milosz

Does the REF63 Midrange drop right in to the mounting space in the G37 door? I've wondered if there was an alternative that would fit right in the door. There's not much room in there.

I see the REF63 is actually called a "full range driver" and they specify response to 20,000 Hz though no curve or dB tolerance is given. What is the idea of using the GS10 tweeters with this driver that also produces treble? Are you using a low-pass filter on the REF63 Midrange to prevent it being fed treble energy? Having two drivers placed relatively close together covering the same high frequency range is going to produce interference between the two drivers, which will result in irregularities in the response that can't be dealt with in DSP. Have you run curves on your system? I'd be interested in seeing them.

Rollo

Very interesting information and discussion here. I find the bose system in my 13 coupe to sound great after adding a SKAR 10 inch sub in 4080 box with a RF sub amp.

Entcee

iTrader: (3)

Curious to know how you wired it (where you got the signal from)

milosz

Adding a good sub makes a huge difference to the Bose system. The Bose system has a fairly clean, fairly detailed sound but it lacks bass quality, and it lacks lowest-octave performance in general. A good sub will add all of this- tight, textured bass and bass depth.

Any further improvements to the Bose system are more complicated to achieve. Adding a pair of MiniDSP balanced 2 x 4's between the head unit and the Bose amplifier, using them as signal processors and NOT as crossovers, will allow tailoring the frequency response and delay timing for about $250. https://www.minidsp.com/products/min...p-balanced-2x4

This enables tailoring the amplitude response either automatically using a measurement mic and a program like OmniMic to calculate biquads, or manually using either "graphic EQ" or parametic EQ controls in the MiniDSP software.

There are some sound reproduction flaws - like ringing, energy storage problems, group delay issues, etc- that can't be solved by EQ and can only be addressed by using superior drivers and, to some degree, driver enclosures and damping. Trying to reach this kind of sound quality in a G37 can get expensive and complicated very rapidly- I've seen photos of someone who put a Focal Utopia BE 3-way setup in the front, making fiberglass "pods" to mount the mid and tweeter in the A-pillar- and of course they used aftermarket amps, custom subwoofer and DSP control. I don't know if they also replaced the center speaker driver, I suppose you'd have to. They did a very good job with this build, but to me it didn't look all that great to have these pods "growing" out of the A pillar, not as clean as the stock G37 interior. I don't know if it would be worth doing this, it would cost upwards to $10,000 to have this done properly, according to some high-end car audio installers I know. And I don't know how it would sound, really, though the Focal Utopia BE Active system enjoys a very good reputation and their home speakers based on the same technology -their proprietary "W" cone material, their beryllium tweeter dome- are among the best audiophile speakers available. So if the placement of the drivers in this build was carefully considered, this type of setup might sound really great- but at a huge price.

You'd almost be better off selling the G37 and buying an Acura TLX with the Krell sound system, one of the best sounding - if not THE best sounding- car audio packages. The TLX is a very nice car, and with the V6 it's pretty fast (not as fast as a G37, but still pretty fast) so if you want this level of car audio coherence and detail, it might be the way to go.

djdiskino

Really interesting observations! Glad to see someone really dissected the audio components of the Infiniti.

I'm over here still struggling to find a remote wire for my JL Audio Amp.. right now it's wired to an ignition on fuse. I tried the suggested one that feeds the Bose amp in the back but it doesn't have enough juice to signal my amp on. Any suggestions?

Landshark

search for sub install threads and the wiring layout for your car (coupe/sedan) ...

you tap off the wires on the connector to the Bose amp - use some Posi-taps. the wires you want are the preamp signal wires from the head unit (before they get EQ'd by the Bose amp).

i used the rear signals, and have since realized the rear speakers are just "fill" and don't get much power - i have to have my sub level at around 75% for my liking. sometime soon, i'll switch my sub inputs to the front speaker amp signal wires.

Entcee

iTrader: (3)

I wasn't asking because I don't know how to do it.... I have tried front, rear and even before the sub amp. The reason I asked is because he said it sounded great. I have not had that experience no matter where I source the signal.

The best sound to me (still not great) is the signal between the Bose amp and the sub amp (same as OP). This is how my subs are currently connected. I am also using a RDL-TX-1A.

