DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 25 Posts

·
Moderator
Joined
·
18,544 Posts
Discussion Starter #1
This video discusses the difference in results of measurement methods and why they should be considered in your analysis of results. Especially when a Target/House Curve is used as your goal. These all are based on the typical methods I see used. The first is using a microphone placed at multiple locations and sine swept, one-at-a-time, then averaged together to form a singular response. The second is a moving microphone average with you holding the microphone in front of you while in the seated area. The third is like the second, except you are not physically in the seat holding the microphone; you are behind the seat holding the microphone on a boom or by hand. These methods all result in a different response. The magnitude of differences will vary from situation to situation. But the point is: a difference exists. When trying to match your response to a target curve these differences can have a dramatic effect on the result of your equalization process. It is important to understand these differences exist and understand how different they are in your specific case. I am simply illustrating the fact these differences do exist.

 

·
Registered
Joined
·
756 Posts
I get very different results if the circles I make around my ears are spinning as the tires do (the mic moves up and down a lot, but not side to side a lot) or when I move the mic in flat circles as if around the perimeter of a small serving tray I'm holding on my shoulder (the mic moves side to side a lot, but not up and down a lot).

I've been trying to concoct some sort of a rotation pattern like 5 horizontal circles then 5 vertical circles to each ear. This is even more food for thought!

I suppose there is something to be said for EQing around my own fat head, I imagine my fat head all up in the ceiling and window and upper door frame is going to do a number on reflections.




273054
 

·
Premium Member
Joined
·
3,814 Posts
The goal of moving the microphone is to account for head shadowing. therefore it makes the most sense for the mic to move side to side as opposed to forward and back in between the 2 points
 
  • Like
Reactions: Don Camillo

·
Registered
Joined
·
8,484 Posts
The goal of moving the microphone is to account for head shadowing. therefore it makes the most sense for the mic to move side to side as opposed to forward and back in between the 2 points
It still doesn't account for head shadowing, because there is no shadow.

Good video, Erin. Personally, I sit in the back seat, and move the mic is "flat circles" like opekone mentioned. I usually set REW to 32 averages, take a measument with the mic still, then do another 32 averages while moving the mic. Then I average those 2 measurements. I really like using REWs RTA averaging.
 

·
Premium Member
Joined
·
795 Posts
Can i wedge the mic MTK.1 between the headrest and seat poking out as far as possible and get a good result while im outside the car?
 

·
Registered
Joined
·
197 Posts
For your Pink noise measurements, do you use Correlated pink noise or Uncorrelated pink noise?
 

·
Registered
Joined
·
197 Posts
Have you ever just picked one of mic placement method and tuned the entire system with just that method. If so, which one gave you the best sounding system, and I'm not referring to winning a competition tune, but the most enjoyable listen to while driving to work tune...
 

·
Registered
Joined
·
756 Posts
Have you ever just picked one of mic placement method and tuned the entire system with just that method. If so, which one gave you the best sounding system, and I'm not referring to winning a competition tune, but the most enjoyable listen to while driving to work tune...
Whatever method you're contemplating just go for it, the only thing you have to lose is your time.
 

·
Registered
Joined
·
197 Posts
So... its like if you find yourself with an ugly chick, just go for it, beauty is just a light switch away, 'cause it all feels the same when the lights are off....
 

·
Premium Member
Joined
·
4,746 Posts
For your Pink noise measurements, do you use Correlated pink noise or Uncorrelated pink noise?
Erin should reply with a better answer.

However:
  • I think correlated would give better noise suppression.
  • Correlated would decorrelate, with a moving mic, as the frequency increases.
 
  • Like
Reactions: tjk_bail

·
Registered
Joined
·
197 Posts
Erin should reply with a better answer.

However:
  • I think correlated would give better noise suppression.
  • Correlated would decorrelate, with a moving mic, as the frequency increases.
When you say "correlated would give better noise suppression", i'm not clear on the narrative that applies to. What type of noise would correlated suppress that Uncorrelated would not suppress?

When you say "Correlated would decorrelate with a moving mic" Well thats over my head too.... is there a difference between decorrelate and uncorrelate?

I've heard that its better to measure and EQ individual speakers with Correlated pink noise, but when tuning-Eq'ing Left&Right speakers together, that Uncorrelated is better. Is their any truth to this?
 

·
Registered
Joined
·
197 Posts
I ran both methods in one sitting. Mic in head rest and the other with oscillating fan. The fan method was way better.
I've been thinking about trying this method... Looks like the mic movement for every measurement would be consistent, which would mean better and more accurate results... I just need to go buy a fan.......
 

·
Premium Member
Joined
·
4,746 Posts
When you say "correlated would give better noise suppression", i'm not clear on the narrative that applies to. What type of noise would correlated suppress that Uncorrelated would not suppress?
...
If the particular s/w is matching the mic against a copy of what was sent to the speakers, then it could have processing gain.


...
When you say "Correlated would decorrelate with a moving mic" Well thats over my head too.... is there a difference between decorrelate and uncorrelate?
...
At 2kHz the wavelength is about 6", so the thing is moving 360 degrees in phase very 6" of fanning.
That is referred to as decor relating, as the movement is causing the correlation to lessen.
It all depends on the time window that processing is happening over, and how much it is moving in phase over that time period.


...
I've heard that its better to measure and EQ individual speakers with Correlated pink noise, but when tuning-Eq'ing Left&Right speakers together, that Uncorrelated is better. Is their any truth to this?
I dunno... I suspect there may be some wisdom in it.
This first two sections were answered with theory, but I have not considered this last one.

I would use uncorrelate if I was fanning, and I suspect if all the speakers are done with fanning, then also use uncorrelated. But if the mic is fixed, then it seems like it should not matter if it is correlated.

And then the s/w implementations are also somewhat opaque.
 

·
Registered
Joined
·
1,065 Posts
Have you ever just picked one of mic placement method and tuned the entire system with just that method. If so, which one gave you the best sounding system, and I'm not referring to winning a competition tune, but the most enjoyable listen to while driving to work tune...
How you measure won’t make any odds to whether it’s a competition tune or for daily, that’s just which house curve you use, the measurement method will be good for one if it’s good for the other 👍🏼
 
1 - 20 of 25 Posts
Top