[Rapture333]

So the other day, I was dropping off a boat amp under warranty for my brother at a popular shop in downtown Miami. While I was waiting for the amp to be replaced (they had to bring over another amp from another shop), the owner wanted to sit in on my car as we were talking about SQ builds. His shop dealt mostly with Focal and my car is mostly a Morel build. The first thing he says as he got into my car was that my sound stage was blurry, my philosophy is that the sound stage should be directly in front of the driver, but he insisted it should be on the center of the dash (so the driver would be sitting to the left of the stage). It didn't sound bad after I let him adjust it, but he also said my front stage was delayed way too much, and that my subs should be delayed as well. Here were my old settings on the DSP:

Old:

FL 8.00 FR 7.15
RL 1.40 RR 0.00
SUB 0.00

And these are the settings he dialed in:

FL 1.25 FR 0.00
RL 0.00 RR 0.00 (didn't tune the rear)
SUB (11.00)

What I thought was interesting, is that I always assumed the front stage should be delayed the most to create the illusion of depth, and that the subs didn't need any time correction at all, and that the fartherst speaker from the driver should remain at 0.00.

However, he insisted that my my front stage was far too heavily delayed and that my rear stage needed to have a higher delay than my front, and that my sub should be time corrected to give it the illusion that it is coming from the dash.

Is there a specific philosophy in time correction that I should follow, or philosophy that yields the clearest image?

 

[gijoe]

I would say he is right that the center should be in the center of the dash, not in front of you. Because of the physical limitations of a vehicle you have to compromise. If you have the image in front of the driver you have a left stage that is physically less than half as wide as the right side. The center should be in the physical center of the stage, regardless of where you sit.

The sub timing should be corrected so that bass has the illusion of coming from in front of you, not behind you. Sometimes the right side of the front stage can be physically further away than the subwoofer is, so there is no fast rule that says the sub should have zero delay.

Do a search for ErinH, he has a good time alignment tool in his signature that works very well.

 

[Jepalan]

I'm interested in how others answer your questions, so I am going to sub to this discussion.
I am no expert on time-alignment. But my general understanding is that one should tune the sound stage to the center of the dash regardless of seating position.

For general time delay settings using measurements from driver's listening position to each speaker, I also thought the sub would get zero delay, the rears would get a bit of delay and the fronts would get the largest delay - this would put everything up front. The stage can then be adjusted with small left-right and level changes using Erin & Robert's tool.

 

[GLN305]

Every judging form I have seen for SQ comps has the center image anchored in the center of the dash/windshield with depth going from just on the dash to out on the hood.. I prefer the center to be there as well so that I can get a solid representation of the entire stage.

 

[Hanatsu]

Up front bass is mainly an level/EQ issue. Phase should be aligned to avoid a null around the crossover point.

OP: Yes - the stage should be in the (acoustic) center of the dash.

 

[Alrojoca]

I'm still learning this, it seems that hu's without an advanced DSP just have the cm and inches settings and no need for calculators or settings in micro seconds, but it helps with math and simple formula steps

Many do not TA the subs and the bass is still in the front, I had an HU with no TA and nobody could tell the sub was in the back. I agree it's more about the sub's phase that may give a different or better result. In my case either phase worked in the front and i could not tell the sub was in the back,the only difference was deeper bass compared to less but wider bass in the front

 

[gijoe]

As frequencies get lower they become omnidirectional, meaning you literally can not hear where they are coming from. Setting HPF low enough on subwoofers helps a ton, but there is usually harmonics and noise (rattling/resonating panels) that occur at the same time as the low notes, thus becoming localizable. I think true frequencies below about 200Hz are omnidirectional. That doesn't mean that you can use a HPF that high though, for reasons I've already mentioned.

 

[fcarpio]

I'm interested in how others answer your questions, so I am going to sub to this discussion.
I am no expert on time-alignment. But my general understanding is that one should tune the sound stage to the center of the dash regardless of seating position.

For general time delay settings using measurements from driver's listening position to each speaker, I also thought the sub would get zero delay, the rears would get a bit of delay and the fronts would get the largest delay - this would put everything up front. The stage can then be adjusted with small left-right and level changes using Erin & Robert's tool.

You want to delay all the speakers so that the sound arrives at the same time to the listener. If you sit to the left of the soundstage, the sound of the left speakers are going to arrive first. To achieve proper time alignment you have to have a reference speaker to align the rest with respect to it. That reference should be the furthest speaker, whether it is a tweeter, mid sub or whatever. Once you find this reference speaker you will delay the rest of the speakers so that their signal arrives at the same time. Then you can shift it to the left or right by playing with the times and levels. Once you do this you can even be more anal and fine tune the alignment by channel/frequency using your EQ.

 

[Hanatsu]

Time arrival isn't THAT critical with subs, getting the waveform in phase is pretty important though. You won't hear if the sub is off by a period in the lower octaves.

 

[14642]

Like others here, I think the center should be in the center. I suppose this will be an endless debate. I know plenty of people who say they prefer the center in front of them, and many OE systems are tuned this way.

Delaying all of the front speakers doesn't move them farther away or create a sense of depth. Delaying the left speaker doesn't move it outside the car and delaying the sub doesn't automatically make it sound like it's coming from the front. In fact, so long as the sub and the front speakers aren't close to 180 degrees out of phase in the range of frequencies they both play, delay on the sub is completely unnecessary.

Use Erin's spreadsheet tool. The tape measure is the quickest and easiest way to set delays. The only time that may not work precisely is with dash mounted speakers that face the windshield. IN that case, you may find that the better distance to use is from the speaker to the windshield to the listening position, but that depends on how high the dash speakers play. The higher they play the more likely you'll need to use the longer measurement.