Here is the beginning of my struggles:
https://www.myg37.com/forums/audio-v...l-options.html

zer099

iTrader: (2)

Thank you very much for this research and footwork in seeing what actual numbers this system produces.

What's your opinion in switching out the midrange in the door for a scanspeak 10F/4424G. Someone else did this a bit back and reported back better balance. They had to do some trimming of the panel to get it to fit was the only downside.

I was thinking of going with the 10F/4424Gs as my mids up front, follow your lead with the polks in the rear as I sat back there once and noticed all i heard was bass (I have a Acoustic Elegance IB10AU in my rear deck powered by a bridged Blaupunkt VA2100).

milosz

Scanspeak makes some very good drivers. The 10F/4424G is a 4 inch driver, my first concern would be 'Does it fit- or does the speaker mount in the door need to be modified to make it fit?' Since someone has used this driver, apparently it does fit or modifications allowing it to fit are possible.

My second concern would be the breakup mode this driver exhibits at 9 kHz. (see the graph, below, supplied by Scanspeak) - The 10F/4424G is a fiberglass cone driver, and fiberglass cones, like metal cones and some poly cones, exhibit a peaky "oilcan resonance" - as you approach this frequency, the cone will start to resonate and store energy, and harmonic distortion will increase significantly as well, not to mention the driver will exhibit an unnatural emphasis in this frequency range.

Now, 'best practice' in speaker design would be to place a low pass filter at least one octave below the resonance frequency. In an active system, that would mean fixing the upper crossover point at 4.5 kHz with the tweeter operating above this point, but not the midrange driver.

The G37 wiring does not run a separate line from the sail-mounted dome tweeters back to the amp module in the trunk, however. There's apparently a capacitor* in series with the sail-mounted tweeter, but the tweeter is wired directly in parallel with the stock Bose midrange. So, unless you ran wires for the tweeter from the doors back to the amplifier, using an electronic crossover to limit the 10F/4424G to operation below 4.5 kHz would leave you with NO high treble.... the other alternative is making a passive 4.5 kHz low pass for the 10F/4424G and wiring it in the door, this would keep the 10F/4424G out of it's problem frequency range, and still allow the high treble to get to the sail mounted dome. This could - if the crossover were designed well- work quite nicely. Passive crossover design isn't the easiest thing to master- using simple cookbook math almost never gives good results, generally you need to iterate a design / build / measure / re-design / build /measure process to get it dialed in. But the 10F/4424G - if it fits! - does look like a good candidate here. ( I am skeptical about getting it to fit, I've looked in there and the space seems quite tight, especially in terms of depth.)

The other issue with the 10F/4424G occurs at the lower end of it's range. The FR starts to roll off below 300 Hz; Scanspeak says to put this driver in a sealed 1 liter enclosure for an F3 of 275 Hz or a vented 1.2 liter box to get down to 125 Hz. Since there's no way to put a 1 liter enclosure inside the door, I imagine the F3 will be high-ish for this driver in the door, the back of the driver will face into a cavity in the door of some unknown volume and would certainly not be sealed air-tight, so it's not possible to calculate the F3 limit for the low frequency performance of this driver- you'd have to measure it in place and from that you would decide the high-pass crossover point to use in your electronic crossover in the trunk between the midrange and door woofer.

FYI I have ordered one of the stock Bose 3.5" door midrange drivers from eBay, and when it comes I will be running some curves etc for that driver which I will post. We'll see how bad / good it is... of course frequency response and distortion curves don't tell the whole story about sound quality, but I think it will be interesting to see what's going on with these 3.5" drivers. I suspect there will be some kind of peak at 6 kHz or above, because these drivers sound a bit "ringy" to me in that range. In the G37 Bose system there is no low-pass filter on these driver, they are used nearly full range. Most of the treble heard in the cabin comes from these drivers, the stock sail-mount tweeters beam a little treble energy into the cabin, but this mostly fills in a little of the high treble; the highs that have the most impact on the sound of the Bose system comes from the 3.5" Bose drivers left / center / right. I have wondered if using a passive low-pass filter on these drivers to limit their treble output somewhat would be beneficial....


Here's the 10F/4424G FR curve






Scanspeak 10F/4424G 4" midrange driver impedance and frequency response curves. Black is on-axis, green and red are progressively more off-axis.

