2016 Porsche Macan S - stealth SQ build

[Ge0]

No, this is not my final install. Just updating my test mule to see what sub I like:














I have a bit more power on tap now than I had with the stock Bose subwoofer amp. I did some preliminary gain setting and experimentation last night before shutting things down for the day. All I can say is DAMN! Now that's what real bass sounds like in this car. I did take some REW plots. However, as luck would have in my PC Blue Screened before I could save them off. Some driver quirk when disconnecting my USB audio interface...


Hopefully more on this topic this evening.


Ge0

[geolemon]

EDIT: sorry, I wrote this up at like 10am and didn't hit submit until nearly 4pm... forgot about it, while working... I see you added some more diagrams and posts. Still valid, I think:

Now that we know sub positions lets study frequency response:
Attachment 11394

Part of me thinks "Eh - factory systems are usually just lip service." Not surprised. Also not surprised that lots of people on a Porsche Macan forum might not understand acoustics and acoustic loading, and might absolutely convince themselves "It must be better firing up!" with no measurements, or even minutely perceivable sound difference. Anyway... not my real point.

So - part of me wonders if Porsche might just have actually done some measuring, exactly like what you are doing...
I say that because there is a 35hz-55hz bump in the response that might just be the usual marketing-requested "Ooh, listen to that BASS!" built-in bass-boost?
And, possibly even - that 100hz. But there's also other possibilities, because that dip in the middle is pretty severe.
But just to be clear - my hypothesis #1 is the possibility "there's some built-in factory EQ/DSP built into this BOSE subwoofer amp". Does anyone know if that's true?

The other thing I'd be wondering is more about that 55hz-70hz dip... because if you were to draw a line straight across that (say - like you'd be inclined to slide some EQ northward in this region), then suddenly we end up with a pretty normal looking sealed-box curve for a little sub like that. So right off the bat, I'm thinking "Hypothesis #1" is probably wrong.

So - Hypothesis 2 would be cancellations. I feel like some of it is probably the natural acoustical NULL that most car interiors get. I know the Macan isn't huge, but it's probably bigger enough inside that my (I'm referencing ErinH's measurements) Civic hatch 70hz NULL could pretty easily be 5hz lower in a Macan. Your roof is a little taller, your back glass is more upright, and it's probably even at least a touch longer inside, too.
Now, if it was JUST that NULL, I'd expect a narrower null dip - even narrower than the dip in the red line at 100hz there - so I suspect there's more going on:

As you mentioned, you also have those midbass drivers up front... which I propose might not even have a high-pass filter on them, or it might be letting them play down to 40 or 50 hz at least. If that were productive, it could cause that slight hump between 40hz-50hz (I say "hump" because if you follow the blue line up, and imagine a standard sealed box frequency response in an anechoic chamber like modeling software simulates, you'd see it rise from 30hz to 40hz looking exactly like that, but then continuing to flatten out - at 45hz I'm picturing the anechoic response to be just under the red line, then being pretty flat to 100hz). So that boost could be EQ - or the boost could be the combined output of the subwoofer and front bass drivers, aligning like they should... but then working against each other starting at about 50hz there, and progressively cancelling more and more... their cancellation unfortunately appearing to JUST start to end at about 60hz, when it rams right into that 65hz NULL, and then they combine productively again by that 80hz point.
Again that's just a hypothesis, but I feel better about hypothesis 2 than hypothesis 1.

You could probably tell which by taking one more set of measurements (blue line - as Porsche designed it) smack between the headrests, and then a third measurement line right at the middle of the back seats.
If you get less cancellation between the headrests, and less cancellation still on the back seat still, then I'd bet that's what it is.

Or - disconnect those front door speakers again and take another front measurement. All should provide clues.

[Ge0]

EDIT: sorry, I wrote this up at like 10am and didn't hit submit until nearly 4pm... forgot about it, while working... I see you added some more diagrams and posts. Still valid, I think:


Part of me thinks "Eh - factory systems are usually just lip service." Not surprised. Also not surprised that lots of people on a Porsche Macan forum might not understand acoustics and acoustic loading, and might absolutely convince themselves "It must be better firing up!" with no measurements, or even minutely perceivable sound difference. Anyway... not my real point.

