When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
To speculate on what **MIGHT** be done to really upgrade the sound, starting with speakers in the doors - I think there might be 6 or 6.5 inch mid-woofers that could work, but you'd need all new electronics too- you'd want to keep most of the deep bass out of these door woofers and rely on trunk mounted subs; then you'd need new midranges and why not tweeters too in the doors, and DSP crossovers adjusted for driver EQ and steep crossover slopes. And POWER. Lots of power. Probably have to modify the door to fit the new midrange drivers, and you'd want to work on the "enclosure" for the door mounted woofers. It would be a lot of work but it could be done. A lot of work and a lot of money. I have not heard them in an automotive setup, but I really like Focal HiFi speakers, and have designed some 2-way home speakers using their Utopia drivers and they are high quality drivers- for some reason, in my imagination, I think that a set of Focal Utopia drivers, properly mounted and with good electronics and good subwoofers, might give excellent results. There are lots of other good hi-end automotive drivers out there - Dynaudio Esotar, Scanspeak, etc.- if you have the $$ to spend on this and want to go "all out" the sky is the limit.
But I have to say, with some appropriate EQ and a good sub, the Bose system sounds pretty darn good. (HAHAHA though as someone who suffers from audiophila, I always want to try one more tweak: I'd like to hear what the Bose system would sound like with a proper crossover between the Bose 3,5 inch midrange and the Bose tweeter- to see what difference spending $10 on inductors and capacitors would make- and I would try it, too, if I wasn't too lazy to remove the door cards. Too lazy and also afraid I'd get impatient and break some plastic mounting clip and never be able to get my door back together properly.)
Milosz - A few posts up (99) I installed part of the crossover you suggested back in January. I need to get back into the front doors to eliminate some buzzing I have at certain frequencies and add the crossover to the tweeter, but installing a LR 12dB 5kHz low-pass on the mid-range produces exceptional results. The output is diminished, similar to placing your hand over the driver to muffle it, the harshness from the cone breakup is gone and I can actually hear the tweeters now - after listening for several months it's almost a day and night difference and well worth the effort. Vocals, brass instruments, electronic music - all dramatically improved. Additionally, when I was in the doors I added sound dampening, fiberglass insulation and took time to lubricate the window mechanism for good measure. With the sound dampening and fiberglass the woofers also took on a warmer sound which is more pronounced and seems to be more accurate.
I have a set of Morel separates and have considered replacing the factory speakers with them. At 4 Ohm, they could theoretically work with the factory amplifier (I have a Zapco 5 channel if needed) however differences in sensitivity would impact staging. And, although I realize sound under 100 Hz is non-directional, I can't imagine any 6.5" driver reproducing the kick and impact of the Bose 10's.
I recommend taking the time to install the 5kHz crossover - you'll be amazed at the difference!
Milosz - A few posts up (99) I installed part of the crossover you suggested back in January. I need to get back into the front doors to eliminate some buzzing I have at certain frequencies and add the crossover to the tweeter, but installing a LR 12dB 5kHz low-pass on the mid-range produces exceptional results. The output is diminished, similar to placing your hand over the driver to muffle it, the harshness from the cone breakup is gone and I can actually hear the tweeters now - after listening for several months it's almost a day and night difference and well worth the effort. Vocals, brass instruments, electronic music - all dramatically improved. Additionally, when I was in the doors I added sound dampening, fiberglass insulation and took time to lubricate the window mechanism for good measure. With the sound dampening and fiberglass the woofers also took on a warmer sound which is more pronounced and seems to be more accurate.
I have a set of Morel separates and have considered replacing the factory speakers with them. At 4 Ohm, they could theoretically work with the factory amplifier (I have a Zapco 5 channel if needed) however differences in sensitivity would impact staging. And, although I realize sound under 100 Hz is non-directional, I can't imagine any 6.5" driver reproducing the kick and impact of the Bose 10's.
I recommend taking the time to install the 5kHz crossover - you'll be amazed at the difference!
-Eric
I'm really sorry that you said that. Now I have something else to be obsessed about. HAHAHAHA Thanks for the report on effectiveness of this modification. I WILL have to do this at some point. That Bose 3.5" driver really measured quite well up to about 5 kHz, and even above 5 kHz it wasn't ragged or peaky- it just had stored energy issues, so it makes sense that a low pass would improve the sound.
Actually, I'd like to Dynamat the whole cabin- firewall, doors and floor. Maybe not the roof. Make the whole car quieter. Don't think I'll ever do that, though.
I agree that it would be hard to find a driver with equal "slam" to replace those low-profile Bose 10-inchers in the front doors. Once you supplement their lowest octave with a taut subwoofer, they really do sound pretty good. There's no substitute for cone area for moving air.
