DiyMobileAudio.com Car Stereo Forum banner
1 - 20 of 20 Posts

·
Registered
Joined
·
5,816 Posts
Discussion Starter · #1 ·
I'd like to open a discussion about path length difference and how it relates to what we try to achieve in our car audio systems. I've been a bit confused about how exactly to define PLD after doing some reading and hopefully this can shed some light on the subject and get you veterans to teach a "noob" some acoustic science.

Traditionally, I've always understood PLD as being the difference in distance/time between driver "A" and driver "B" in relation to the listener. However, after doing some reading and trying to learn some more about acoustics it appears that there is more to it. Specifically, direct sound versus reflected sound. Since sound radiated from a speaker does not travel in a small straight laser beam, but instead disperses out from the cone, there are a few things to consider when installing drivers in a car. One of the main things would be reflection. When we install a driver near an interference object (all those hard surfaces on the vehicle), you end up with the signal reaching its destination at different times, even from the same speaker. This is because you will get the "direct" sound which travels straight from point a to point b, and you will also get "reflected" sound which bounces off of objects in your car and eventually makes is to your ears.

Does this also = path length difference? The difference in time between direct sound reaching the ears, and reflected sound reaching the ears?

Here is a nice article I have been reading...
Acoustics and Psychoacoustics: Introduction to sound - Part 6 | Audio DesignLine


To expand on this as well, since Time Alignment cannot fix the path length difference between direct and reflected sound, it makes on-axis aiming with as little interference between the driver and listener appear to be the best solution for minimal reflection. This is why I have been testing a-pillar/dash pods with fullrange drivers aimed on axis to the listening position.

I have a few other questions along these lines as well, but since this is alot to go through, I'll wait for responses first before I dig deeper.

Thanks
 

·
Registered
Joined
·
513 Posts
I think it's a great idea to discuss these since these are the two main topics that really differentiate car audio from mainstream acoustics (home theatre, studios, even concert venues, etc.).

One thing I'm a bit confused about is whether minimizing reflections is truly ideal in a car. At first, a lot of the acoustics white papers etc. I looked at talk about how important it is to have the direct sound be much more prominent (higher amplitude) than the reflected, and to avoid reflected sound and diffraction.

BUT I've also read some sources that imply that we may actually WANT diffraction, reflections, and diffusion in order to create a sense of depth. In other words, human beings seem to interpret sound that has been diffracted off surfaces, with significant reflections following the initial direct sound, as coming from further away. This seems to be somewhat confirmed by some of the cars I've listened to -- for example, domes deep in the kicks (which causes a lot of diffraction and later reflections) seem to sound further away to me than ID horns (which maximize direct vs. reflected sound). That's totally subjective though and it might just happen to be the cars I've heard with different setups tend to be tuned differently.

In other environments (home theatre, etc.) there's no concern about pushing the stage further back, because you're far enough away from the speakers. In a car, maybe we're so close to the speakers that we might need the diffraction and reflections to push the stage further away.

Again, I'm kinda confused on how these fit together so I'd love to learn more. Andy, I'd love to get some of your knowledge on this. Thoughts?
 

·
Registered
Joined
·
5,816 Posts
Discussion Starter · #4 ·
Those are some of the things I was thinking as well when raising this topic dbiegel. Perception of depth/space based on reflection and delay of direct sound vs reflected sound.
I would think that while this may make there seem to be more depth, that you would end up losing tonality and focused staging as a result, no ?
 

·
Registered
Joined
·
513 Posts
bump. Anyone? anyone? Bueller? Bueller?
 

·
Registered
Joined
·
13 Posts
Here's a good google book that gets pretty deep into direct vs reflected sound. I did a search in it for direct vs reflected sound, I've never linked a search in a book so I don't know if it will work right or not but chapter 3 is what I think you'd be interested in.

This direct vs reflected idea is the basis behind "open baffle" speakers (that is to say that people actually really enjoy reflections with their music, which is contrary to a lot of conventional audio design). Probably preaching to the choir, but check out linkwitzlab.com, specifically the link for "issues in speaker design". Linkwitz's Orion is, from what I have seen, the reference design for open baffle speakers. The reason I am talking about these speakers, which normally operate with a LOT of room is because of the fact that they embrace reflected sound rather than try to eliminate it and some of the concepts could be useful in a nearfield like a car.