So - part of me wonders if Porsche might just have actually done some measuring, exactly like what you are doing...
I say that because there is a 35hz-55hz bump in the response that might just be the usual marketing-requested "Ooh, listen to that BASS!" built-in bass-boost?
And, possibly even - that 100hz. But there's also other possibilities, because that dip in the middle is pretty severe.
But just to be clear - my hypothesis #1 is the possibility "there's some built-in factory EQ/DSP built into this BOSE subwoofer amp". Does anyone know if that's true?

The other thing I'd be wondering is more about that 55hz-70hz dip... because if you were to draw a line straight across that (say - like you'd be inclined to slide some EQ northward in this region), then suddenly we end up with a pretty normal looking sealed-box curve for a little sub like that. So right off the bat, I'm thinking "Hypothesis #1" is probably wrong.

So - Hypothesis 2 would be cancellations. I feel like some of it is probably the natural acoustical NULL that most car interiors get. I know the Macan isn't huge, but it's probably bigger enough inside that my (I'm referencing ErinH's measurements) Civic hatch 70hz NULL could pretty easily be 5hz lower in a Macan. Your roof is a little taller, your back glass is more upright, and it's probably even at least a touch longer inside, too.
Now, if it was JUST that NULL, I'd expect a narrower null dip - even narrower than the dip in the red line at 100hz there - so I suspect there's more going on:

As you mentioned, you also have those midbass drivers up front... which I propose might not even have a high-pass filter on them, or it might be letting them play down to 40 or 50 hz at least. If that were productive, it could cause that slight hump between 40hz-50hz (I say "hump" because if you follow the blue line up, and imagine a standard sealed box frequency response in an anechoic chamber like modeling software simulates, you'd see it rise from 30hz to 40hz looking exactly like that, but then continuing to flatten out - at 45hz I'm picturing the anechoic response to be just under the red line, then being pretty flat to 100hz). So that boost could be EQ - or the boost could be the combined output of the subwoofer and front bass drivers, aligning like they should... but then working against each other starting at about 50hz there, and progressively cancelling more and more... their cancellation unfortunately appearing to JUST start to end at about 60hz, when it rams right into that 65hz NULL, and then they combine productively again by that 80hz point.
Again that's just a hypothesis, but I feel better about hypothesis 2 than hypothesis 1.

You could probably tell which by taking one more set of measurements (blue line - as Porsche designed it) smack between the headrests, and then a third measurement line right at the middle of the back seats.
If you get less cancellation between the headrests, and less cancellation still on the back seat still, then I'd bet that's what it is.

Or - disconnect those front door speakers again and take another front measurement. All should provide clues.

I have a few amp guts postings of the factory amps. The main amp has a fairly powerful DSP. The sub amp does not. However, the sub amp gets its signal from the main amp. So, I'm sure the signal going to the sub is heavily processed.

Like I eluded to in my post, I think there are some significant boosts and cancellations going on between the front midbass speakers and the sub. One of my first graphs was with the sub disconnected. The other with it connected.



Next, I find it amazing the boost in response amplitude you get by embedding a subwoofer under the floor vs. letting it fire into an open cabin:



Who would have thunk it?

Ge0

[geolemon]

You surely could be right. Like I said - that was my initial suspicion as well. But dusting off my old pro installer / manager hat - that's one of those things that needs to shift from "maybe" to "yes/no for sure"...

I don't think you want to move forward in a direction without knowing that, because so much (and such different) potential mitigation hinges on it.

I know you have an O-scope - so if I were in your shoes, before doing anything else, I'd run a sweep right from your head unit, and I'd watch the amplitude at the subwoofer terminals. If you see amplitude fluctuating like there's an EQ curve, then I'd repeat the test at the amplifier inputs.
Then you'll know, for certain:
a) is there factory EQ mucking things up? Or is it cancellations and natural nulls?
b) If there's factory EQ curves, are they in the head unit, or in the amps themselves?

