Southern California Test Bin - Car Audio | DiyMobileAudio.com | Car Stereo Forum

Reply
 
LinkBack Thread Tools Display Modes
Old 05-11-2011   #1
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Southern California Test Bin

DIYMA has been the home of many enthusiasts seeking quality sound for over half a decade. Many of the members have to come to treasure 3rd party tests and the scientific method. With the revival of speaker testing on this site, we can help too by testing the final product.

I have set up a few simple tests to capture a generous amount of data from any member vehicle. The last So. Cal. meet was a lavish expression of technical might, with a wide variety of testing gear. Since I frequent the local meets this is where most of the volunteer cars will come from. That's not to say I will only test So.Cal. vehicles, I'll test anybody's car, stock cars etc. With the latest generation of testing equipment we can perform the majority of speaker tests in a car with no headaches. The only requirement for the test is a standard CD player.

Next I will list the basic layout of the tests. I can submit a post later on data interpretation and cater to each of you as needed.



Data
Info Sheet

I will have these available at the meets as a printed PDF file. For those that want to type in ahead of time send me a PM so I can send you either the empty PDF or the Lyx file.
Enter power per sheet entry line in continuous RMS at the speaker terminal.



Frequency response


-SPL is minimum 90db A-weighted L+R (to get good noise rejection and emulate human hearing)

-use sine sweep for best quality data

-minimum 30 samples averaged (left ear, left front, straight ahead, right front, right ear; minimum 6 of each)

-3 tests total, left speakers, right speakers, all speakers

-1/6th octave smoothing (one step better than human hearing, ears pickup 1/3)

-5db steps for the DB axis

-12hz-20,000hz resolution



Harmonic Distortion testing:


-test at 90db A-weighted first, and then move on in 5db steps until the owner stops giving me the ok. All cars will do fine at 90db imo, even if your crossovers and EQ are all out of wack. Based on the first result I will go ahead and test at louder level if the owner agrees.

-test right speaker only. If I test the entire system some people will have their left speaker turned down lower compared to the right because they are closer. This will give them a disadvantage in HD. The same car could get better HD by simply equalizing the gains between left and right. 5.1 guys can get exceptions.
There is no reason to believe the left speaker will have different HD. The drivers and crossovers are the same. Reflections would be different but that reduces my testing variance as well, which is good.

-in the same plot: 2nd,3rd,4th,5th and 2nd-5th order harmonics along with the FR

-maximum HD test is at 120db, mic clips a little after that

-valid resolution 12hz - up to 10,000hz depending on the test (basically only the frequencies that matter, where HD is audible)


Miscellaneous

So at the minimum there will be three files per car: info, FR, HD @90db. This will be the base. From there it depends on how loud you want to be tested for HD. As always, all liability is on the owner, I can teach you how to click buttons if you want to do it all yourself.

Previously I said that people would be able to re-test if they want to tune based on the initial results and improve their car. To that extent I talked to Ant about letting me edit my own posts indefinitely in this thread. He agreed to take the editing requests upon himself so, while clunky there is a system to update now.

I will always appreciate your input on how to improve testing, so if in doubt, let it out.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 05-11-2011 at 03:12 AM..
cvjoint is offline   Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Old 05-11-2011   #2
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

My car as tested:



Power is continuous per line, so every tweeter gets 90w, and all three subs get 750w together.



HD @90db


HD @95db


HD @100db


HD @105db


HD @110db


Legend for FR Graph
Red Right Speaker
Blue Left Speaker
Black L+R

Legend for HD Graphs
Blue THD
Red 2nd
Purple 3rd
Green 4th
Blue 5th

Distortion should increase as the SPL increases. If it doesn't then there is too much background noise to accurately measure. You can see the 90db data is pretty noisy. From 95db on the distortion stays the same or increases as expected. BK noise was in the low 70 db unweighted, that was on top of a 6 floor parking structure with no other vehicles or people around.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #3
DIYMA 500 Club
 
Member
Join Date: Jun 2008
Location: Los Angeles
Age: 65
Posts: 347

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 142 CraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enough


iTrader: (0)



Default Re: Southern California Test Bin

Are Erin's tests, "A weighted" ?
I couldn't find any info regarding weighting in his threads.