* I assume that Bose has a capacitor in series with the sail-mounted tweeter, forming a first-order crossover. I can't imagine they don't. But, I have never seen it....

Landshark

i just have a powered Kicker "truck box" .. sealed enclosure because i want the bass tight and responsive, not "boomy" and flubby. tapped the rear preamp outs, and it sounds fine. the Bose highs arent that great IMHO, but sounds good through Viper4Android app when tweaked.

its not going to shake the neighborhood, but thats not what i was going for. like i said, i have the sub level around 75%, so at some point i'll switch to the front preamp outs, but it sounds good to me - big improvement.

Entcee

iTrader: (3)

I have a sealed box too, with 2 10s. Gain is almost all the way up.

I get noise, like alternator whine, but does not change pitch with RPM. This noise comes through whichever speakers I tap. That's one reason I connected to the factory sub signal, so I can unplug the factory sub... No more noise.

No matter which source I used, the bass sounds (and even feels) delayed slightly compared to the rest of the music.

I did not like the rear speakers tapped... You would be surprised at how much bass there is in the phone ringing!

future62

Thanks a million for this. I just got a G37S sedan with the upgraded system.... I was impressed with it at first, but there are two things that bother me:

1, the bass is good, up to a point. I've heard this in the past with Nissan/Bose systems (had it in my 350Z, with an aftermarket radio feeding the Bose system), and only recently have they fixed it (I had a 2016 Maxima rental a few months ago). Basically what happens is the volume goes up at different rates across the frequency band. So above 100Hz or so, you can crank it until your ears bleed. But below that the volume is capped pretty early. I listen to a good bit of rap and electronic music so it sucks.

2, it's interesting that the graph is that flat... maybe my hearing is toast, but I find the tone of the system to be a little... bright? To the point that it's harsh. I'd love to be able to dial things back in the 3k-6k range. I do have it in driver center stage which may make the difference... I'll switch it off and see if that helps.

Problem 2 is expensive to fix as you said... but for problem 1, would you recommend throwing a sub in the rear deck location? The challenge there would be getting a clean + constant volume signal below 100Hz... I don't even know if the non-sub channels go that low. Have you noticed the gain shift with your sub setup?

milosz

My suggestion for #1 is to disconnect the connector to the rear deck subwoofer amp, and tap into it's audio signal and add a decent 12" sealed sub and adequate amp in the trunk. Judicious adjustment will give you tight bass . Here's what I did https://www.myg37.com/forums/audio-v...ml#post4091839

RE: #2 yes it measures flat but the highs do sound..... wrong.... too hot... I think this is due to RINGING somewhere in the treble by those 3.5" "midrange" drivers. These are located: driver door, dash center, passenger door. Picture of the driver shown below. I believe there is an electronic high-pass filter and Bose amplifier driving these; I do not know is any Bose EQ is applied. They appear to be fed all frequencies above their high-pass crossover point. The dome tweeters in the sails (photo below) have a capacitor in series with them, they are driven by the same amplifier that drives the "midrange" driver. Midrange driver does NOT have a low-pass to filter out the high treble. There is not separate speaker wiring from the Bose amplifier to the tweeter- the tweeter just works off the feed to the midrange. So, both the tweeter and the midrange are reproducing the highs... if you cover the tweeter with your hand while the system is playing, you can hear that the highest treble is cut off but that the ~4kHz to ~ 11 kHz treble is still audible, because it is coming from the 3.5" drivers.

Typically, in efforts to extend the treble of midrange drivers, manufacturers will resort to stiffer, lighter cones.... this usually results in an "oilcan" resonance peak at the top end of their useable range. With driver cones of aluminum, fiberglas or certain plastics this peak can be quite large. With paper, the peak can be controlled to a degree, but nonetheless the resonance is still there. As the signal driving the speaker approaches this resonance peak the cone will begin to misbehave in several ways. You'll get more harmonic distortion, and often the cone will "store energy" at this frequency so that even after the signal is gone, the cone will continue to vibrate for a while- it will "ring."

Both the distortion and the ringing can add to a "bright" sound that may not show up on an frequency vs amplitude plot. This is partly because frequency response plots are made by sweeping an audio signal from 20 Hz to 20,000 Hz and the drive signal will only remain at the cone's resonant frequency for a very brief time as it sweeps through. This doesn't given enough time to fully excite the resonance, a cone at resonance acts as an integrator, storing energy for a period of time, and in the case of a swept frequency graph, the drive signal is not feeding energy into the cone at it's resonance frequency for very long at all.