 

[truckguy]

I've always struggled.with time alignment. I've used Erin's app and thought the missing time alignment on the furthest speaker was an error. Lol. I'll have try again tonight.

My other question is when I do see some people post their time alignments it seems like their numbers at times are extremely large. No where near what it would be by measuring with a tape measure. What affect does that have on your stage? Is there a calculator for that or how do you figure out those numbers in your own vehicle.

 

[Hanatsu]

My time alignment numbers is:

Left mid: 1,5ms
Right mid; 0ms
Left midrange: 1,8ms
Right midrange: 0,38ms
Left tweeter: 2ms
Right tweeter: 0,48ms
Sub; 0ms (90deg phase+)

It's within 10% of what I measured with a tape measure. If it's not, then you got reflection issues or something.

 

[truckguy]

Thanks Hanatsu. I've been following your suggestions in the optimal RTA thread and it's helped out a lot. Also downloaded the test audio files in there. Much appreciated!

 

[ErinH]

his numbers seem reasonable to me. even the sub.


center of anything is defined by the middle of the left and right boundaries. left and right acoustical boundaries in a stereo system. so, the center should be between those two points. typically this is close to the center of the car but there's always some give or take there because most car audio systems don't expand the same amount outside the of the physical boundaries (ie; typically the left may sound 'wider' than the right). this is reflection driven and due to the incidence angle so therefore can be tweaked via physical position/aiming but you are going to have to understand there are various tradeoffs as well.


phase/time on the sub is all about SPL at the crossover. the reason I delay the sub is because it's easier than delaying everything else to get the desired result.

 

[Jepalan]

Time arrival isn't THAT critical with subs, getting the waveform in phase is pretty important though. You won't hear if the sub is off by a period in the lower octaves.

Hanatsu and ErinH both stressed the importance of crossover phase alignment over time alignment for the subwoofer.

I have used ErinH's alignment tool, read his excellent SQ 101 post and read Hantsu's RTA and other excellent tutorial posts. I want to make sure I understand these concepts clearly.

Sorry - LONG post. Read on.

I like using examples and pictures. This discussion is about *subwoofer time-alignment and phase alignment of the sub-mid crossover point*. SO, I am assuming a mono setup for simplicity.

Figure 1 below shows the spectrogram of a single kickdrum *as recorded* in our media file. Frequency goes up the vertical axis and time goes horizontally left to right. Lots of energy is spread across the sub-mid crossover point when the mallet hits the drum. This energy decays into mostly low frequencies as time goes on.

Lets assume a sub-mid crossover point of 100 Hz for this discussion.
Let's also assume the sub is way in the back of an SUV and therefore is farthest from the listener.

Figure 2 below shows a hypothetical example of the same kickdrum as heard by the driver in the actual car install. In this case we see a phase null at 100 Hz and we see the subwoofer frequencies are delayed (maybe sub is in the rear quarter panel of an SUV so takes longer for sub energy to reach the driver than the mids in the front doors).

Now Figure 3 below shows the same kickdrum after additional delay is applied to the subwoofer until the phase null disappears. In this case, the sub is not time aligned, but has been delayed far enough to eliminate the phase null.

Finally, figure 4 below shows the same kickdrum, but this time additional delay has been added to all the front speakers until the null disappears, which results in elimination of the phase null *and* proper time alignment of the sub with the mids.

QUESTION: Am I correct in assuming that Hanatsu and ErinH are saying Figure 3 and Figure 4 are equally valid as far as car audio install is concerned?

(EDIT: forgot figure 4. added it)

 

[fcarpio]

^^^ Interesting.

 

[Jepalan]

Bump - anybody?

 

[Hanatsu]

Have some trouble reading the graph (is the x axis in milliseconds?). Is the sub delayed by 1s in the example? That is a problem if so. One period at 100Hz is 0,01s (10ms). If it's 1ms delay then it certainly doesn't matter, this is exactly the same as discussing how audible group delay is at subwoofer frequencies.

Sent from my Samsung Galaxy S5 using Tapatalk

 

[Jepalan]

Have some trouble reading the graph (is the x axis in milliseconds?). Is the sub delayed by 1s in the example? That is a problem if so. One period at 100Hz is 0,01s (10ms). If it's 1ms delay then it certainly doesn't matter, this is exactly the same as discussing how audible group delay is at subwoofer frequencies.

Sent from my Samsung Galaxy S5 using Tapatalk

Sorry, I should have been clearer in my description. The x-axis is in seconds, total time of graph is 0.5s. In the figure three, sub is meant to be delayed .010s (10mS) or 1 period of 100Hz.

BUT, please realize these are photoshop'd waveforms to illustrate a concept and create discussion, the only real measurement is figure 1.

I am familiar with some of the discussions of group delay at sub freqs, but will search a bit and do some reading.

One thought I had was to time align like this:

1) Use measurements to set delays on all drivers with farthest getting 0mS as previously discussed
2) Add a bulk delay to *all* drivers that puts the closest driver near the max delay capability of your DSP or HU. By adding the same amount to all drivers, you have not changed their relationship to one another from the measured distances (TA is relative).
3) NOW you can go back and tweak the subwoofer phase easily because you can move it forward & backward in time relative to the other drivers to find the nearest shift point that fixes any remaining crossover null/peak issues (In this example I am assuming sub is farthest again and was 0mS ref at start).

With this method you can ensure the sub is both phase aligned & time aligned, and are less likely to run out of processor delay.

 

[Jepalan]

This thread is a good read -> http://www.diymobileaudio.com/forum/966750-post1.html

<EDIT: removed my incorrect comments about room modes>