I know the FR graphs that Andy has posted are "unweighted".

2007 Corvette Coupe
Audison Lrx 3.1k & 4.1k amps, JBL;MS-8,P462 Ctr,P562 Rears,Peerless SLS 6.5, Faital Pro 3"mids,Dyn; MD102, Morel Ultimo 10 sub w/Boxology Encl.,Kenwood DNX9140 Nav/HU.
CraigE is offline   Reply With Quote Quick reply to this message
 
Old 05-11-2011   #4
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

None of the tests are a-weighed per se. I only use the weights in figuring out what spl im testing at. I start in the spl mode with both speakers and turn the volume up until i hit at least 90 db a weighted. Then I move into FR and HD testing modes which are unweighed.
My tests, just like Erins are unweighed then. But why weigh the spl?

Erin tests single speakers in a controled environment, no eq, no reflections etc. I test a complete car, with eq if any, multiple speakers, with the interior reflections and all. In my car where the FR has been tuned flat +- 2.5db it wont matter if i weigh the response. In some cars however the subs are way loud, i've seen it as high as 30 db over the tweeters. if i dont weigh the spl that user will only get meaningful data for his subs. Besides, a-weighted is how we perceive spl.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #5
DIYMA 500 Club
 
Member
Join Date: Jun 2008
Location: Los Angeles
Age: 65
Posts: 347

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 142 CraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enoughCraigE will become famous soon enough


iTrader: (0)



Default Re: Southern California Test Bin

Quote:
Originally Posted by cvjoint View Post
if i dont weigh the spl that user will only get meaningful data for his subs.
Makes sense. Thanks for the detailed explanation.
I'm not following the above though.
How would a slightly different setting on the volume knob result in meaningful data ONLY for the subs ?

2007 Corvette Coupe
Audison Lrx 3.1k & 4.1k amps, JBL;MS-8,P462 Ctr,P562 Rears,Peerless SLS 6.5, Faital Pro 3"mids,Dyn; MD102, Morel Ultimo 10 sub w/Boxology Encl.,Kenwood DNX9140 Nav/HU.
CraigE is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #6
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Quote:
Originally Posted by CraigE View Post
Makes sense. Thanks for the detailed explanation.
I'm not following the above though.
How would a slightly different setting on the volume knob result in meaningful data ONLY for the subs ?
I might get down to my car one of these days for an example of this. I have a gain knob for the subwoofer up front that will do just fine.

Until then imagine Bob the audiophile. He has the subwoofer playing 30db louder than the rest of his speakers. I put the test signal on and turn the volume knob up until I hit 90db unweighted for the first tests. His subwoofer is going to be the main reference for the microphone. While his subs are doing 105db or so the rest of his speakers are still around 75db somewhere. The mic will give me an SPL of 90db so I start testing.

The problems he will have is that the other speakers are playing so low they will never have a chance of beating the background noise. The frequency response over 100hz will be composed out of the background noise so it will be misleading and the harmonic distortion will easily be in the 5% range even though his full range front speakers are barely breaking a sweat. Simply put the signal to noise ration will be too low for the midrange and tweeters at common SPL levels that I test everyone at.

If the A-weight is on when I find my DB levels on the volume readout the mids and tweeters will get a higher weight in the SPL readings. This will require a decent amount of power on mids and tweeters to get to the common SPL readings.

Of course I could do unweighted SPL and beat outside noise the usual way, with lots of power. But in a sense that's what I do anyway with the A-weight.

This is the most kosher way I could think of to find the reference output level for everyone's car (90db,95db, ... , 120db). It is also the industry standard:

Occupational noise exposure. - 1910.95

Note "measured on the A scale of a standard sound level meter at slow response." That's exactly what I do, A-weight and slow response to read it properly.

