[Todd Cinci]

I have read Andy's TA approach many times and I am still new to DSP.. Learning more each month. However, I have a 2016 GMC truck with 3 way Focals... Midbass and Mid in factory locations and tweeter is in front on Mid on Dash about an inch cut into dash facing Windshield (I know not the best location, but wanted clean look) and don't have the mechanical ability to modify A -Pillar) . Also have a few difference curves to try as well, but always use Andy's test gear curve as reference. Every single time I do it, I have to drop the level on the passenger side Mid and tweeter - 5 in order to get center imagine. So frustrating because I know that throws off the imagine all together as well... I have done measurements each time and know I am correct on that and input data info JL TWK 88... However, today I adjusted the level on the right side again just a little, but I also decreased the delay on the left side by .21ms and boom, the imagine centered... However, i am the kind of guy that has to understand why... so I ask WHY?... why would that work and am I screwing anything up by doing that imagine wise.. sounds good, but I guess need to compare songs... Could it be because of the mid and tweeter in the dash on the right side are reflecting causing it to throw off the measured TA delay or?? Like the bound in that sound is hitting ears first before left? I just would rather keep the REW frequency on both side the same (match) and not have to move levels.. however, I know in Andy's article, he says never adjust delay to get center imagine.. so confused...

 

[mzmtg]

At frequencies with wavelengths shorter than the size of your head, your brain localizes sound on volume difference, not arrival time.

 

[Todd Cinci]

At frequencies with wavelengths shorter than the size of your head, your brain localizes sound on volume difference, not arrival time.

So, would that mean that my Delay was off to begin with?? By changing that, it is the only time in the last 6 month that the imagine has that spark clarity that I have been wanting... just not sure if its the correct move... More of less looking for someone to say that is a nomal adjustment, or that adjustment was wrong and the great sound is just in my head.. ha ha

 

[GrM]

If the adjustment made it sound better it is. The measurements etc. are just to take as much of the guess work out of it as possible. Chances are TA was off to begin with, that’s the most logical explanation anyway. But your ears don’t lie (unless you’re so used to it being wrong right sounds wrong), if messing with TA improved imaging it’s closer to being correct than it was

 

[Todd Cinci]

If the adjustment made it sound better it is. The measurements etc. are just to take as much of the guess work out of it as possible. Chances are TA was off to begin with, that’s the most logical explanation anyway. But your ears don’t lie (unless you’re so used to it being wrong right sounds wrong), if messing with TA improved imaging it’s closer to being correct than it was

Thank you.... I figured go with what sounds good.. just wasn't sure how much that would mess with true imaging between songs... Guess time will tell when I listen... I have attached snap shot of the changes verse original..

 

[GrM]

Do you really have a 17”+ difference between left and right tweeters/mids? I’d double check the measurements with a friend or piece of string and tape, that doesn’t sound right unless it’s a particularly wide car and odd placement. But if you have the center image where you want it and there’s no cancellation at the listening position that’s all that really matters, but might be worth checking out just the same

 

[Todd Cinci]

Do you really have a 17”+ difference between left and right tweeters/mids? I’d double check the measurements with a friend or piece of string and tape, that doesn’t sound right unless it’s a particularly wide car and odd placement. But if you have the center image where you want it and there’s no cancellation at the listening position that’s all that really matters, but might be worth checking out just the same

Yes. It’s a truck. So the dash is long. Unfortunately.

 

[Todd Cinci]

Yes. It’s a truck. So the dash is long. Unfortunately.

Also how would I know if any cancellations? Place all the test CD and all sounds good unless something I am missing.

 

[GrM]

Ideally you’ll use a mic but if you can find an ascending tone, frequency sweep, that’ll play each tone for a few seconds and if no frequency is quieter than the one before (they sound louder the higher the frequency) that should do. Probably try to find one that’s at least 1/3 octave.

edit: if you have trouble finding one online you can download REW onto a laptop, that has all sorts of test tones and sweeps. But I can’t image it’ll take much finding one on google

 

[gijoe]

You should not be using TA to move the center image, you should be using levels. That said, it's possible that either the TA was not set correctly in the first place, or that you have a different expectation of where center is. The accepted standard for center is between the speakers, not directly in front of the driver. The center image should sound like it's coming from out in front of you, but in a left hand drive car that would place the center out towards the front center of the hood (where you'd find a hood orniment), basically to the right of you as the primary listener.

 

[Todd Cinci]

You should not be using TA to move the center image, you should be using levels. That said, it's possible that either the TA was not set correctly in the first place, or that you have a different expectation of where center is. The accepted standard for center is between the speakers, not directly in front of the driver. The center image should sound like it's coming from out in front of you, but in a left hand drive car that would place the center out towards the front center of the hood (where you'd find a hood orniment), basically to the right of you as the primary listener.

The TWK 88 does all the math for delays but have checked multiple times to verify. And if I used levels to move to center (which for me is dead center. Not left or right. I could never get it. But bay adding less delay (hense adding more inches) to tweeter and mid. The center moved. But still had to move levels as well. Just wasn’t sure if that’s normal in truck since passenger side is so far away. On original rew based on perfect according to mic and room wizard. Would have to move right side negative 7 or 8 to get close but still would pull to right.