Here's another link to letters by Linkwitz and Mike Gough of B&W. It's mainly talking about measurements on a Magnepan loudspeaker and some explanations for a couple of ugly bumps in the nearfield response graphs (click the measurements link at the bottom of the page). Gough's comments about dipole roll-off are interesting but not terribly useful for this discussion, but Linkwitz has a couple of things to say about nearfield environments. Now that I think about it though, a dipole is really just clever usage of reflected sound, which is what we're talking about anyways right?

I may have missed the point of the discussion, but I think this is what you guys are talking about. I don't claim to be an expert on the subject, but it is an interesting subject and there is a good deal of literature if you're talking about home audio, but almost nothing about cars. Perhaps purposely leaking the rear radiation to the front in an a-pillar configuration could yield some good results in deepening the soundstage?
 

·
Registered
Joined
·
3,903 Posts
At Paisley Park studios (Prince) they have a recording room made out of granite that has a cling that raises and lowers to alter the reverb/echo/reflection. So if you're listening to a Price album with this effect in a car, you're listening to purposely reverberated recorded music in a small room full of more reflected sound. And people claim to know what this "SQ" should sound like......

All I can say from my own experience is that in order to know the truth, you really need to significantly alter the space if you truly want to hear accurate sound reproduction outside of actually talking to the artist and seeing the recording studio. Those that do not do this, yet make claims that such and such speaker doesn't sound "right" or "accurate" need to find a different hobby.

Here's a small example of what I'm talking about. A nice chuck of OCF strategically placed where the phantom image and stage is in my car with corner loaded RR tweeters...



I've tried several different places and positions and it's crazy how much of what we hear is incidental sound (ie hearing the car) and NOT coming from the speakers themselves. I'm most definitely not an expert, nor do I understand half of what I read when it comes to the technical aspects, but I'm not afraid to get out and try and fail 10x's for every success.

I don't think it's even possible to know for certain what your speakers *should* sound like in a car. Maybe in a home or studio, but not a chance in a moving vehicle. Car audio "SQ" competition and judging is altered reality. If you're into making your car sound like some of these top vehicles that just sit still and sound good, you're insane!! :)
 

·
Registered
Joined
·
172 Posts
I believe in a car what can muddle up your soundstage is the timing of what reaches your ear first; direct sound or reflected sound. This is something that time alignment can't fix. Seems the if majority of what reaches our ears first is direct sound we are far better off than reflected sound reaching us first, unless that is what we're striving for. I originally had my mids in my dash firing up at the windshield so most of my midrange frequencies were reflected and whatever was coming from the cone at 60° off axis was secondary. After testing dash pods aimed much more on axis I noticed it really helped my imaging and decided that was the better choice.

If anything it goes to show how important it is to choose drivers based on their on/off axis characteristics and where we plan to install them. If you want to bounce your tweeters/mids off of the windshield then you want drivers that don't perform well off axis so you don't get a lot of extra "direct" sound. Whereas if you plan on aiming drivers on axis then a driver that plays well off axis will just add more reflection issues.
 

·
Registered
Joined
·
5,816 Posts
Discussion Starter · #9 ·
The way I'M thinking of this is that direct sound is point to point, from the cone to your ears. So even if it is in the dash firing to the windshield, the direct sound is what goes from the cone to your ears...everything that hits the windshield would be reflected sound, and would therefore arrive later. As far off axis as you can get, there will always be some "direct" sound that reaches your ears first.
Think about if you layed on your back in the desert. You have one midrange above you firing straight up at the sky...no reflections, right? Would you still hear it? Sure.

Acencsu: Thanks for the link. Theres some good info in there I'm reading through. :)
 

·
Registered
Joined
·
172 Posts
Think about if you layed on your back in the desert. You have one midrange above you firing straight up at the sky...no reflections, right? Would you still hear it? Sure.
Of course you'd hear something, though it would most likely be very diffuse and not much in the high frequency department. It would come down to how much that diffuse and muffled high frequency sound would ultimately influence the sound coming directly at you from a speaker on axis. Think of a dome midrange on your dash firing upwards at the windshield. That would be a reflection nightmare due to the excellent off axis response and whatever is bouncing off the windshield, dash, a pillars, etc...

The way I'M thinking of this is that direct sound is point to point, from the cone to your ears. So even if it is in the dash firing to the windshield, the direct sound is what goes from the cone to your ears...everything that hits the windshield would be reflected sound, and would therefore arrive later. As far off axis as you can get, there will always be some "direct" sound that reaches your ears first.
This is the point I was making. A speaker that performs poorly off axis won't have as much output and high frequency information if you were to listen to it say 60° off axis. So while you are getting some of the point to point sound from a speaker aimed at the windshield, assuming it was a poor off axis performer, the sound coming off of the windshield should be of much higher ouput along with higher frequencies so it would "hopefully" have more of an effect on what you're hearing than the direct sound coming at you off axis. If you were to use a driver that performs well off axis in this same scenario then you would be getting a lot more direct sound and higher frequencies coming directly at you along with the windshield reflections which would have a greater chance at affecting the overall sound.

