DiyMobileAudio.com Car Stereo Forum banner

321 - 340 of 837 Posts

·
Registered
Joined
·
3,277 Posts
Re: Jaguar XKR, Jbl, Audiofrog, Acoustic Elegance & iphone source.

Don't ever go as close as 1cm of the drivers with the APL measurements, you need to keep the mic away from the direct response of the drivers. The plots in post 53 show that you got interference and the "noise" in the IR at 200ms show that your measurements are corrupt.

Mic should be pointed towards windshield, between 15-50cm, move the mic like you paint a picket fence up and down going to the next one etc, try collect 10-15points on each vertical line...

I will do a video on this soon.

Nice build btw ^^


Sent from my iPhone 6 using Tapatalk.
Ever make that vid?

Sent from my SGH-T999 using Tapatalk
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #323
Im planning on doing something very similiar in my car. My inner sheet metal is damn near non existent so i should have good room. Ccf on the whole inner side of the door card. Ccf on the edges of the inner door metal. Fiberglass a negative of each. Epoxy both together. See what kinda volume im looking at.

Any tips or pros/cons with this approach?


Pic stolen from google but if you look past the plastic sheet. You can see the giant hole.

Mine is currently very well deadened/sealed now but I just wanted to see if I could make an actual enclosure. Feel free to pm me if you dont want to clutter your thread




You're lucky with all that space! Seems there's is a transverse in the door no?
Tips? well here's what I learned for my next project:

- Wood is way easier then fiberglass, and in order: plywood > mdf > fb
A shell enclosure of this size in 2 parts like you plan will be super heavy. Especially with large flat surfaces that would need big thickness or reinforcement. A wood enclosure with the right bracing is a bit more work on planning, but a lot of weight saved in total.
Same for time, burden etc, wood is much nicer to work with than FB I think. Of course it can be a combination if it makes sense on complex areas (but if you don't need big size, maybe you can copy Gary's doors?).
- See what you can leverage from the existing stuff: like using the door panel pocket volume, the door panel shape etc to create one wall already. But using the door panel, as a side of the enclosure, will produce vibrations that you'll probably feel in the armrest or near areas. That's why I finally isolated mine completely from what I can touch.
- Target the biggest volume possible for some flexibility later if you want to swap drivers. You'll spend a lot of time here, so why not be ready for later? And maybe think about a system that would allow bigger driver. But don't chase the last cu in, keep it simple.
- Carefully plan the way you'll fill / close / mount / re-open if needed. Check 100 times during the process if everything fits, glass motor/rail clearance, door panels and switch mounting, door closing & hinge effort etc
- If you can, and if your doors are symmetrical, build both sides in same time, helps on measurements and saves time.
- Be ready to restart from scratch, in fact plan for it :D. At least one quick draft box would teach a lot on issues you'll have to face.
 

·
Registered
Joined
·
3,277 Posts
You're lucky with all that space! Seems there's is a transverse in the door no?
Tips? well here's what I learned for my next project:

- Wood is way easier then fiberglass, and in order: plywood > mdf > fb
A shell enclosure of this size in 2 parts like you plan will be super heavy. Especially with large flat surfaces that would need big thickness or reinforcement. A wood enclosure with the right bracing is a bit more work on planning, but a lot of weight saved in total.
Same for time, burden etc, wood is much nicer to work with than FB I think. Of course it can be a combination if it makes sense on complex areas (but if you don't need big size, maybe you can copy Gary's doors?).
- See what you can leverage from the existing stuff: like using the door panel pocket volume, the door panel shape etc to create one wall already. But using the door panel, as a side of the enclosure, will produce vibrations that you'll probably feel in the armrest or near areas. That's why I finally isolated mine completely from what I can touch.
- Target the biggest volume possible for some flexibility later if you want to swap drivers. You'll spend a lot of time here, so why not be ready for later? And maybe think about a system that would allow bigger driver. But don't chase the last cu in, keep it simple.
- Carefully plan the way you'll fill / close / mount / re-open if needed. Check 100 times during the process if everything fits, glass motor/rail clearance, door panels and switch mounting, door closing & hinge effort etc
- If you can, and if your doors are symmetrical, build both sides in same time, helps on measurements and saves time.
- Be ready to restart from scratch, in fact plan for it :D. At least one quick draft box would teach a lot on issues you'll have to face.
I always thought fiberglass was used because of its light weight?

