The "Share your tune measurements and methods" thread
I'm really curious to see what the measured response of other users systems "looks" like... I know the measured response doesn't tell the whole story and that everyone "hears" differently and all of that, but I still find it helpful to see the measured response of other peoples systems.... For whatever reason, I've found that it's pretty rare for people to post actual measurements of their systems and I'm not quite sure why that is (maybe they don't want people being critical of it?). The first thing that newer folks look for advice with is their measured response, simply because they usually don't have anything else to compare theirs to in order to see where they are at when it comes to overall response measurements.
I use rear speakers in my car (bandpassed and lower levels), but for simplciity, I'm going to start the thread by posting measurements of my front channels only. I've come to realize that when rear speakers are involved, the response of the front channels are the main thing - and they should match your curve without rear speakers playing. While rear speakers do add something, they are just "extra" and should not be included in the measurements (in my opinion). I will, however, show some example of what the rear speakers do to the response - and with different delay values (I've found that the amount of delay really changes the overall response).
Anyway, to start things off, here is the current measured response of my front channels (6x9 midbass speakers in doors,and 3.5" coaxial speakers in dash - factory speaker locations in my Dodge Challenger). I also have a small 8" under-seat sub (JBL BassPro SL). No other subs in car. Currently, I'm acoustically crossing the door/dash speakers at 400hz with 24dB LR slopes. I don't have individual speaker measurements here yet, but I will eventually post those as well, including the tuning "process" that *I* use (which I'm not claiming is even "correct" - just how I currently do it).
I measure with a UMIK-1 using REW. I use a 90-degree calibration file and hold the MIC straight up and down and move in from ear to ear and also do some figure-8's with the tip of the MIC around each ear (since I figure that my ears are never going to be in exactly the same place all of the time).
Here is the measured response of my current tune. The midbass is "exaggerated" a litte when sitting in my garage, because I've find that when I'm driving, a lot of the midbass is "masked" by road noise - so setting it to be a little higher makes things sound better while actually driving, which is the ultimate goal, IMO (at least for me).... I also have a pretty big "notch" between 2k and 5k as that stops harshness, especially at higher volumes - at least for me.
https://uploads.tapatalk-cdn.com/202...5377e7c5fe.jpg
All criticism is welcome! I would love to see the measured response of other peoples systems!
Re: The "Share your tune measurements and methods" thread
I will post a graph of my overall soon.... but mine is no where near as good as yours.... shit, my goal is have my graph look as good as yours... ha....as your is pretty dam good ! That "notch" between 2k and 5k is a good one... seems like the smaller the vehicle (and the closer the speakers are to your listening position, the more that type notch is needed)....
what smoothing did you use? 1/6th?
Re: The "Share your tune measurements and methods" thread
Quote:
Originally Posted by
Brego
I will post a graph of my overall soon.... but mine is no where near as good as yours.... shit, my goal is have my graph look as good as yours... ha....as your is pretty dam good ! That "notch" between 2k and 5k is a good one... seems like the smaller the vehicle (and the closer the speakers are to your listening position, the more that type notch is needed)....
what smoothing did you use? 1/6th?
Erm... 1/3 it says in his measurement name lol ;)
Re: The "Share your tune measurements and methods" thread
I will share various measurements and what they did and didn’t show over the last year and a bit I have when I get my laptop in a bit as I have some time on my hands with the virus kicking about
Re: The "Share your tune measurements and methods" thread
I posted a lot of Tuning related information in my old car's build thread. Starting with Post #11 here: https://www.caraudiojunkies.com/show...ght=2006+CIVIC
But, unfortunately, I guess those photos are no longer available. Still, there's a few good posts in between in there. So, I can say I definitely had no problem sharing... at least back then. Once I start actually tuning in my car again I'll make sure to post that info in my new build log.
Also, you guys might be interested in this if you haven't seen it:
https://www.caraudiojunkies.com/show...r-RTA-purposes
Re: The "Share your tune measurements and methods" thread
Your overall curve looks extremely similar to mine. My sub region is boosted slightly more than the typical JBL/Audiofrog/Whitledge targets for personal preference. I also notch the 2k-5k region by ear after I hit a linear target curve. It usually ends up 3-6db lower than the surrounding 1k and 6k regions in order not to be fatiguing or irritating to my ears. What I find odd is that having a ruler flat response on my home audio system doesn't produce the 2k-5k harshness that I experience in car audio systems. I've tuned probably around 40 different cars now, with all types of speakers, amps, and DSPs. Literally ALL of them have some problem areas from 2k-5k. No matter what target curve is used either. Trying to determine what causes this, and whether or not I should build a custom target curve that accounts for this at the start rather than applying the changes by ear later.