I'm open to suggestions. If you guys want to do unweighted I'm ok with that too. I just can't find any reason to do it.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #7
DIYMA 500 Club
 
Member
Join Date: Jan 2008
Location: South Bay, So Cal
Posts: 436

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 148 emrliquidlife will become famous soon enoughemrliquidlife will become famous soon enoughemrliquidlife will become famous soon enoughemrliquidlife will become famous soon enoughemrliquidlife will become famous soon enoughemrliquidlife will become famous soon enough


iTrader: (6)



Default Re: Southern California Test Bin

This is exciting. Heck, I'm about to put in Massive Audio RK6. I would love to get before and after plots between the new speakers and my old Polk Momos.

When is our next meet?!

Ed

I run a mail order aquarium products business. If you want the hook up PM me.
emrliquidlife is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #8
DIYMA 500 Club
 
circa40's Avatar
 
Member
Join Date: Jan 2008
Location: Anaheim, Ca
Posts: 1,151

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 154 circa40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enough

Send a message via AIM to circa40

iTrader: (29)



Default Re: Southern California Test Bin

I'll patiently wait in line

+1 on the next meet
circa40 is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #9
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

The next meet should be in the middle of June sometime. Ed, it will be pretty tough to filter out the effect of changing one speaker in a complete system. You have to keep crossovers, gains, EQ. etc unchanged from install to install. Are you sure you can prevent yourself from fiddling with it from meet to meet?

Vin you are IN. I still haven't checked out your "new" car. "new" to me.

Cajuner I've been toying around with the idea of cabin gain testing. Since my car is a convertible I've been testing it under various settings. I've learned the difference between top up and top down in the S2000 but I'm afraid I'm still a long way from doing a proper transfer function test. There are many issues, like the soft top folds over the subwoofers and acts as a barrier and then well, my soft top is vinyl, not a hardtop. Results may vary.

If anyone has a sub box that can be easily taken out I won't mind testing your transfer function. There are many of us interested in mapping out the cabin gain of various cars. It would be neat to do testing inside the community for this. My car is IB, and therefore a pain in the ass for working this out or I would do it with my gear.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-11-2011   #10
Moderator
 
Bikinpunk
Join Date: Feb 2007
Location: North Alabama
Age: 36
Posts: 18,143

Thanks: 317
Thanked 1,161 Times in 550 Posts
Rep Power: 26019 ErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond repute

Send a message via AIM to ErinH

iTrader: (100)



Default Re: Southern California Test Bin

FWIW, a home theater sub makes testing transfer function easy since all you need is a wall plug in and extension cord. It's how I do mine.
A small word of advice: remember that the in car measurement wil still be affected by the location and positioning, so try to put the sub in the car in the same fashion it would be installed. Basically goes back to the questions about "which way should I fire my sub", etc.

Cool thread.

Edit:
No, I don't use weighting in the car. I used to think it helped but it doesn't...at least not for RTA purposes. I also don't weigh speaker measurements. Just answering the Q.

Time Delay Calculator:
http://tracerite.com/calc.html
ErinH is offline   Reply With Quote Quick reply to this message
Old 05-12-2011   #11
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Good point on the positioning of the sub. We can test a few locations to see how that plays in too. Al we really need is someone with an easily maneuvrable sub box (and long wire, or extension). The HT sub idea would work too but we need to plug in and figure out a way to get the signal to it.

I'll see if I can show a comparison on SPL readings for various FRs using different weights. Either way all cars will get tested with the same technique so in that sense it will be fair. I just wanted to pick the technique that would give more useful results. It all boils down to having a good signal to noise ratio. For example, in my HD tests above the signal to noise ratio is generally to low below 40hz to properly read off distortion. Another clue is that distortion at say 12hz is lower as output is increased.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-14-2011   #12
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Pressurization: not sure how this plays in. From my previous testing in the hardtop Accord, the window down tests were about 3db louder in sub frequencies. The frequency response for the sub was virtually identical. Now in the S2000, every little change in window level changes the FR. I presume because the windows are a much greater part of cabin gain in a soft-top.