Its too cold and its gonna be too cold to fiberglass for quite awhile. Lame. I have projects to try!

Sent from my SGH-T999 using Tapatalk
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #327
Yep, I'm not a specialist on all this, just my experience with small/mid size boxes. Mdf and fb are probably close, it depends more of the shape.
But plywood is for sure another world. Much happier with the last ones, for sound and for weight.
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #328
Btw here's the sub box now:


made of 2 of these: Atrend 12SKS Slim Series Sealed Subwoofer Enclosure
Had just one before, now it's maybe 2.4/2.5 cu ft

I will measure and tune a bit more, but I'm not quite satisfied.
The box is braced a bit but is still too soft. But not much I can do externally to reinforce it now, not much space left with the car frame, the seats etc.
It's sitting on an old mastress so vibrations are filtered (it's strapped to the back).

So at the end it's only slightly better, impact and integration are great, really awesome, but it feels a bit slow sometime, or undefined on the bottom end.
Maybe I need to rise the HP, check for distortion?

It's crossed at 120hz, and being here at maybe 1ft of my head it can also be disturbing. Stage is sometime pulled back but it's a trade off.
But on big impacts it can be painfull to my right ear, too much pressure.
I wonder... if it was on the left, right behind me and not in the center, could it be better? or reversed, firing to the back panel, I have only 1" clearance there for the cone due to some chassis braces.

I may try my old ib12au here, but not sure the box is big enough for it. On winisd there's not much difference compared to the gb12.
Or maybe another sub? I searched a bit but don't know much about SQ subs.
Maybe IDQ, treo, memphis, gti?
Or 2 or 3 10", to work in this "small" volume and spread the bass a little more.
 

·
Registered
Joined
·
8,524 Posts
Btw here's the sub box now:


made of 2 of these: Atrend 12SKS Slim Series Sealed Subwoofer Enclosure
Had just one before, now it's maybe 2.4/2.5 cu ft

I will measure and tune a bit more, but I'm not quite satisfied.
The box is braced a bit but is still too soft. But not much I can do externally to reinforce it now, not much space left with the car frame, the seats etc.
It's sitting on an old mastress so vibrations are filtered (it's strapped to the back).

So at the end it's only slightly better, impact and integration are great, really awesome, but it feels a bit slow sometime, or undefined on the bottom end.
Maybe I need to rise the HP, check for distortion?

It's crossed at 120hz, and being here at maybe 1ft of my head it can also be disturbing. Stage is sometime pulled back but it's a trade off.
But on big impacts it can be painfull to my right ear, too much pressure.
I wonder... if it was on the left, right behind me and not in the center, could it be better? or reversed, firing to the back panel, I have only 1" clearance there for the cone due to some chassis braces.

I may try my old ib12au here, but not sure the box is big enough for it. On winisd there's not much difference compared to the gb12.
Or maybe another sub? I searched a bit but don't know much about SQ subs.
Maybe IDQ, treo, memphis, gti?
Or 2 or 3 10", to work in this "small" volume and spread the bass a little more.
I'd maybe try moving it away from you to the passenger side as well. Maybe removing enough carpet from the left side to cut out a new hole. Then sealing up the center one temporarily from the outside with a scrap piece of mdf. And if possible just to see, I would throw in a L shaped round port that fires out the drivers side into your seat back. That should lower distortion by given you the low end without introducing the upper harmonics you get if the box is sealed and you EQ the low end in.

That concept also lets you rotate the box clockwise 180 degrees and see how it sounds with the sub to your back, ported and not.

If it doesn't work you can always go back by using the initial MDF cover piece to cover the hole from the inside and then laying/securing the cutout carpeted piece onto it. And similar for the port hole.
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #330
I'd maybe try moving it away from you to the passenger side as well. Maybe removing enough carpet from the left side to cut out a new hole. Then sealing up the center one temporarily from the outside with a scrap piece of mdf. And if possible just to see, I would throw in a L shaped round port that fires out the drivers side into your seat back. That should lower distortion by given you the low end without introducing the upper harmonics you get if the box is sealed and you EQ the low end in.

That concept also lets you rotate the box clockwise 180 degrees and see how it sounds with the sub to your back, ported and not.

If it doesn't work you can always go back by using the initial MDF cover piece to cover the hole from the inside and then laying/securing the cutout carpeted piece onto it. And similar for the port hole.
Great t3sn4f2, thx for the feedback.
You're right I should try that before messing with everything else. And left or right firing would be easy to test.
About ported I'm not sure, well just not comfortable, I didn't have great experience before. But I should keep this in mind.