I'm really sorry that you said that. Now I have something else to be obsessed about. HAHAHAHA Thanks for the report on effectiveness of this modification. I WILL have to do this at some point. That Bose 3.5" driver really measured quite well up to about 5 kHz, and even above 5 kHz it wasn't ragged or peaky- it just had stored energy issues, so it makes sense that a low pass would improve the sound.
Actually, I'd like to Dynamat the whole cabin- firewall, doors and floor. Maybe not the roof. Make the whole car quieter. Don't think I'll ever do that, though.
I agree that it would be hard to find a driver with equal "slam" to replace those low-profile Bose 10-inchers in the front doors. Once you supplement their lowest octave with a taut subwoofer, they really do sound pretty good. There's no substitute for cone area for moving air.
Yes - you really must. It took me about 4 hours to get into both doors, remove the regulator so I could access the door skin, install insulation, damping and crossover.
I purchased components from Madisound:
I would assume the Sonic Barrier would need to be wrapped in plastic to prevent mold? It's in a very wet environment and I don't see in the listing that it's resistant to mold/mildew. This is the second vehicle where I've installed compressed fiberglass, actually fiberglass ceiling tile where I remove the pebble backing, cut to fit, wrap in thin plastic, then adhere to the door with aluminum tape. The idea is the same as the Sonic Barrier - some question the effectiveness when wrapped in plastic and of course the concern with breathing in airborne fiberglass - the plastic prevents moisture and particles from becoming airborne. I learned about this process from an acoustical engineer on a Honda Accord forum and it makes a significant difference in sound quality through absorption, eliminating standing waves, and decreasing outside noise - not a quality speaker enclosure but certainly better than a metal box.
I also have a sub to install to help with the low end and add some impact to the system. I have some test music that would indicate the factory system plays down to around 25 Hz however the output really falls off below 35. I'll be using an early series Alpine 12" Type R in a sealed enclosure driven by a Zapco amp which is a fairly accurate and controlled driver. I'll leave the factory system intact, including sub, and will use an Audiocontrol LCQ-1 taking signal from the factory sub which will also allow me to remotely control the sub volume from the driver's seat.
MiniDSP unit in cold weather - second unit I tried works fine
I bought another MiniDSP balanced 2x4 unit and used it to replace the original one I installed in 2018, and this MiniDSP seems to work just fine in cold weather so far. We've had 17° F (which is -8.3° C) here in Chicago yesterday and there were no issues with the MiniDSP in my G37.
The one I had bought in 2018 would **NOT** work when ambient dropped below about 45° F (7° C) - it would not output any audio. Just no output at all. I talked to MiniDSP and they said "well this thing is not intended to be used in a car, the parts are not spec'd to work below 0° C..."
But I am happy to report that the second one I tried works fine.
I'd still like to put a decent crossover between the mid and tweeter in the front doors as described elsewhere in this thread but I am afraid to remove the door cards, I have a history of breaking those plastic panel clips.....maybe I'll pay an autosound shop to put some Dynamat in the doors and allow me to get in there to solder in the crossovers while they are working.
I'd still like to put a decent crossover between the mid and tweeter in the front doors as described elsewhere in this thread but I am afraid to remove the door cards, I have a history of breaking those plastic panel clips.....maybe I'll pay an autosound shop to put some Dynamat in the doors and allow me to get in there to solder in the crossovers while they are working.
Just adding the low pass to the sqawker made a world of difference - I still need to put the high pass on the tweeter as some of the remaining harshness could be distortion at the lower end or the inexpensive electrolytic cap used by Bose.
Door panels are very easy - video below. Only challenge for me was one clip directly below the sail panel - otherwise very easy.
I know absolutely nothing about car stereo but do have a basic bedroom studio with studio monitors on foam isolation pads, studio sub, and audio interface. The more I've built out my bedroom studio setup, the more disappointing my G37's Bose factory "studio on wheels" becomes. Ever since purchasing the car, I've always noticed how harsh the upper midrange can sound with modern pop/electronic music which substantially takes away from the warmth that I know this system is capable of producing on more symphonic music. And I know there's something wrong since this that absolutely does not happen on my studio monitors at home. This thread proves all that scientifically.
I'd like to attempt this LPF crossover install to cutoff the 5khz+ entering the midrange drivers due to the positive results of everyone who's attempted this in the thread.
Does anyone know which parts I can use. I know llebcire attached his list below: However, I've noticed madisonspeakerstore no longer carries the Carli MET 3.9 mfd Mylar capacitor. Would I be able to use a higher MFD capacitance one instead? Or one of a different material? Any links, more detailed instructions, or readings would be appreciated for my noob ***. Or would y'all suggest I take it to a car stereo shop for them to do the install.