Using my own sub enclosure+sub: This will save me some time you are right, but on the other hand I can see these guys wanting the tests done with their own sub box if possible. It caters better to every individual car. You basically get to see your own sub perform outside the car, that can be neat to see. I don't mind testing twice for those guys. I do have a 10" Seas Prestige I can donate for the cause. I need someone to donate a small 10" box. I can wire it up and do the testing from there. If it's standardized and I use the same output level I can even overlap the results from everybody's car and make a cabin gain plot like Andy from Harman did. My car uses industrial quick disconnects so I can power this thing from my car in a heartbeat and just drop it in several locations in another car.

Testing cabin gain over 130hz up: not sure I agree with you on this one. Reflections play a big role once we move up in frequency. I will already give everyone a very good idea of the overall response of their car with reflections right where their speakers are from the HD and FR plots. To that extent if they have a sub in the cabin they will already have a test for how it behaves there. Now, the Seas woofer does extend up to 800hz at least and I don't mind testing wherever people want but if I were to standardize a few tests it would be like 5 locations in the trunk. Maybe you can go over this idea more in depth with me, but I don't see how reflections won't screw everything up in high frequencies. I suppose I can gate my impulse response but it's too small of an environment.

I like all the ideas guys, keep 'em coming. I have many tests in store that I haven't mentioned yet like reverberation, bass decay, phase, and time alignment. For now I figured I just start with the meat, FR and HD is an insane amount of information on the final response. The other stuff, like cabin gain, reflections, reverb, decay will tell you WHY you get the results you do but in essence all those effects are captured in the plots I laid out as an example. Once we saturate the thread with these tests we can add on more.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-19-2011   #13
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Robustness Checks

These are a few tests to showcase the reliability of the measurements under stress tests. Like any other scientific work stability tests are done to insure that the results are not majorly impacted by changing environment conditions , inherent variability in the software or hardware, test design, test parameters but the same predictions ensue in repeated samples. In practice, robustness checks are generally done to appease the journal referee (that's you here).

SPL weighting scheme

In this test I provide two frequency response graphs. One with my usual car tune done for the road (bass gain 0), and the second with the bass gain knob turned to max. In both cases I provide the measured SPL under 'no weight' and 'A-weight' conditions. I keep the volume fixed on the headunit to get a base output.



Business as usual FR (bass knob at MIN):

SPL no weight:
88.5db
SPL a-weighted:
89db

Bottom heavy scenario (bass knob at MAX):

SPL no weight:
94.5db
SPL a-weighted:
95db

Conclusion:
It seems like my Clarion bass gain knob can provide 10db worth of attenuation/boost for the sub amplifier. Under both weighing schemes the overall SPL increases 6db. This suggests the weight scheme does not affect the SPL level at which I will test, at least in the context of 10db variations in sub output.

Another result is that bass heavy cars should be tested at higher output to get a better S/N ratio for HD results to be more useful. In this test a 1% increase in sub only gain results in .6% increase in overall SPL. For very bass heavy cars 90db overall SPL might mean only 60db worth of output on the front speakers, that's competing against background noise.

Course of action: I can do a-weight or unweighted. It doesn't seem to matter much (.5db). I like a-weight because it's the industry standard, but I can be swayed either way. I want to see more opinions on this otherwise I'm sticking to the original choice.

Sample to sample variation in FR
Right speaker averaged frequency response (30+samples) sample 1.
Right speaker averaged frequency response (30+samples) sample 2. Taken 30 minutes later using roughly the same pic positions.



Conclusion:

With the exception of sub 30hz response where the variance was 2db, the two samples are almost identical. Over 30hz the output varies at most by 1db and on average by a fraction of a db. This suggests that the FR testing method is highly robust to slight changes in mic positions, temps etc I take this to mean I can give you reliable measurements that show up over and over even if I test your car in different days. If you want to tune your car between meets you'll see the change in your EQ. work and not in my testing procedure. This is clutch! Note: left speaker variance is twice as much due to shifting feet position. In some cars this variance will be greater (think kick panel tweeters) in others it will be as low as the right (think dash mounted midbass through tweeter).