There are always reflection issues even in a large room. It would come down to how much would those reflections ultimately affect the sound based on their intensity, arrival time, etc.. Unless we listen to a speaker that has no baffle while we're floating 1000 feet in the air there will ALWAYS be reflections. I think that we just try to minimize their impact with the proper driver in the install location of our choosing. Or maximize their impact by using the reflections purposely in our favor.
 

·
Registered
Joined
·
5,816 Posts
Discussion Starter · #11 ·
Ok, so we are on the same page then.

What I dont understand then is when someone says that they mount their drivers off axis for more depth. If mounting on axis yields the most accurate response from your drivers and minimizes reflection, wont the audio be truer to the original recording than trying to "simulate" space through reflection?
 

·
Registered
Joined
·
3,033 Posts
I'm obviously new and don't have much empirical data to add to this issue. However, I'm just now getting to tune my installation and I'm having a terrible time with reflection vs. direct path. I've noticed that T/A doesn't suffice my need as much as I'd like and that's most likely because I'm shooting for a system based on off-axis response(assumption).

With that said...off axis angles are different from each side and therefore L/R tuning seems more important than L=R.

Maybe the correct variables we should be looking at is speaker response/design vs. application and the necessary install techniques as well as tuning techniques that are needed to resolve sound issues.

On-axis would seem to benefit greatly from T/A. Off-axis?...depends on what's going on with reflection.

Does this seem coherent? I apologize if I'm off base I'm really trying to learn about all of this. Thus far, it seems that if I had used an on-axis approach it would be easier to tune instead of trying to attenuate and "fix" off-axis waves that are mingling with reflections.
 

·
Registered
Joined
·
172 Posts
T/A works for both on and off axis applications it's the reflections that kill. Theoretically if you were to do the bounce off the windshield install as I used as an example you could attempt to T/A to the distance including that reflection. So instead of measuring to the driver itself, you'd measure the path that the reflecting sound would travel. That way you're not T/A to sound that hits you first(from the speaker) and actually T/A to the reflected sound. That would be in theory only though. If you had a very diffuse point source like a dome midrange then you'd have to T/A to the driver.

T/A is great for PLD, but does zero for reflections. That has to be controlled primarily through install. In my previous example you could put some sort of foam barrier between you and the speaker so most of the off-axis(sound we don't want) would be absorbed and the sound reflecting off of the windshield would still reach our ears. Can't block it all, but could help and would be worth testing.

Bottom line is first and foremost we should always install our speakers with the most equal PLD's we can achieve(dash pods for a deep dash, kick panels, etc..) then use T/A to bandaid it even more. Then deal with reflections separately. I have my mids/tweeters installed in dash pods aiming between my front seats so most of the sound isn't reflected off of the windshield and the majority of the sound is coming straight towards me. I will be wrapping my a pillars with grill cloth soon to help combat reflections from my tweeters.
 

·
Registered
Joined
·
172 Posts
My guess is that the idea is that the sound would reach their ears after it's reflected so the PLD is actually longer? Using my windshield analogy it could create a few more inches of distance by bouncing off something than coming straight from the point source. However it'd have to be done right as the unwanted reflections could destroy your imaging, cause comb filtering, etc.. in an attempt just to make your soundstage "deeper". This method would require you to T/A the path of the reflected sound instead of the actual driver or it probably wouldn't work right.

What I dont understand then is when someone says that they mount their drivers off axis for more depth. If mounting on axis yields the most accurate response from your drivers and minimizes reflection, wont the audio be truer to the original recording than trying to "simulate" space through reflection?
 

·
Registered
Joined
·
1,488 Posts
Guys you really need to look at the radiation pattern as a function of freq. If you do, you will likely decide that on and off axis only have effects in certain freq bands (beyond where the output is omnidirectional). IMO - This is critical to understanding how to maximize SQ in a highly reflective environment like a car.
 

·
Registered
Joined
·
3,033 Posts
Guys you really need to look at the radiation pattern as a function of freq. If you do, you will likely decide that on and off axis only have effects in certain freq bands (beyond where the output is omnidirectional). IMO - This is critical to understanding how to maximize SQ in a highly reflective environment like a car.