Note (b) helps you know whether you can trust your signal, when you swap in the Zapco for more testing.

I think if it was truly EQ, then I think the dip you see there wouldn't be such a dip...
I *do* think they may have said "eh, screw it" and NOT included any compensatory EQ ...but if they put the effort to include the EQ, you'd have more of a ripple than a canyon there.

Only one way to know - sweep it.

Sent from my LM-G710 using Tapatalk

[Ge0]

You surely could be right. Like I said - that was my initial suspicion as well. But dusting off my old pro installer / manager hat - that's one of those things that needs to shift from "maybe" to "yes/no for sure"...

I don't think you want to move forward in a direction without knowing that, because so much (and such different) potential mitigation hinges on it.

I know you have an O-scope - so if I were in your shoes, before doing anything else, I'd run a sweep right from your head unit, and I'd watch the amplitude at the subwoofer terminals. If you see amplitude fluctuating like there's an EQ curve, then I'd repeat the test at the amplifier inputs.
Then you'll know, for certain:
a) is there factory EQ mucking things up? Or is it cancellations and natural nulls?
b) If there's factory EQ curves, are they in the head unit, or in the amps themselves?

Note (b) helps you know whether you can trust your signal, when you swap in the Zapco for more testing.

I think if it was truly EQ, then I think the dip you see there wouldn't be such a dip...
I *do* think they may have said "eh, screw it" and NOT included any compensatory EQ ...but if they put the effort to include the EQ, you'd have more of a ripple than a canyon there.

Only one way to know - sweep it.

Sent from my LM-G710 using Tapatalk

I have a factory system that is all digital fiber optic until you reach the speaker outputs. So, its hard to tap into the amplifier inputs. I do have a line level out going to the sub amp though. I know I need to scope this. I can use REW to do a slow frequency sweep and just watch the amplitude on the sub amplifier input. I'll let you know what I find.

Unfortunately my Picoscope is sitting on my desk at work. The building is in government mandated lock down. I do have an older Techtronix hendheld though. I can't take screen captures but probably do not need them here.

Ge0

[Ge0]

I did some sub testing over the last few days. As you can see I have the JL Audio 10TW3-D4 mounted to a 1.25 cu/ft sealed enclosure pointing upwards in my rear hatch as it would be if i mounted it under the floor.




Here is the response graph with and without the pathetic factory sub:


The factory sub does nothing more than add a huge boost at 80Hz over the front door midbass??? Strange.

Anyway, if we add a serious amp and a decent sub we can make improvements. Here is the JL Audio 10TW3-D4 powered by a Zapco DC Reference 500.1 in red. The factory system with no sub in green:



As you can see we gain some amplitude in the bass range AND extend low frequency response by a considerable margin. The listening experience was good with a noticeable amount of low frequency extension with songs that required it. Bass was much less Boomy and more controlled.

Now let's keep all amp settings the same and switch subs. The Sundown Audio SD4-d4 is also on my list. This is a strange little sucker that my friends encouraged me to try. Sundown SD4-D4 in Blue:


I was impressed with the measurements. My initial subjective thoughts on this sub were "man I need to turn the bass down". This sub is efficient.

But, before I could get a chance to reduce output gain by -3dB for another evaluation my amp over heated and failed. CRAP!!! Technical difficulties preventing me from making a decision...

More on this later while I get my hardware issues sorted out. Maybe an amp repair. Maybe a new amp.

In the mean time I am impressed with the Sundown Audio SD4 given it is nearly $100 cheaper than the JL Audio 10TW3-D4

Ge0

[Whiterabbit]

Id pick the sundown.

[Ge0]

Fun with circles:

I have an idea on how to mount the door based midranges.

Attachment 11479



'

Make a pattern:











Lesson learned. Don't cut ABS plastic with a dull Jig Saw blade at 1:30am and many beers in you.



Correction time:




To be continued...

Ge0


Attachment 11478

[Ge0]

Id pick the sundown.

My initial thoughts as well. More than enough bass for me and easier to mount given my space restrictions.

Ge0

[Whiterabbit]

More bass no matter the curve means easier to EQ down. I’ll pick output every time.