Sample to sample variation in HD:

I forgot to turn the bass knob down in one case so judge 125hz up.




Conclusion:
A bit more variance than the FR, but very much alike from sample to sample. This is a test at 95db. At higher output HD plots look even more similar from plot to plot. I can't emphasize the importance of S/N ratio. We will test in a high noise environment. The better your sound deadening and higher the output level the better the quality of the data.

Variation due to test signal type:

I chose the sine sweep test signal for higher noise rejection and better quality results. Here I provide two FR plots, one tested with the sine sweep and one with pseudo-noise.



Conclusion:
Results are nearly identical, you most likely see the variation due to changing mic position or my body than variation due to test signal. I'll continue to use sine sweep because theoretically it is better even if we can't see the difference here.


If there are any other worries about testing quality let me know, I'll test if I can to see if the concerns are a real problem or not. Good or bad results I'll post.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-19-2011   #14
Moderator
 
Bikinpunk
Join Date: Feb 2007
Location: North Alabama
Age: 36
Posts: 18,143

Thanks: 317
Thanked 1,161 Times in 550 Posts
Rep Power: 26019 ErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond reputeErinH has a reputation beyond repute

Send a message via AIM to ErinH

iTrader: (100)



Default Re: Southern California Test Bin

Looks good to me. I appreciate the work you're putting in to document this. Sadly, I don't think many will really care as long as you can give them a graph. Do you plan to teach while you measure? Would certainly helps those whose cars you are measuring. Spreading the knowledge, dude. Four thumbs up.

Time Delay Calculator:
http://tracerite.com/calc.html
ErinH is offline   Reply With Quote Quick reply to this message
Old 05-19-2011   #15
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

I like to think the teaching goes both ways. I can certainly learn a lot from other people's cars, and after all people like Npdang are locals After all this time I still only know a fraction of his know how.

It would be really tough to help all people tune since that's a major time investment but I will most definitely go over the results with anyone. If we get to test all the cars out there and we run out of tests to do with this setup we can move to tuning. For now, posting results will give us an idea of the sort of target curves people like to listen to and the dynamic range capabilities.

I was thinking of testing my mom's Hyunday Accent. It should be fun to see what it can (or rather can't) do.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 05-24-2011   #16
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Tutorial

Frequency Response Plot

Plots sound frequency in HZ on the horizontal vs. output level or SPL on the vertical axis. We can generally hear music 20hz-20,000hz, and SPL changes of about 1 db.

Things to look for:
1. The left FR (in blue) should as close as possible to the right FR (in red). If they are close to each other the car will image better.

2. The FR (left, right, or together) should have little variance over the target response curve. This is where subjectivity will come in. From the point of view of reference sound and minimizing linear distortion the FR should be flat. Here is an example of a tune for a flat frequency response, RED is the target curve:



My belief is that "flat" sounds great in a low background noise environment and a properly mastered track. Think about a listening session of Diana Krall in the parking lot.

However, for one reason or another one may want a slightly different target response. I find it useful for combating road noise, which in my car starts 1,000hz and has a general rise all the way down to 20hz. Some may find it useful to equalize further to better suit a particular type of mastering.
Here's an FR tuned for the road, RED is the target response:



I think the important thing to take away here is the the FR should have very little variance around a target curve. The choice of target curve is subjective, the rest is not.

3. The summed FR (in black) should have low variance and fit the target curve well. This one is useful to pick up constructive or destructive interference. A scenario could exist where left and right speakers are fairly well tuned individually but when played together there are large peaks and dips. Not much you can do here but rebuild the speaker baffles to aim them differently.

