JBL L Series (1990s)

[BMWCCA]

Your DX4200 has a tone generator?

[Titanium Dome]

Your DX4200 has a tone generator?

Yes, it does, when you hook it up to a an AE2000! (Kinda left that part out, didn't I?)

[JBLAddict]

the detailed illustration on the effect of placement prompted me to download some test tones, and experiment with frequency response and see if I could validate any perceived differences in LF in my L7s. Pardon my ignorance if any of this sounds remedial/obvious to the more experienced, I was simply looking for some data to work with given the equipment I have on hand

in short, I set the speakers in two positions, one at 15" from the backwall, one 36" from the back wall (both 36" from the side walls). 15" being the WAF spot, 36" the JBLAF spot

I played .mp3 files (from a USB into my Oppo) of single tones starting at 30Hz up to 90Hz in 10Hz increments, and then one at 120, 150, 180, 210, then measured the response from the primary seating position with the RS/SPL.

I'll post the numbers later, but the results are interesting. In some cases I measured (and heard), 4-5dB difference between the two positions by changing the tone 10Hz from a point where the two positions were previously equal, wow talk about sensitivity! If I moved my seating position closer/further the effect magnified further, 7-8dB

The most compelling result comes from walking around the room (and up and down) where the measurements and audible tones rise and fall up to 10dB (similar to what Doug posted).....the nulls and peaks I've seen in the calculators really came to life with this kind of isolation, and opened my eyes to how sensitive both speaker and listener position are to the listening experience.

From what I observed, altering the position of the speaker and/or listener, changes the response of multiple frequencies (in some cases in opposing directions) so finding a flat position seems impossible? One could pick a single position of the speaker and listener and use EQ to flatten (if that is one's goal), but I'm wondering how you go about using speaker position without EQ to do anything other than try to elminate the biggest nulls and peaks?

I'm sure I'm just hitting the tip of the iceberg on this complex science and the world of calibration tools out there, would love to hear some feedback on whether my conclusions hold any water. In some ways, it explains to me why I've enjoyed the L7 both close and far from the wall, as in each position, the peaks simply shift to a different frequency, question is which one does my ear prefer?

[localhost127]

in short, I set the speakers in two positions, one at 15" from the backwall, one 36" from the back wall (both 36" from the side walls). 15" being the WAF spot, 36" the JBLAF spot

I'll post the numbers later, but the results are interesting. In some cases I measured (and heard), 4-5dB difference between the two positions by changing the tone 10Hz from a point where the two positions were previously equal, wow talk about sensitivity! If I moved my seating position closer/further the effect magnified further, 7-8dB

The most compelling result comes from walking around the room (and up and down) where the measurements and audible tones rise and fall up to 10dB (similar to what Doug posted).....the nulls and peaks I've seen in the calculators really came to life with this kind of isolation, and opened my eyes to how sensitive both speaker and listener position are to the listening experience.

I'm sure I'm just hitting the tip of the iceberg on this complex science and the world of calibration tools out there, would love to hear some feedback on whether my conclusions hold any water. In some ways, it explains to me why I've enjoyed the L7 both close and far from the wall, as in each position, the peaks simply shift to a different frequency, question is which one does my ear prefer?

all of this is likely due to SBIR off the front wall (and side walls). as you move the L7s closer to the front wall, you increase the SBIR frequency, as the distance between the wall and the L7 is the distance of 1/4freq (+all multiples of that frequency), which will result in a null. same thing for any reflections off the front wall arriving in phase with the original source and resulting in a peak.

you will also have peaks + nulls from reflections off the back wall that is also adding to this mix. this is why a few pages back i pretty much said the best practices for placement of your L7s is irrelevant, and you should really just spend all day with different placements and measuring frequency response to find the real world best placement. the L7 documentation gives you a good start, but even that isn't anywhere near enough information to get optimal placement...you could still spend all day fine tuning in one specific/general area.

From what I observed, altering the position of the speaker and/or listener, changes the response of multiple frequencies (in some cases in opposing directions) so finding a flat position seems impossible? One could pick a single position of the speaker and listener and use EQ to flatten (if that is one's goal), but I'm wondering how you go about using speaker position without EQ to do anything other than try to elminate the biggest nulls and peaks?

right - you need to start adding porous bass traps or pressure based (helmholtz, etc) to combat the peaks and nulls. getting placement right (to the best performance possible) will be a good start. but freq response is only one part of the story. it's the decay times (modal ringing) that really bother me the most. more so than a choppy freq response. bass traps will also tame these and can be measured with a waterfall (time domain) plot.

eq can't do much of anything for infinite nulls, let alone curing any modal ringing/decay issues. not to mention, eq will still only benefit the single listening position, not the entire room (especially true if you have a large sweet spot). that's why placement and good room treatment come first, and eq is just icing on the cake (hopefully for just taming some peaks below 40hz

cheers, (and thanks for the photos, Ti Dome!)

[gferrell]

It really gets complicated while dancing around the room to the music.

[audiomagnate]

into an acoustical nightmare of a room. A sparsely furnished maybe 12 X 12 x 8 box and they sound ridiculous. Whatever bass resonance they have to start with is out of control. They sounded great in my irregularly shaped master BR, but they do no love a cube. These little things can get some serious bass vibes going. EDIT: Pulling them way out into the room helps dramatically, but this is my daughter's BR, not a near field mixing room. They actually sound great in the new position.

[JBLAddict]

Hz 15" 36"
30 89 88
40 93 91
50 91 88
60 85 83
70 90 91
80 92 94
90 88 91
the following mp3's were from a different web site so the baseline output might be different from above
120 82 88
150 82 80
180 83 86
210 81 77
280 85 87

[localhost127]

Hz 15" 36"
30 89 88
40 93 91
50 91 88
60 85 83
70 90 91
80 92 94
90 88 91
the following mp3's were from a different web site so the baseline output might be different from above
120 82 88
150 82 80
180 83 86
210 81 77
280 85 87

unfortunately, there really is no conclusion to draw from such data.
i bet it was fun though
wish i was able to play 30hz > 90dB in my place .... damn condo.