 

[Todd Cinci]

The TWK 88 does all the math for delays but have checked multiple times to verify. And if I used levels to move to center (which for me is dead center. Not left or right. I could never get it. But bay adding less delay (hense adding more inches) to tweeter and mid. The center moved. But still had to move levels as well. Just wasn’t sure if that’s normal in truck since passenger side is so far away. On original rew based on perfect according to mic and room wizard. Would have to move right side negative 7 or 8 to get close but still would pull to right.

So basically after rew sides match. If I play test cd. Test gear. It is all right side. I am active 3 way. Weird

 

[gijoe]

I would carefully measure each of the speakers to the listening positions and make sure that the TWK is in the right ballpark. I wouldn't trust the auto TA (if that's what you did), I've never seen it work well, or at least not repeatably.

If you are looking to have the center directly in front of you then you will be drastically squishing the left side of the stage, and stretching out the right side, that's why we don't set center to directly in front of the listener.

 

[Todd Cinci]

I would carefully measure each of the speakers to the listening positions and make sure that the TWK is in the right ballpark. I wouldn't trust the auto TA (if that's what you did), I've never seen it work well, or at least not repeatably.

If you are looking to have the center directly in front of you then you will be drastically squishing the left side of the stage, and stretching out the right side, that's why we don't set center to directly in front of the listener.

Thank you for all your help. Basically I measure to test gear mic. Input in TWK and the delays are auto set. Therefore, are you saying most people still adjust from there even if the measurements are correct? Guess trying to validate me lowering the delay on left side by .22ms. Is that normal. The measurement as exact. The delay set in ms from TWK is stop on based on the measurements. Checked other sources. But still not centered

 

[gijoe]

Thank you for all your help. Basically I measure to test gear mic. Input in TWK and the delays are auto set. Therefore, are you saying most people still adjust from there even if the measurements are correct? Guess trying to validate me lowering the delay on left side by .22ms. Is that normal. The measurement as exact. The delay set in ms from TWK is stop on based on the measurements. Checked other sources. But still not centered

No, if the tape measurements are accurate, there is no adjusting to TA, it's a set it and forget it thing. Moving the stage would then be done by adjusting levels, this is why getting both sides to have the same frequency response is important. If they don't match then different frequencies will float left to right.

TA should be set and forget.

 

[GB4runner]

As above mentioned I have had a great result in adjusting left and right to match exactly in level.
Center image seems pretty darn good to me.

 

[Todd Cinci]

No, if the tape measurements are accurate, there is no adjusting to TA, it's a set it and forget it thing. Moving the stage would then be done by adjusting levels, this is why getting both sides to have the same frequency response is important. If they don't match then different frequencies will float left to right.

TA should be set and forget.

Okay. Makes sense. Thank you all your responses Just weird that the stage seemed so much better. By adjust that left delay slightly. That’s why I wanted to verify. So if what your saying is the case then I have a few questions. When I use pink noise to look for center, I **** off all speakers and adjust tweeters. Then mid, then mid bass. Or should I do all at once?? Can I also increase let side to try and match or just - the right side. Just when I do right side only it may go down -8 before it gets close. Mostly in tweeters. Could it be becuase they are facing up verse in a pod. The mid in factory location usually only have to adjust slightly and mid bass not really at all. All speakers are crossed under beaming as well.

 

[Todd Cinci]

As above mentioned I have had a great result in adjusting left and right to match exactly in level.
Center image seems pretty darn good to me.

Thank you. Helps a lot. Thought I could only adjust right and would mess with imagine if I adjust both levels.

 

[GB4runner]

In my case and I am no expert. I started with measurements. After a ton of reading and doing it the "wrong" way I started with my midbass on drivers side everything else muted. Took measurement with RTA, matched the passenger side everything muted. Then just bounced back and forth between midbass to mids to tweeters until all were matched level wise. I am using a zapco HDSp V16 just to put that out there. I used some test tracks that go from left to right, center, slightly off cnter to the left and right. Sounds pretty good to me but I am waiting for the event nearest me in August so an "expert" can quantify if my system actually sounds good.
For now it sounds good to me

 

[Todd Cinci]

In my case and I am no expert. I started with measurements. After a ton of reading and doing it the "wrong" way I started with my midbass on drivers side everything else muted. Took measurement with RTA, matched the passenger side everything muted. Then just bounced back and forth between midbass to mids to tweeters until all were matched level wise. I am using a zapco HDSp V16 just to put that out there. I used some test tracks that go from left to right, center, slightly off cnter to the left and right. Sounds pretty good to me but I am waiting for the event nearest me in August so an "expert" can quantify if my system actually sounds good.
For now it sounds good to me

Thank you. I do the same with REW. And match each side to curve. However if in am understanding you correctly, are you saying to match levels, you use mic and take DB level to verify on RTA then make sure each side and speaker matches from that perspective or am I way off? So if left mid reads 80, yiu make sure right reads 80 etc? Or spl level??