: Aug 2011
Location: New Zealand
Posts: 1,278
Re: Upper Frequencies

Quote:
You forgot to tell us what mic orientation you were using for these measurements
.

For the 0 degrees the mic was pointed horizontal with slight upwards.
For the 90 degrees the mic was pointed vertical.

Quote:
That said, do any of the graphs represent what you’re actually hearing? IOW, do the highs sound “right,” or do they sound soft, like the first three graphs show?
Depending on my hearing the top three represent more what i am hearing, soft.

Quote:
I assume you got that from the Downloads page? It’s bad information, especially since our stock calibration file is 0-degree.
Got it from the site below.

http://www.hifizine.com/2012/09/dayt...spectrum-labs/

Thank you
Phillips is offline

 Bookmarks

 Tags frequencies , upper

Message:
Options

## Register Now

Random Question
Random Question #2

User Name:
OR

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.