Exactly, that's what I'm realizing now that I'm at the point of tuning. Attempting to make assumptions about frequency response from the different radiating angles is almost pointless. I'm finding that my results are better off without utilizing T/A at the begining. Therefore, I've tried tailoring the T/A for "final" adjustment.

It helps and it doesn't...it all depends on the frequencies/song. That being said, I haven't added EQ to the tuning process yet either. Every time I go through the process I'm attempting to single out an attribute I dont fully understand so that I can expand my understanding through experience and hindsight.

More and more it seems that T/A is best left for the final tweak. However, depending on the installation, axis response, and reflections it may not suffice your needs.

Off-axis installation seems to add so many more variables when trying to understand PLD + reflections.
 

·
Registered
Joined
·
522 Posts
I believe Time alignment should be the very first tweak, not the final tweak. If you EQ first, then time align, I guarantee you will be working on the EQ again. If your tweeters, etc are out of phase because of time alignment issues, that will effect both stage and tonality.

The best way to time align while accounting for reflections is to use a measurement software, like MLS, that can actually show you if the drivers are time aligned correctly.

whether a driver is on or off axis certainly does make a big difference and has to be taken into account in many ways. Let me give an example. I have Dynaudio dome midrange in the kickpanels, tweets in the a pillars. The driver side dome is terribly off axis. When I measure the response on right vs left midrange, there is a huge difference between the two, most of it having to do with the fact that the midrange that I am hearing from the driver dome is mostly off axis. The on axis output of the speaker is bouncing off the console and getting to my ears much later.

After setting correct time alignment, I EQ to try and get the response of the left and right to match as much as possible, attempting to use cuts rather than boosts. Not trying to EQ for tonality yet, just getting the two to match as much as possible.

When I RTA the system, one thing that I noted was a significant dropout in the upper midrange. Dynaudio says that the crossover between midrange and tweeter should be 6 db per octave at 4KHz. I am sure that is probably the correct crossover point if mid and tweet are mounted close to each other on a baffle and are on axis with the listener. But currently for me, that is not the situation. The tweets are slightly off axis in the a pillars and the mids are in kicks as previously stated.

When I look at the 60 degree off axis response for my midranges, and I can quickly see why I have a valley in my system response in the upper midrange. It is because the off axis output of the midrange suffers severely. For now, I am not moving the midranges but plan on trying them in the a pillar eventually. Instead, taking into account the response that I am getting, I cross the tweets over at 3K instead of 4K. I check the RTA and viola!, I just took care of the lull in the response without having to boost the mids on the EQ.

Soundstage especially presence, immediatley improved as well.

This is where understanding what is actually going on with your drivers in your particular setup can make a huge difference in the tuning.

I guess my point is that there are so many variables in the car, we have to be willing to look at all of them and understand what is going on if we are going to use all of the tools at our disposal to end up with a good sounding vehicle.
 

·
Registered
Joined
·
3,033 Posts
I think the difference is that I'm not using an RTA. That's why I said trying to make assumptions about my reflections are pointless :(

I'm using active now instead of passive to do exactly what you mentioned. I'm going to attempt to resolve the issue at hand through higher cross over points for my tweets.

I'll try to think about the use of T/A at the begining. It would seem easier to setup the stage and imaging prior to using T/A and then move it? However, if reflections are ruining the ability to "move" the stage maybe it will help me take into account those reflections earlier on?
 

·
Registered
Joined
·
522 Posts
If what you are saying is work on speaker placement to get the best sound prior to time alignment, then sure, that is certainly the best way to go. But before any other tuning with an EQ or tweaking crossover points to deal with other issues, I would time align.
 

·
Registered
Joined
·
3,903 Posts
If what you are saying is work on speaker placement to get the best sound prior to time alignment, then sure, that is certainly the best way to go. But before any other tuning with an EQ or tweaking crossover points to deal with other issues, I would time align.
You do this with frequencies that we know that we are most sensitive to amplitude and not time? Ie ILD's vs ITD's? Ie above ~2khz (tweeters)?

Make your speakers produce sound that hits your ears within .0001 ms of each other, but you still cannot remove the oddly shaped room you're hearing also.

IMO, it's not that reflection is the enemy or what's the most important way to tune, it's realizing how much compromise you have to make to get what you want out of your system. It's all about expectation level...and I see people struggle with this all the time in this hobby.
 
1 - 20 of 20 Posts
Top