In this test you can see that I have some sort of dip at 125hz when looking at the summed black FR:



Things that don't matter:
SPL level at which I test, as long as the background noise is very low by comparison. FR plots should look the same as long as there is a good signal to noise ratio across a wide variety of SPL levels.


Harmonic Distortion

Unlike FR, HD is a test for nonlinear distortion. It plots the fundamental FR and distortion separately. On top, in black, you have the fundamental, the pure sine sweep. On the bottom you find the 2nd order through 5th order total distortion THD in blue, and separately 2nd (in red), 3rd (in purple), 4th (in green), 5th (in light blue).

Example: a tweeter is fed a 2,000hz test tone. Ideally we would want it to reproduce only the 2,000hz tone. However, all speakers will produce harmonics, in this case the 2nd order harmonic is at 4,000hz. 3rd harmonic would be at 8,000hz, 4th will be at 16,000 hz and 5th at 32,000hz. The fifth harmonic is not audible.

Exhibit A
marks the difference between the fundamental which is at 95db at 2,000hz and the THD (blue line) which is at 55db or -40db:



Use this scale to convert from db differences between fundamental and chosen harmonic:

percent decibel
100 0
50 −6
20 −14
10 −20
5 −26
2 −34
1 −40
0.5 −46
0.2 −54
0.1 −60

That means that in my setup THD is -40db at 2,000hz or about 1% distortion. Always interpret THD reading at an output level, here my SPL is 100db for the test. At higher SPL the THD will be higher. This type of distortion increases with output.

I highlighted three areas of the HD plot where the distortion is highest. Often times you can pinpoint exactly what the problem is, sometimes however it's not that easy.

Exhibit B shows a rise in distortion at 125hz. This is most likely due to the high pass filter on the midbass drivers set at 100hz. I suspect a lot of car systems will show this behavior. The excursion requirements near the bottom of the frequency range are higher and distortion is likely to be higher as a result. I could move the high pass higher at 125hz but that would require a higher subwoofer crossover point and more localization problems. Here the tradeoff is between nonlinear distortion and sound staging (front to back localization).

Exhibit D
presents a similar challenge. My planar tweeter has a high pass at 1,600hz. The bump in distortion at 2,000hz is most likely due to the Kaladex diaphragm limits of excursion. Here too the tradeoff is between nonlinear distortion and sound staging (stage height).

Exhibit C
is somewhat baffling at first. You would think that the midbass driver would not distort significantly near the top of it's range. Unlike the previous cases the excursion is very low right in the middle of it's frequency band. It is also far from overheating at 100db, needing merely 7watts to get there. The culprit here is some heaving EQ. I boosted these frequencies by as much as 8db to flatten the frequency response. I dropped the EQ and the HD dropped as well. Here there is a tradeoff between linear distortion (frequency response) and non linear distortion (HD).

The basic idea is to measure the difference in DB from the fundamental to the harmonic. In loudspeakers you will find that 2nd and 3rd order distortion is the most abundant. For the most part I could drop the 4th and 5th harmonics from the test and the THD figures would be about the same, only slightly lower. There are higher harmonics than the 5th obviously but you can see they are not likely to enter into audible distortion significantly.

Not all nonlinear distortion is harmonic distortion. The speaker could produce artifacts that are not exactly multiples of the signal played. We are not testing for intermodulation distortion, as it requires more advanced measuring software. Think of the results as a lower bound on the distortion in your car. There is likely a bit more, even though HD dominates from what I've seen in third party testing. Generally, a speaker with high HD will have high intermodulation distortion as well.

Here is a website that you can use to test the audibility of nonlinear distortion:
http://www.klippel.de/listeningtest/lt/
I recommend you test the "full range" music selections (composite, not just BL and LE). Your ability will vary based on:
-Speaker quality
-Background noise
-Level of ear training
-Duration of the test

I can pick up distortion as low as 1.5% with test tones and 3% with music now, but I am getting better. Try for yourself.