Thanks for the quick reply and showing me that part #. Any reasonllebcire purchased 4x of each? Wouldn't it just be 1x capacitor + 1x Inductor on each side to create the 2nd order crossover? Meaning he really only needed 2x inductor and 2x capacitor? Sry I'm a real noob,
I'm assuming this inductor would have the same specs as the one mentioned in the following?
Finally, once I receive the parts it would just be a matter of either soldering/wiring the inductor in series between the driver/tweeter and then wiring the capacitor in parallel between the driver and tweeter?
4 capacitors - one in the low pass one in the high pass times 2 channels, maybe?
I don't know I don't have the schematic for the crossover he made.
And the inductor - as long as it has the same inductance in millihenries ( mH) it will be fine.
If you are trying to improve "warmth" in your G37 Bose system, this crossover will reduce the shrill quality of the highs to a degree, but I still think you need to EQ out a 9 KHZ emphasis that the system seems to have; additionally there is a lack of energy in the 500 Hz area where the woofer crosses to the midrange, I found using a parametric EQ 450 Hz 1/2 octave wide about +3 dB or so helps. Makes voice and guitar sound richer, but I still think the system lacks warmth. I think the warmth thing comes from a lack below 400 Hz but you can't really boost it much because the woofer / door will store energy around that range and instead of warmth you'll get boom and mud.
Eq can make it sound a LITTLE warmer but you won't end up with the kind of warm sound you can get in a room from a monitor speaker. The car interior / doors/ etc just have resonances of various kinds that make it really hard to get the lower midrange / upper bass to sound "warm" without sounding terribly muddy.
Let me help - for crossover design, there are very specific components needed to attenuate frequencies based on the impedance of the driver you're targeting. While most speakers used by Bose for car audio are 2 ohm, @milosz tested and found the drivers used for the mid and tweeter in the G37 are 4 ohm. What I designed was a very simple crossover - crossovers in high end audio systems are tweaked based on the parameters of the drivers used as a 4 ohm speaker isn't 4 ohm across all frequencies. Additionally, there are many different crossovers and combinations of crossovers that can be used based on desired effect - I chose a 12 dB Linkwitz-Riley (LR) based on how it attenuates and flattens the response - you can use an online calculator to see how different crossover types change the sound. 12 dB means that the crossover attenuates the targeted frequencies at 12 dB per octave (see attached diagram) and is "sharper" than a 6 dB per octave but not as sharp as a 18 or 24 dB per octave. You'll also see that with the crossover I chose, the output at 9 KHz is down at least 3 dB. 12 dB is simpler (less components) and doesn't cause the phase shifts associated with other types. For the actual components, the are preferences on how the components impact sound quality. Most people agree that using an inexpensive electrolytic capacitor negatively impacts high frequencies (when the cap in inline with the high range driver) whereas when the cap is connected to the negative feed (12 dB low pass) the sound quality is irrelevant - I chose the same components to ensure they were all similar. I chose the caps I did based on the material and the cost - you could (and many high end home companies do) spend quite a bit more money on components for sound quality. Same with the inductors although typically not as concerning as the caps due to the frequencies involved. Lastly - regarding the sound quality of the lows and mids - I tackle this by installing additional peel/stick for vibration and compressed fiberglass which impacts the lows and mids (similar to a home speaker) - I caution this could be dangerous if precautions aren't taken and have a thread here.
Crossover above - response curve. 12 dB per octave means that for a crossover centered at 5 KHz, the output should be -12 dB at 10 KHz (low pass) or -12 dB at 2.5 KHz (high pass) and to accomplish this the crossover starts affecting frequencies well outside the crossover point - below the mid range starts to attenuate (slightly) <400 Hz. As the crossover slope increases (18 & 24 dB) the range of impacted frequencies shrinks as the cutoff is much sharper.
Let me help - for crossover design, there are very specific components needed to attenuate frequencies based on the impedance of the driver you're targeting. While most speakers used by Bose for car audio are 2 ohm, @milosz tested and found the drivers used for the mid and tweeter in the G37 are 4 ohm. What I designed was a very simple crossover - crossovers in high end audio systems are tweaked based on the parameters of the drivers used as a 4 ohm speaker isn't 4 ohm across all frequencies. Additionally, there are many different crossovers and combinations of crossovers that can be used based on desired effect - I chose a 12 dB Linkwitz-Riley (LR) based on how it attenuates and flattens the response - you can use an online calculator to see how different crossover types change the sound. 12 dB means that the crossover attenuates the targeted frequencies at 12 dB per octave (see attached diagram) and is "sharper" than a 6 dB per octave but not as sharp as a 18 or 24 dB per octave. You'll also see that with the crossover I chose, the output at 9 KHz is down at least 3 dB. 12 dB is simpler (less components) and doesn't cause the phase shifts associated with other types. For the actual components, the are preferences on how the components impact sound quality. Most people agree that using an inexpensive electrolytic capacitor negatively impacts high frequencies (when the cap in inline with the high range driver) whereas when the cap is connected to the negative feed (12 dB low pass) the sound quality is irrelevant - I chose the same components to ensure they were all similar. I chose the caps I did based on the material and the cost - you could (and many high end home companies do) spend quite a bit more money on components for sound quality. Same with the inductors although typically not as concerning as the caps due to the frequencies involved. Lastly - regarding the sound quality of the lows and mids - I tackle this by installing additional peel/stick for vibration and compressed fiberglass which impacts the lows and mids (similar to a home speaker) - I caution this could be dangerous if precautions aren't taken and have a thread here.
