Measurement Interpretation Help - AVS Forum
DIY Speakers and Subs > Measurement Interpretation Help
mefistofelez's Avatar mefistofelez 10:02 PM 05-15-2013
Greetings all,

I have performed measurements on a TD-15M in a 107 l sealed enclosure and I need some help with interpreting the measurement's results. All the measurements were made with HolmImpulse, raw data, free space, grass surface, 3 m (117 in) distance between the microphone and the center of the driver. The center of the driver was 16.5 in from the ground.



The first image shows the microphone positioned at height of 44 in. The calculated cancellation frequency is 586 Hz; the measured cancellation frequency is about 562 Hz. I attribute the "smear" in the cancellation to scatter from uneven surface.



The second image shows the microphone positioned at height of 34 in. The calculated cancellation frequency is 739 Hz; the measured cancellation frequency is about 732 Hz.



The third image shows the microphone positioned at height of 17 in. The calculated cancellation frequency is 1436 Hz; the measured cancellation frequency is about 1441 Hz.

Based on the foregoing, I would argue, that I have an understanding of the cancellation due to reflection of the sound from the ground as a function of a difference between the position of the microphone in relationship to the driver.

Where I am confused is what are the dips at approximately 48 Hz, 98 Hz, 153 Hz, 215 Hz, 271 Hz, and so on. The difference between the dips is roughly 50 Hz.

Any help would be greatly appreciated.

Kindest regards,

M

cookieattk's Avatar cookieattk 11:03 PM 05-15-2013
just take some near field measurements
tuxedocivic's Avatar tuxedocivic 11:27 PM 05-15-2013
There is likely a reflection somewhere related to 50hz. Try adjusting the gate to 20ms and see what you get. That'll make the data not useful below about 100hz and very not useful below 50hz. We're you near a house or a car?

And ya, why such a measurement. Why not ground plane? Move the mic to a ground plane position or raise the speaker and microphone up and place the mic 1m away and gate that measurement.
mefistofelez's Avatar mefistofelez 10:52 AM 05-16-2013
Hi cookieattk,

thank you for the reply. What information will I gain from a near-field measurement? Here is a measurement with the same conditions as in the third measurement in the previous message except the distance is now a 1 m. It seems to me that the details of the response is attenuated compared to the longer distance.



Hi tuxedocivic,

thank you for the reply. No I was not close to any structure. But I will try your suggestion.

Why such a measurement? Well, first I was trying to get familiar with the measurement software and the measurement methodology. Furthermore, I was trying to understand how is a cancellation due to a reflection dependent as a function of different positions of the microphone with respect to the driver.

I am little leery of ground plane measurement, because the parameters of the ground plane, i.e., reflection, scattering, and the like are unknown. So I am proposing to next measure by placing the speaker on its back and suspend the microphone above. Yes, I understand that I should bury the speaker level with the ground, but the proposed methodology should give me an approximation of 2 pi response.

Kindest regards,

M
djarchow's Avatar djarchow 11:06 AM 05-16-2013
You should be able to look at the impulse response see where your reflections are occurring. then gate the measurement appropriately. I normally do ground plane measurements in my large driveway and even there the lowest I can usually go is 200 Hz before reflections from the garage, trees etc begin to contaminate the measurement. Below that I do a nearfield measurement of the woofer (and port if there is one) and scale and combine the measurements to get a full range response.
Bill Fitzmaurice's Avatar Bill Fitzmaurice 11:54 AM 05-16-2013
Measure ground plane and your problems will go away. There's no good reason to elevate the mic.
mefistofelez's Avatar mefistofelez 12:51 PM 05-16-2013
Hi djarchow,

following your and tuxedocivic's advice regarding the gating, I arrived at possible suspect; there was a big boulder behind the speaker. But at the low frequencies the sound is omnidirectional so this may be the the culprit. I will move the speaker an re-measure.

Hi Bill,

thank you for the reply. The different distances and heights were intentional, so that I could understand how is a cancellation due to a reflection dependent as a function of different positions of the microphone with respect to the driver.

My main issue was the ripples in the 40 Hz to 250 Hz bandwidth. But, these might be explained as noted above once I re-do the measurement.

Kindest regards,

M
djarchow's Avatar djarchow 01:27 PM 05-16-2013
Quote:
Originally Posted by mefistofelez View Post

Hi djarchow,

following your and tuxedocivic's advice regarding the gating, I arrived at possible suspect; there was a big boulder behind the speaker. But at the low frequencies the sound is omnidirectional so this may be the the culprit. I will move the speaker an re-measure.

Hi Bill,

thank you for the reply. The different distances and heights were intentional, so that I could understand how is a cancellation due to a reflection dependent as a function of different positions of the microphone with respect to the driver.

My main issue was the ripples in the 40 Hz to 250 Hz bandwidth. But, these might be explained as noted above once I re-do the measurement.

Kindest regards,

M

To get a clean measurement down to 40 Hz without reflections messing up your data you need to have roughly 9 meters of clear space around the speaker and mic.

Regards,

Dennis
tuxedocivic's Avatar tuxedocivic 01:33 PM 05-16-2013
It also helps to move the mic closer so that the direct sound swamps the reflected sound. 3m is pretty far out. 2m is normal for GP. 1m is somewhat close for GP but I don't see an issue and is usually perfect for a normal sized driver/cab when raised up.
mefistofelez's Avatar mefistofelez 02:32 PM 05-16-2013
Hi Dennis,

I think that I can arrange for 9 m space.

Hi tuxedocivic,

what is GP please?

Kindest regards,

M
tuxedocivic's Avatar tuxedocivic 03:13 PM 05-16-2013
GP is Ground Plane
Bill Fitzmaurice's Avatar Bill Fitzmaurice 04:04 PM 05-16-2013
Quote:
Originally Posted by tuxedocivic View Post

1m is somewhat close for GP but I don't see an issue .
That close to a large cab/driver you can get line source characteristics that throw the result off. I prefer to measure at 4m to take the cab out of the equation. So long as there are no boundaries within 15m reflections shouldn't be a problem, but you can gate to be sure of it.
tuxedocivic's Avatar tuxedocivic 04:53 PM 05-16-2013
It depends what you're doing. If you're interested in LF, I don't see an issue. But ya, if you want to see cab effects then 1m is to close for a large cab/driver like this 15". But then I never bother with GP to get that info cause the mic on the ground can be problematic. Above 150hz I raise the mic and speaker and gate. So I don't see an issue with GP at 1m.
Up
Mobile  Desktop