Things that don't matter:
The shape of the fundamental. Look at the FR plot for the frequency response.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 05-24-2011 at 05:30 PM..
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-13-2011   #17
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

New Test

2005 Hyundai Accent 4 door GLS

Tested door speakers on the right:



Pioneer head - lowest offering, all stock speakers.



HD @ 90db


HD @ 95db


HD @ 100db



I will comment on this one since it is not a DIYMA member car.

First thing to note is that Hyundai is basically using woofers for a fullrange job. Note how the frequency response takes a dive after 3,500hz. Maybe they should have used a whizzer cone like I've seen on Hondas. The stock system needs a tweeter badly. Consider the size of this thing too, 5.25", the FR should be more extended.

On the bright side these things are hella efficient. Nothing beats the good ole paper cone. It's also stiff to the touch, which is more than you can ask from oem. 100db is easily attained with these guys and HU power. Because of their efficiency, they do fairly well in the HD department. Very clean midrange, 100hz up is decent for oem.

The takeaway message here is that OEM can do a good job with plain jane paper cones. Skipping on a tweeter however is bound to be very noticeable. Great sensitivity, average HD, poor FR.

Stick around for round two results where I install HI Vi Autosound coaxial drivers in the same car (left speaker in photo).

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 06-13-2011 at 01:21 AM..
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-17-2011   #18
DIYMA 500 Club
 
Newbie
Join Date: Apr 2011
Location: la
Posts: 53

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 119 93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of


iTrader: (3)



Default Re: Southern California Test Bin

Wow this is a really cool service thanks! I hope I can attend the next SoCal meet (I can bring a set of Hertz ML 165's if you're interested in testing them) As I understand you do both measurements inside a car with the complete system, and outside with the individual component? Sorry this is new to me.
93accordlxwhite is offline   Reply With Quote Quick reply to this message
Old 06-19-2011   #19
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Quote:
Originally Posted by 93accordlxwhite View Post
Wow this is a really cool service thanks! I hope I can attend the next SoCal meet (I can bring a set of Hertz ML 165's if you're interested in testing them) As I understand you do both measurements inside a car with the complete system, and outside with the individual component? Sorry this is new to me.
This gear is generally used to test speakers on a giant plywood board. Imagine the speaker is mounted on a endless wall so the front and rear wave are completely separated yet there is no actual box sealing the back. That sort of test would give you a standardized picture of what the speaker will sound like. That's basically what Erin does. You can't just hold the speaker on a table and test it.

What I do is test the entire car with everything installed. It's more of a final product evaluation kinda thing. If you are willing to bring a giant wall to the meet on which the speaker is mounted I won't mind testing it. Even then the background noise is a bit much in a parking lot full of enthusiasts to get any useful information. You can ship the drivers to Erin for this type of testing.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-21-2011   #20
DIYMA 500 Club
 
Newbie
Join Date: Apr 2011
Location: la
Posts: 53

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 119 93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of93accordlxwhite has much to be proud of


iTrader: (3)



Default Re: Southern California Test Bin

Oh I see, thanks for the clarification.
93accordlxwhite is offline   Reply With Quote Quick reply to this message
Old 06-21-2011   #21
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

New Test

2005 Hyundai Accent 4 door GLS

Pioneer head - lowest offering, 5.25" Hi-Vi Autosound coaxials in front doors. Compare to stock speakers test above.

Tested door speakers :



FR


HD @ 90db


HD @ 95db


HD @ 100db



I will comment on this one since it is not a DIYMA member car.

The FR is fairly flat without EQ. despite the fact that the crossover network is probably very basic and includes a simple HP filter for the tweeter. The 3/4" dome tweeter does a good job all the way to 20khz, although they could have used a higher sensitivity driver here. On the low end there is no more low end sensitivity than the stock drivers, kinda disappointing.

Efficiency is OK, not as good as the stockers. Cone stifeness is adequate. HD is poor imo, not better than stock. To me this is indicative of a very basic woofer. In fact I doubt there is much different from the stock Hyundai paper cones. Both are fairly tiny overhung simple motors. The coax I suppose has a disadvantage as it has to hold the tweeter in the middle whereas the stock woofer has all that surface area and no reflections.