Crossover above - response curve. 12 dB per octave means that for a crossover centered at 5 KHz, the output should be -12 dB at 10 KHz (low pass) or -12 dB at 2.5 KHz (high pass) and to accomplish this the crossover starts affecting frequencies well outside the crossover point - below the mid range starts to attenuate (slightly) <400 Hz. As the crossover slope increases (18 & 24 dB) the range of impacted frequencies shrinks as the cutoff is much sharper.
Just to clarify for people looking at this schematic - the software used to generate the values for the crossover says "4 ohm woofer / 4 ohm tweeter" but in this instance the "woofer" is actually the G37 / Bose midrange driver. The software template has "woofer" in it but in fact here in this crossover it is the 3" midrange driver.
I now have an instrument to measure driver impedance; I'll post curve for the 3" mid and the tweeter later today. We may want to recalculate the crossover once we see the curve, as it is the impedance at the crossover point of 5,000 Hz that we need to plug into our equations.
ALSO NOTE that the calculated value for the capacitors is 3.98 uF. Most capacitors used for crossovers have a 10% tolerance (though some have tighter tolerance - 5%, 2% or even 1%) And with crossovers, the math will calculate a value but manufacturers only make certain values, and it is usually not possible to find the exact value specified by the crossover math. In this case the math says 3.98 uF. No manufacturers sell a capacitor with a value of 3.98 uF. There's 3.9 and 4.0 uf caps available. So I'd say in this instance we want a 4.0 uF film cap (though a 3.9 uF will work OK too,) like https://www.mouser.com/ProductDetail...Vsoqq17alGE%3D
Well here we have the impedance plots of these two drivers. You would label the 3-inch mid a 4 ohm driver because that's it's lowest impedance in it's operating range above 500 Hz. Free air resonance is just over 200 Hz meaning this thing can actually work below 500 Hz to a degree, though probably limited by power handling and distortion as you go lower in frequency below 500 Hz to maybe 300 Hz. The crossover between the door-mounted 10 inch low-profile woofer and this 3 inch mid in the Bose trunk-mounted amp is at about 500 Hz.
I think we will want to recalculate the crossover component values based on what we now know about the impedance at 5 kHz.
Given that the impedance used in the calculation (above) was 4 ohms, but the actual impedance is more like 9~10 ohms, the crossover above will actually give us a crossover at around 2.4 kHz.
That may actually be a GOOD THING. If you built this crossover and it SOUNDS good, then... it IS good. 2.5 kHz is an entirely reasonable point for a crossover to this Bose mylar dome tweeter. The 3-inch mid can work higher than 2.5 kHz, but likely the dome will be better behaved.The response plot (https://www.myg37.com/forums/audio-v...l#&gid=1&pid=2) shows the tweeter has good response from 2 kHz and up, and also acceptable distortion levels from 2 kHz on up, too. So this crossover which turns out serendipitously to be approx 2.5 kHz could be JUST RIGHT.
At a crossover point of 5 kHz this thing has an impedance of 10 ohms.
G37 Bose 3 inch midrange impedance plot
And here we have the tweeter. Above 10,000 it approaches 4 ohms (5.8 ohms actually) so again you'd typically call it a nominal 4 ohm driver. At 5 kHz it has an impedance of about 9.3 ohms.
I'll point out that when I installed the crossover, I only installed the low pass on the mid as I didn't have time to take apart the tweeter and re-wire it. Bose epoxied a small capacitor to the tweeter housing and my goal is to go back and add the high pass later. From your earlier posts, I gathered the harshness was attributed to the mid range playing without a low pass and starting to break up above 5k. What I installed made a significant difference in sound quality and I can definitely hear the tweeters now with most of the harshness gone (referenced changes I heard in different tracks).
Since you have access to another set of drivers, are you planning on installing a crossover on your test set outside the door?
05-21-2019, 08:25 AM
