As of now, I have T/A setup by physically measuring the distance of all speakers to the driver's head. (tweets/mids in a two-way active) Then, I subtract the greater distances on the passenger side from the shorter distances on the drivers side. This gives me a net number to delay the passenger side tweet and another delay for the mid. Therefore, as of now, I have zero delay on the driver side and the net delay (drivers side minus passenger side) to delay the passenger side. It sounds good and is relatively centered although not rock as vocals wander from left to right and the imaging sounds centered high but on the rear view mirror not out on the hood centered like I want. Reading through this thread, I realize my methods are very flawed so I tried the OP instructions but...If you're working with a 3sixty.2 you may struggle to find the sweet spot using this method, as there is only 5'-0" of delay available from the interface.
The wavelength associated with the crossover point you select may be much geater than 5'. Depending on the phase shifts from the various filters you may have in play, it can shift you outside the available delay window.
With the 65 hz crossover point I use the wavelength is 15'-10", so you can see how hard this could be. Even flipping polarity on the sub is only going to help by 1/2 wavelength, or 7'-1 1", which still might be out of reach.
You could get lucky, but in my case I needed 6'-0" on the passenger side and 7'-6" on the drivers side. The results are epic and definately the biggest breakthough in SQ I've ever had in my setup, and this thread was the driving force that got me to dig into it. Thanks Greg!
If you're still using a 3sixty.2 and need more than 5'-0" of delay feel free to pm me. If there's enough interest I'll put together a thread on it.
Yes, I searched AND googled for pink noise track to use this method since the one on the first post isn't available for download. Anyone have a download link for a pink noise track, white noise and frequency sweep so I can try this out?
Was going thru this for the first time and found this little gem. Gotta hand it to ya Erin, this is pretty ingenious. I don't have any measuring equipment so have to do it by ear (and can't seem to hear the Doppler effect Greg is talking about with pink noise) and while it's harder at the sub frequencies, it's very easy to do in the upper end. I played a test tone at my mid/tweet crossover point, left side only with the tweeter out of phase. By adding delay to the tweeter I could hear the tone get louder and then quieter (I had to turn my head slightly to hear it). I left it at the quietest setting and tested it by flipping the tweeter back to normal phase. In phase it was much louder and quieter out of phase.been meaning to post an idea for a long time now but never did. Finally put it to the test tonight. You can see my results here:
I hope this helps.
OP here. You're doing it right. Every single system is different, but 10cm doesn't sound unusual to me at all for tweeter delay. What will tell you once and for all if you're correct, is that the "repeating pattern" will exhibit the Doppler effect. For example, if represented by numbers, the Doppler effect will rise and drop in frequency like this: 4-3-2-(1)-2-3-4 OR 1-2-3-(4)-3-2-1. Find the midpoint, and you're at optimal alignment. See upcoming post for more info on sub/mid.I just noticed this thread and have been testing the method for a couple of times now. I'm using a Pioneer 80prs from which I set the delay in cm. At first I couldn't hear anything when testing with sub and passenger mid. I then tested with passenger mid and tweeter with both delays set to max (400cm) first. Here's what I noticed: when I started to add delay to the tweeter quickly, I noticed a clear pattern which was occurring very frequently. I then started to locate the midpoint of the first pattern and found that at 390cm, the extra noise disappeared. Then I started looking into the next ones and found these spots: 377.5cm, 366.25cm, 342.50cm and so on. All of the quiet areas are only 2 steps wide (one step is 1.25cm).
Could someone comment if I'm listening to the right stuff or not? I'm curious about the first spot as I understood that that is the one that should be used, but it's only 10cm delay. Previously I have had around 40-50cm of delay between the mid and the tweeter and noticed that many people here also have similar kind of delay. So am I supposed to set delay of only 10cm or am I completely lost with the method?
After practicing with mid/tweet, I returned back to the sub and passenger mid. I'm still having problems with hearing the effect. Based on my practice, I however did think that I heard some spot where the background rumble was a little more quiet. It's really hard to determine the points where the sound changes, but I ended up with values 172.50cm and 138.75cm. So I took the midpoint of those and set it as the delay to the passenger mid (156.25cm). Does that look ok, meaning the distance and the width of the area? I had little time to listen to the drivers side mid and couldn't locate the area yet. I have to keep on trying.
Any comments are appreciated.
Is this method of T/A applicable if using a CTR channel? Would the front midbass be the point of reference in this case. Your instructions are great!More info for those that are having trouble aligning sub/midbass. First, understand what makes the sub/midbass alignment so difficult: you need to hit the delay key many times to reveal the Doppler effect. Consequently, you can't hear a difference in sound each time you press the delay button. For example, aligning the mid/tweeter may occur like this:
Where 5 to 1 represents the Doppler effect getting lower in frequency with each delay step, and 1 to 5 represents a rising frequency. The midpoint (1) represents optimal alignment. Now, in contrast, the sub/mid bass alignment might look like this:
In short, the process is much more “spread out” at low frequencies; this is why it's so difficult to hear changes. With this in mind, here are some setup tips to make the Doppler effect easier to hear. During alignment, play the sub and mid bass full-range for maximum bandwidth overlap. More overlap = stronger Doppler effect. Next, adjust sub and/or mid bass levels so that the rumble from each are roughly the same volume. Now turn up the HU volume pretty loud, to make the Doppler effect easier to hear. If your mid bass is straining, apply a 50/60Hz high pass to cut low bass. The car should be off so as not to add its own low frequencies to the mix. Now adjust mid bass delay quickly, to make frequency changes more apparent. In doing so, the pattern may occur more like this:
First things first, you just want to FIND the pattern. Once you do, slowly and methodically find the midpoint; count the number of delay steps in the pattern, and divide by 2. Once you're successful, further refine your alignment by reducing the amount of sub/mid bass frequency overlap. With practice, you will be able to align them with their normal crossover frequencies - and this is when you will realize the full benefit of the method. The results are well worth the effort.
As a counterexample, just use 150Hz as your frequency, 4 cycles as your time, and calculate x ms of "delay." You'll see it will be a smaller number than 40ms. So if you delay by 40ms, you may be 1440deg out of phase at 100Hz but you will be more than 1440deg out of phase at 150Hz. Basically you wouldn't get that "suck-out" sound if things were not actually aligned correctly at all frequencies (unless you have some really bizarre/horrible group delay causing massive phase issues etc etc etc).For example at 100hz, 4 cycles are 40ms
Thank you so much. I didn't look at it that way... however now that you pointed it out, its so obvious. Lol. CheersSure - if you followed these instructions using only 100Hz tones you would definitely have this issue. Since you are listening to pink noise you will be listening to all frequencies from 20-20k.
If you are 1440 degrees out of phase at 100Hz with 40ms of delay you will be more or less out of phase at other frequencies with the same level of delay.
As a counterexample, just use 150Hz as your frequency, 4 cycles as your time, and calculate x ms of "delay." You'll see it will be a smaller number than 40ms. So if you delay by 40ms, you may be 1440deg out of phase at 100Hz but you will be more than 1440deg out of phase at 150Hz. Basically you wouldn't get that "suck-out" sound if things were not actually aligned correctly at all frequencies (unless you have some really bizarre/horrible group delay causing massive phase issues etc etc etc).