[JBLAddict]

Perhaps, but for me validated my experience that having them closer to the wall did in my room/seating position not cause any glaring audible anomalies. and possibly in my configuration, the lesser output from 70-120Hz might make the 15" position more balanced

was running pre-amp about 54% to the 240WPC soundcraftsmen, lot of power in that puppy, clearly

The fact that the L7 is only 3dB down from 40Hz to 30Hz is as testament to the LE120H-1 in that cabinet. The '92 Audio Review of the L7 I posted a few pages back which stated the L7 "LF keeps up with the best speakers he's reviewed as well as several subwoofers" is well backed by my data.

[Titanium Dome]

unfortunately, there really is no conclusion to draw from such data.
i bet it was fun though
wish i was able to play 30hz > 90dB in my place .... damn condo.

Well actually there is. There's an inversion of relative output between 60 and 70 Hz depending on distance to a reflecting surface.

I wouldn't make much of any changes above 90 Hz, but with respect to the changes below that, further investigation might further elucidate the transition point in terms of distance to the wall. Of course, this would be true of most speakers with rear-firing ports in varying degrees, but it's unusual in the JBL world for a side-firing woofer to be part of the equation.

[localhost127]

and possibly in my configuration, the lesser output from 70-120Hz might make the 15" position more balanced

was running pre-amp about 54% to the 240WPC soundcraftsmen, lot of power in that puppy, clearly

The fact that the L7 is only 3dB down from 40Hz to 30Hz is as testament to the LE120H-1 in that cabinet. The '92 Audio Review of the L7 I posted a few pages back which stated the L7 "LF keeps up with the best speakers he's reviewed as well as several subwoofers" is well backed by my data.

no, the data is not of high enough resolution where you can make this comment. you cannot assume linear coorelation between your data points. you could have a massive peak or null that is only 3hz wide (e.g. 65-68 could be an infinite null), but because of your resolution (in 10hz increments)), you never see this. it's the same problem people have when measuring frequency response of the low end and then applying ridiculous smoothing... you really need to see the measurements of each frequency in 1Hz steps on the low end, because things can vary widely due to room modes, LBIR, SBIR, etc.

The fact that the L7 is only 3dB down from 40Hz to 30Hz is as testament to the LE120H-1 in that cabinet.

again - this is completely false. you cannot say that the L7 is only 3dB down from 40hz to 30hz. you can only say there is 3dB difference between EXACTLY 30hz and EXACTLY 40hz in your room (with a massive host of variables). with your data, you do not know what is happening between 30 and 40hz. there could be a 20dB peak at 35hz. the data does not support any conclusion that really offers any proof. also, you cannot attest dropping 3dB from 40hz to 30hz to the L7s driver, because it could actually be -15dB, but maybe due to your room dimensions, you have a mode at 30Hz, which is then bumping the response to a -3dB difference. if you measured this in an anechoic chamber, then you could make that claim. while a fun exercise, you cannot draw much of any conclusion from the data.

[grumpy]

I spent the last week in an anechoic chamber (albeit microwave)...
still damn annoying audio-wise. That's pretty close to the last place
I'd want to actually listen to a stereo system. Great place to test for
anomalous stuff (vs design), but ultimately a "field test" is required,
with all of it's lumps and bumps.

I would say that warble tone testing would be an improvement over
plain discrete tones if attempting to glean some idea of a system/room
response by ear, but just saying it's crap and you're full of it too
isn't helping anyone.

[Eaulive]

I spent the last week in an anechoic chamber (albeit microwave)...
still damn annoying audio-wise. That's pretty close to the last place
I'd want to actually listen to a stereo system. Great place to test for
anomalous stuff (vs design), but ultimately a "field test" is required,
with all of it's lumps and bumps.

I would say that warble tone testing would be an improvement over
plain discrete tones if attempting to glean some idea of a system/room
response by ear, but just saying it's crap and you're full of it too
isn't helping anyone.

I got used to hear pink noise and very often I have a better idea of a system by playing a pink noise than playing some unknown musical material

[JBLAddict]

yes, help would be helpful

[localhost127]

I would say that warble tone testing would be an improvement over
plain discrete tones if attempting to glean some idea of a system/room
response by ear, but just saying it's crap and you're full of it too
isn't helping anyone.

i dont understand? my post was in no way negative towards the user. i didn't say it was "crap" --- i said you are "unable to draw a conclusion" from the test and gave an explanation as to why.

if someone can tell me what concrete conclusions you are able to draw from taking measurements every 10hz (regarding the low end), im all ears.

a standard frequency response plot (with no smoothing) and a waterfall plot will give you the full image on the low end.

there's too many variables going on in the room with SBIR, LBIR, room modes, nulls based on distance between the user and the rear wall (1/4wavelenght distance cancellations) etc, for you to make any conclusion based on moving the speakers from one location to another.

for example - what if by moving it from one location to another (and taking measurements every 10hz), you "think" you are getting a smoother response, but what if in actuality you have created big (+10dB) peaks and infinite nulls at frequencies of which you didnt measure (63hz, 72hz, etc).. ?? there is not enough data to conclude whether you made things better or worse. there is simply not enough data points.

edit: and of course, if one person were to determine the optimal distance from the front (and or side) walls for their L7s, that in no way is transferable to another room, as the room is dictating the low frequency response.

also, by moving the L7s closer to the front wall, you are increasing the frequency of SBIR, which then makes it easier to treat with porous absorption.