So what do you get from a basic coax aftermarket speaker? A tweeter. I presume this is why it sounds a lot better. The benefits of 2 way shine through. On the other hand, never underestimate the paper cone.

For more info on the HI-Vi, I have a review thread:
https://www.diymobileaudio.com/forum/...evel-gold.html

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 06-21-2011 at 04:35 PM..
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-25-2011   #22
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

New Tests 06/25/2011



FR


[email protected]


[email protected]


[email protected]

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 06-26-2011 at 07:48 PM..
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-25-2011   #23
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

New Tests 06/25/2011

David - Jeep Grand Cherokee

FR


[email protected]


[email protected]


[email protected]


[email protected]


On this car we did something different. The HP crossover on the fullrange was raised from 200hz to 250hz.
[email protected]


Kenny (calicant) - Mercedes SL 550 stock Bose system

FR


[email protected]



With this car we did something different as well. There is a plot of the FR with the top down vs. top up. This is a hardtop convertible vehicle btw. There was lots of road noise so it's fairly noisy unfortunately.

Black is top down:



A reminder: There should be an increase in HD with higher test levels. If you see the opposite pattern that means the background noise is too high. Most tests at 90db are very noisy as you can tell. At 95db they start making more sense.

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"

Last edited by cvjoint; 06-25-2011 at 10:12 PM..
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-26-2011   #24
DIYMA 500 Club
 
cvjoint's Avatar
 
Enthusiast
Join Date: Mar 2006
Location: Californication
Posts: 4,701

Thanks: 7
Thanked 8 Times in 7 Posts
Rep Power: 214 cvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enoughcvjoint will become famous soon enough


iTrader: (18)



Default Re: Southern California Test Bin

Neel - Acura TL
Note: We couldn't use the balance control on this car today. Therefore, there is no separate left and right FR. The harmonic distortion tests are also a summed response.

FR


[email protected]



[email protected]


[email protected]


[email protected]



[email protected]

2015 Corvette Z06 Z07:
Head: H800 + Galaxy S7 edge with AptX and optical
Amplifiers: 4,000w+ Zeff class G/H power
2way + Pro logic II steering Satori 6.5" fullrange rears
High: AMT line array;Mid: SS 25W/8567-SE 10";Low: 4x Seas L26RO4Y 10"
cvjoint is offline   Reply With Quote Quick reply to this message
Old 06-26-2011   #25
DIYMA 500 Club
 
circa40's Avatar
 
Member
Join Date: Jan 2008
Location: Anaheim, Ca
Posts: 1,151

Thanks: 0
Thanked 0 Times in 0 Posts
Rep Power: 154 circa40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enoughcirca40 will become famous soon enough

Send a message via AIM to circa40

iTrader: (29)



Default Re: Southern California Test Bin

I cant wait to see if independent L/R EQ will help. After all this time with the P99, I just figured out that you can do individual L/R EQ hahaha, thanks for the help George!

...but first, I have to install different amps (yeah again)

Oh BTW, my mid HP is 315hz
circa40 is offline   Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Reply

Bookmarks

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Car Audio | DiyMobileAudio.com | Car Stereo Forum forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:
Location
Where you live
First name
Last name
Do you work in the 12 Volt industry?
Do you work in the 12 Volt industry?
Youtube Channel
Enter Your Youtube Username, And It will Be Linked To In Your Posts!
12V Industry - Company Name
Your company name.
12V Industry - Your Position
Your position in your company.

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Display Modes

Posting Rules
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Powered by vBadvanced CMPS v3.2.3

All times are GMT -6. The time now is 02:10 PM.


Powered by vBulletin®
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Search Engine Optimization by vBSEO 3.6.0
Powered by vBadvanced CMPS v3.2.3
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.

Ad Management by RedTyger

Home | User CP | Members List | New Posts | ITrader | Faq | Post Spy