The energy the cone stores at it's "problem" frequency will "ring back out" of the cone, either at the resonance frequency, or at the resonance frequency and harmonic multiples. I suspect something like this is going on with these Bose "midrange" drivers- they are ringing and this is adding extra treble energy to music playback. This won't show up on a FR graph, but WILL show up under various types of time-domain analysis such as a waterfall plot.

I have gotten one of these "midrange" drivers (and also one of the soft plastic dome tweeters) and I will be setting these up in my lab and testing them in isolation to learn more about their electro-mechanical behavior, just out of curiosity. I'll post that stuff when I get around to doing it.

AT ANY RATE, #2 problem - hot sounding treble- how to fix? There are several things to try. One is to EQ out a little of the treble at the frequencies at which the driver rings. This is not really a solution, but it can help a little. Also, in addition to cutting the EQ of the treble at this particular point, boosting the frequencies below this point a little can help too, this can mask the problem. HOWEVER, EQ can only help a little, and really drastic levels of EQ will make the sound especially bad, so be judicious with EQ. I am going to try this approach, I have bought two MiniDSP 2x4 balanced which I will use as EQ in the LEFT / RIGHT and CENTER signals feeding the Bose amp. I'll write this up when I get around to installing those.

Depending on what frequency this ringing is happening at, it may be productive to put a passive low-pass crossover filter in series with the driver, filtering out the treble from the midrange cone but allowing it to be fed to that soft done tweeter. That **MAY** be a good fix. (However the dome tweeter may have it's own issues! I have to test one...)

But anyway adding a passive 2nd order lowpass in series with the midrange driver might be an inexpensive and fairly easy-to-do fix (if it does the trick!) - there's certainly room in the door for a few capacitors or inductors, and since the frequency is pretty high, they won't be physically very large, and should fit easily.

It might even be a good idea to find a better tweeter to replace the one in the sail, and then, with a passive low-pass crossover filter keeping the higher treble out of the midrange driver, the better tweeter might give the cleaner, more balanced sound we want. I haven't looked, but I will bet there are a few decent tweeters that can fit in that tiny space. The soft plastic dome of the existing Bose tweeter, well, I have a personal prejudice against this type of tweeter dome material, I've always associated it with "hard" treble sound. But frankly, given the position, I'm not sure that the tweeter really contributes that much to the overall sound.

The REAL fix would be to use a different, better mid and tweeter altogether. This is hard, because the space is so limited. Not many drivers will fit. And really I don't know what frequency Bose is crossing over to the midrange driver at. Finding a driver that will match the efficiency of the existing mid and so stay in balance with the door-mount 10" bass drivers will be tricky. Of course you could always pad down the 10 inch drivers but then your total system volume level might not be high enough for your tastes.

Adding a DSP crossover / DSP EQ and separate amplifiers for new midrange drivers and tweeters.... now we're adding a LOT of complexity and cost, in addition to the difficulty of finding drivers that will fit. You could build new pods for new drivers etc etc now we're moving into an area which will change the look of the cabin, probably not for the better....

So my approach is to first try everything I can short of replacing those midrange drivers... if they still remain sonic problem-sources, then I'll have to try other drivers I guess, but I really don't want to go down that road.

It's going to take me at least a month to do my measurements on the Bose drivers in my workshop, and to install and tweak the MiniDSP 2x4's. Stay tuned....










NOTES:

The audio signals in the Bose system wiring are BALANCED PAIRS, and keeping everything balanced can really help eliminate noise and hum. See What?s the Difference Between Balanced and Unbalanced? : Aviom Blog


MiniDSP 2x4 balanced is not expensive ($125 see https://www.minidsp.com/products/min...p-balanced-2x4) and can be used as a stereo biamp crossover or as a stereo EQ, it's adjusted by connecting a notebook PC to it by USB and adjusting the parameters either by loading corrective "biquads" derived from software analysis of measured response, or manually by adjusting the unit as a ten band parametric equalizer. Lots of flexibility, good audio quality and not very expensive. Not really plug-and-play however.