I know the wavelengths are too long to have real effects on ITD, but I still wonder if it's related.
I mean the box is so closed to the right ear, almost the same distance than my ear to ear, and almost on the same axis.
And clearly I can sometime feel a much bigger pressure on the right ear, or can it be due to harmonics?
I got a Q of 0.58 with filling in the big box, 0.8 empty. Can it have some effect too?

That's why I wanted to try behind driver seat first, to get same distance to each ear.
But I'll try both, and in any case the "corner" load should help a bit.


Yesterday I’ve finally built the small boxes for the 8.75" (SW223BD02_03):



They’re probably not a good replacement in my case, but they could be nice woofers, they play quite high.
And I got a Q of 0.7 in small boxes without filling! On Winisd they look nice, very similar to the GB except for max output and Xmax, and 3 of these would still be smaller!

Well first the new side holecut!
 

·
Registered
Joined
·
8,524 Posts
I know the wavelengths are too long to have real effects on ITD, but I still wonder if it's related.
I mean the box is so closed to the right ear, almost the same distance than my ear to ear, and almost on the same axis.
And clearly I can sometime feel a much bigger pressure on the right ear, or can it be due to harmonics?
I got a Q of 0.58 with filling in the big box, 0.8 empty. Can it have some effect too?

That's why I wanted to try behind driver seat first, to get same distance to each ear.
But I'll try both, and in any case the "corner" load should help a bit.
Hhmmm, hard to even begin to speculate on what it could be at this point. I'm sure after you try the sub experimentation you'll see a big change that points you in the right direction.

Oh.....

Btw, what's optical variable??
It's a digital output that can be attenuated via the master volume control of the head unit. As opposed to a "fixed output". You don't see that hardly anywhere other than on a PC. The head unit oabeieo mentioned is the first one I know of that implemented that feature. Many other units can control the master volume from the head unit when using the digital output, Alpines mostly. But they do so by leaving the digital output untouched and sending control signals to an analog volume control on a compatible DSP. That's a better way of doing it since you don't loose resolution.
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #332 (Edited)
Hhmmm, hard to even begin to speculate on what it could be at this point. I'm sure after you try the sub experimentation you'll see a big change that points you in the right direction.
Yes I need to try, wanted to start today but finally spent some time in the car tuning...
With another method, new xo points allowed by a bit of eq on the C-dsp, but not too agressive. So now it's "acoustic" [email protected] Instead of 120hz.
Then dirac with a smaller volume for the measurements. A lot of sweeps in total, so it's easier to duplicate like that. Something like a cube of 1ft.

Still sounds pretty good, but different. I lost on impact what I gained on stage.
It seems a bit wider on left, but not sure.
So it's a little bit less "chesty", but the midbasess working harder keep 95% of the stuff in front. Not bad, cleaner maybe, but a tad less fun.
Well I'll see during this week, my ears are done.
Maybe the sweet spot is 100hz.

Oh and I also rised the hp on the sub, dirac capted some big humps below 20hz (but it was noisy today outside).
And I still want to try the bmx higher, I'm sure they would be happy playing up to 1600hz.
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #333
It's a digital output that can be attenuated via the master volume control of the head unit. As opposed to a "fixed output". You don't see that hardly anywhere other than on a PC. The head unit oabeieo mentioned is the first one I know of that implemented that feature. Many other units can control the master volume from the head unit when using the digital output, Alpines mostly. But they do so by leaving the digital output untouched and sending control signals to an analog volume control on a compatible DSP. That's a better way of doing it since you don't loose resolution.
Got it Thx!
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #334 (Edited)
Jus FYI, here’s the last result:

BEFORE (120-800/48db no EQ):




AFTER (80-800/24db) EQ on sub & midbass around XO, none on horns:




- Even if the 8 sweeps*placements used for measurements were not exaclty the same, the horns look the same
- Sub-midbass transition is smoother (it is flat at the nose point used for TA)
- big peaks on midbass are tamed a bit, so the curves are closer to target
- with this EQ, I was able to re-adjust levels, so less boost everywhere.
- bottom sub is not as cleaned. I guess from external noise.