Re: The "Share your tune measurements and methods" thread
Quote:
Originally Posted by
mattkim1337
Your overall curve looks extremely similar to mine. My sub region is boosted slightly more than the typical JBL/Audiofrog/Whitledge targets for personal preference. I also notch the 2k-5k region by ear after I hit a linear target curve. It usually ends up 3-6db lower than the surrounding 1k and 6k regions in order not to be fatiguing or irritating to my ears. What I find odd is that having a ruler flat response on my home audio system doesn't produce the 2k-5k harshness that I experience in car audio systems. I've tuned probably around 40 different cars now, with all types of speakers, amps, and DSPs. Literally ALL of them have some problem areas from 2k-5k. No matter what target curve is used either. Trying to determine what causes this, and whether or not I should build a custom target curve that accounts for this at the start rather than applying the changes by ear later.
I'm still messing with that 2k-5k notch - as well as the 6k+ "shape". I've found that small differences (even .5dB) can make noticable differences up in that range. I'm guessing that the harshness in that range is due to reflections in the vehicle environment. Just a guess though.
I've also found that the overall sound while sitting in a car in your garage is very different than the overall sound while you're out driving around. So I'll have what I think is the "perfect" tune, until I go take an actual drive. :-)
What I plan on doing is this... Once I get my overall system response to exactly where I want it, I'm going to use REW to create individual-speaker "target curves" from the *actual* measured response of each individual speaker. Then, I'm going to re-EQ each individual speaker from scratch using those individual-speaker curves. This *should* reduce the amount of EQ needed to achieve the results I'm looking for. Basically, right now, I'm using tons of EQ bands to "tweak" the response as I go - but I'm sure I'll be able to EQ the system to reach the same response will a lot less EQ once I get the response exactly where I want it.
For example, originally, I EQ'd each speaker individually to match the curve - but then when I played L+R speaker pairs, I noticed that phase and/or reflection issues (that I can't control) required me to do more "speaker pair" EQ in order for the L+R speaker pairs to match the curve. Once the speaker-pairs matched the curve, I then had to use even more "speaker pair" EQ to make the overall response match the curve (again, due to phase and/or reflection issues when all speakers are playing). So when all is said and done, if I make new, individual-speaker target curves from the *actual* individual speaker measurements, I *should*, in theory, be able to EQ each speaker individually to that target curve and the end-result wil be an "all spekaer" response that matches the curve with the least amount of EQ possible.
At least that is my theory. :-) I'll let you know how it all works out (and will document the process along the way to "show" what I did).
Re: The "Share your tune measurements and methods" thread
Quote:
Originally Posted by
mattkim1337
Your overall curve looks extremely similar to mine. My sub region is boosted slightly more than the typical JBL/Audiofrog/Whitledge targets for personal preference. I also notch the 2k-5k region by ear after I hit a linear target curve. It usually ends up 3-6db lower than the surrounding 1k and 6k regions in order not to be fatiguing or irritating to my ears. What I find odd is that having a ruler flat response on my home audio system doesn't produce the 2k-5k harshness that I experience in car audio systems. I've tuned probably around 40 different cars now, with all types of speakers, amps, and DSPs. Literally ALL of them have some problem areas from 2k-5k. No matter what target curve is used either. Trying to determine what causes this, and whether or not I should build a custom target curve that accounts for this at the start rather than applying the changes by ear later.
I'm excited to see others noticing the that 2k-5k region too.
I'm really curious what causes that too!
Re: The "Share your tune measurements and methods" thread
Quote:
Originally Posted by
Justin Zazzi
I'm excited to see others noticing the that 2k-5k region too.
I'm really curious what causes that too!
Hmmm...I've always chalked it up to boundary-loading/waveguide/horn-loading effect of the drivers in close proximity to reflective surfaces & "corners" in the vehicle environment. In a typical home audio setup, that energy is much lower in level before it reaches any reflective surfaces and we are not usually in such a nearfield environment. The energy from that very sensitive range of our hearing is dissipated into a much larger space and not nearly as "reinforced" by close boundaries in a typical home listening environment.
But I'd definitely be interested in learning what else may be contributing to this.
2 Attachment(s)
Re: The "Share your tune measurements and methods" thread
I apparently didn't save my tweeter measurements or speaker pairs/overall on this day, but here is my most recent tune. On average this uses like 3-5 bands of EQ for each speaker.
I use a 90 degree cal file from cross spectrum labs and a dayton usb mic and REW. I wave the mic completely at random. Currently I sit in the drivers seat and do this, but after talking with Erin on his post and Justin, I think I am going to try sitting in the rear seat and doing this.
I use Justin's tuning companion to load house curves
I use whatever slope, crossover type, frequency needed to match the acoustic 24 db LR curve and protect the speakers.
My current sub to midbass crossover is 100hz. For some vehicle/reflection reason I can't get anything to play much under 100hz without falling off from my doors in my listening position (they play way below when measured close to the speaker).
My midbass to midrange is 350hz
My midrange to tweeter is 3500hz
First picture is just the speakers, the second includes the house curves loaded
Attachment 9842
Attachment 9841