There are other issues besides the "hot treble" - one which I've identified is a little too much "mid-bass" which makes radio announcers sound quite "chesty"- it's an emphasis (and some ringing) around 200 Hz I think and maybe also a slight lack of 500~700 Hz energy. I think this one CAN be tamed by a little judicious EQ. This is coming from the door-mounted woofers.

future62

Wow, thanks for that in depth info.

I would not be too scared to replace the mids in the dash, but I'd have to know 2 things... 1, what the impedance is (which knowing Bose would be 2 ohms) and 2 if those drivers are getting a flat signal or a signal tweaked around the speakers. Either would be pretty much a deal breaker. Is there any way you could test the frequency response of the signal to each channel rather than the final output? Or do you know which speakers get "flat" signals and which ones get tweaked ones?

For the sub I do like Dayton... I have some of their HT sub boxes in my house. But I'd probably go with that shallow Kicker Comp RT or try something else that uses passive radiators. I guess since the rear seat doesn't fold down it doesn't matter but I have done the 50lb box thing and I'd rather not again. You don't think an IB friendly 10 in the rear deck would work?

Also do you think anything can be done with the 10s in the doors? I was thinking about replacing them with some shallow 8s, like the Earthquake Sound SWSs.

In any case definitely looking forward to whatever you find out.... this is great info!

milosz

All the drivers in the Bose system are 2 ohm, and high efficiency too with light cones and pretty strong neodymium magnets.

I don't know if replacing the Bose 10-inch door woofers with other drivers will yield an improvement in sound. I just don't know. Those drivers handle a lot of the lower mids / mids, and so maybe you'd have an improvement there in clarity and definition with some other driver. I think those 10 inch drivers have a pretty low Fs though, and in a door mount- which is not really an enclosure at all- I think your lowest bass performance is going to be closely linked to Fs, and I'll bet you those 10 inch drivers have a lower Fs than any 8 inch driver you can put in there. This is just a guess, though, I don't know what the Fs (free-air resonance frequency) of the Bose drivers is.

The Earthquake Sound SWSs 8X has an Fs (they call it Fo) of 30 Hz. The Earthquake Sound SWSs have greater excursion, but also have fairly low efficiency at 86 dB - I'll bet you the Bose woofers have closer to 90 dB efficiency, maye even higher. Also since the Earthquake Sound SWSs seem to be designed to be used as under-seat dedicated subs and so I wonder how good their definition and detail will be if used up into the lower midrange the way the Bose 10-inch drivers are. I'd like to at lease see a FR graph of the Earthquake Sound SWSs.

It might be tricky to get those Earthquake Sound SWSs mounted in the G37 doors, the Earthquake Sound SWSs have a mounting depth of 2 1/16" I wonder if that's more than the Bose 10 inch pancake. The Earthquake Sound SWSs ARE 2-ohm, however, so that part is good. I wish I could hear them in a G37....

What I've heard regarding replacing the 10-inch shallow profile Bose driver (subwoofer, so-called) in the rear shelf is that it's hard to get a driver in there... there's not much clearance between the deck and window, dropping a driver down in there is hard / impossible depending on the height of the driver. And using a low profile driver in there, I dunno, would it be worth it? Wouldn't it sound quite similar (if marginally better-) than the 10 inch low profile Bose driver in there?

For bass, I always prefer the sound of a sealed box, there's better transient performance and lower group delay down low. But that's just me, I went with what I knew would give me the sound I want.

I will get some frequency response plots of the left / right / center outputs of the Bose head unit when I break into those lines to install my MiniDSP's. I also think it would be useful to try and measure the signal going from the Bose trunk-mounted amplifier to the door-mounted mid, and also measure what the FR looks like of the signal going from the Bose amp to the door-mounted 10-inch driver. Having both the head-unit line level FR and the output FR of the amp driving the speakers will tell you if there's EQ and where it is being done. I suspect the EQ comes from the amplifier module, Bose usually pairs an amplifier with speakers. And I think the head unit is actually mfg. by Clarion, not Bose.

The Bose amplifiers have integrated circuits inside that are labelled BOSE that look to be the "class D" amplifier control chips- they drive pairs of MOSFETS mounted on smallish heat sinks with decent size series inductors, as you'd expect for output circuits of "class D" amps- and I think the BOSE chips could have EQ baked into them also, frequencies and level selectable by SMD resistors or capacitors, or which there are many inside the amps.