Something I noticed yesterday during the sweeps, while checking my previous setting of last week.
At quite high level (near mic clipping), the woofers were "flopping" around 80Hz.
Like reaching Xmax maybe? The VBRRR sound for 0.5sec, but not bottoming
But I couldn't see anything on distortion plots, and didn't really hear it with real music for one week.

Now with the new setting, no more FLBLBLBL :p
EQ was definitely needed here around XO, at least to optimize the place of each boost from Dirac.

Also I compared both settings today (a bit slow since I have to adjust 2 sets on both units, Cdsp and ddrc)
> new one is cleaner for sure
> the loss of impact is very subtile, but largely compensated by cleaner sound.
> stage is the same or too close to notice any difference

EDIT: another thing, I now tune the C-dsp to flat (trying SUBTERFUSE advice) but... what a pain with the horns, it is so loud!
 

·
Registered
Joined
·
4,185 Posts
Yeah those scenarios were more for me to build the idea. I need to think like that.
Ok I forgot the stop band only, it would work with both, with same effects, and I tried that with good results.

Don't forget jazzi's spreadsheets for the crossover targets, really handy to see the XO curve to reach.
Especially used with auto EQ, really powerful!

I said B2 harder, but only on driver Eq, because you tune the average in this case, and it can be real crazy. Especially with spfr method.
But yeah I never really used input eq before (always had the apl too for this).
Or barely to finetune L/R, and if I went out of peqs per driver. But I see that in your method it should not be used for one driver.

But bigger point I didn't realized: you use RTA. I use sweeps only.
Rta kind of includes some effect from the room already.
That will push the tuning into the grey zone, mix of both methods.

Grey zone that I was talking about, that each method use some bits of the other one too in a way.
don't know why but I didn't like the results with rta. Like if too many dips and peaks were masked.

Let's move this conversation over here . I didn't want hijack
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #337
Definitely want to know what works best ....

Are you using Dirac as a measurement method now strictly instead of rew ?

Not that it matters , just curious, does Dirac have RTA functions?
That's the,point.
I didn't know how to start so I tried 8 sweeps on rew for midbass (and sub too, with some changes!), using the same placeemtn than for dirac. Then eq lightly.
For horns I didn't do anything, I know now the levels they want compared to others.
Horns are kind of hard to eq, at least to measure, between 8nsweeps, spfr etc it's wild.
But they sound ok to me like that, just managed by dirac.
So yeah this conversation is all about this!

So a lot fo variables:
- quality of the baseline for reference
- mic placement = size of the "cube" for dirac
- eq or not before
- method used for eq (other thread)
- target shape and levels
- gains things (but I don't care)
 

·
Registered
Joined
·
4,185 Posts
Have you just tried using plain old RTA with PN pink first on everything?

I find it a massive pain in the butt to use sweeps to try to tune,
I always use sweeps to analyze what's going on acoustically but not actually for setting my baseline tune. Seems like the RTA is not that obsolete after all especially set to like 32 averages or even 16 averages are usually use 16 because it responds faster, once I have a pretty flat baseline then I start doing sweeps but for the most part to try and EQS system just off sweeps man that would take a long time and even with a whole bunch of averages I bet it plain old RTA would just do a better Sounding job for setting a baseline response
 

·
Registered
Joined
·
1,999 Posts
Discussion Starter #339
Yeah I tried, but no great results.
First I hate the sound, can't stand it more than 10 sec.
But also I don't trust it enough, ok it averages to big number pretty fast but I don't beleive it catches all informations that fast. Also it catches all stuff around, and I tune outside, not on the street but it can get noisy there.
But the main reason is that it masks the deepest dips and peaks.
Not that I absolutely wan them to fix them, but I want to decide myself.
Well I'll try again

But yeah sweeps are a pain too, it's so long... :laugh:
 

·
Registered
Joined
·
4,185 Posts
Yeah I tried, but no great results.
First I hate the sound, can't stand it more than 10 sec.
But also I don't trust it enough, ok it averages to big number pretty fast but I don't beleive it catches all informations that fast. Also it catches all stuff around, and I tune outside, not on the street but it can get noisy there.
But the main reason is that it masks the deepest dips and peaks.
Not that I absolutely wan them to fix them, but I want to decide myself.
Well I'll try again

But yeah sweeps are a pain too, it's so long... :laugh:
Interesting....

Are you using plain old RTA with bins ?
Or is it spectrograph RTA or something?
 
321 - 340 of 837 Posts
Top