The New Advent Loudspeaker Review (Vintage Speaker)

[Dennis Murphy]

Perhaps it would have been clearer if I had stated I designed a real crossover, i.e.

 

[Mr. Widget]

I guess we are talking about different speakers. I measured an unmodified L100A.

 

[Dennis Murphy]

I guess we are talking about different speakers. I measured an unmodified L100A.

We're getting confused. My plot is for the unmodified JBL. Given the rudimentary crossover, there's no way it will measure differently than what I posted. You can see the individual raw driver measurements in the plot. I redid the crossover to try and improve matters. It's still far from perfect and bringing down the mids and highs has the effect of accenting the poor box tuning, which you can see by my measurement of the revised 100:

 

[Mr. Widget]

We're getting confused. My plot is for the unmodified JBL. Given the rudimentary crossover, there's no way it will measure differently than what I posted. You can see the individual raw driver measurements in the plot. I redid the crossover to try and improve matters. It's still far from perfect and bringing down the mids and highs has the effect of accenting the poor box tuning, which you can see by my measurement of the revised 100:

Their box tuning was intentional. JBL used to always put in that bump. Not my idea of a good design, but they sold a pile of these things!

Well, based on your comments the measurement is anechoic above 250Hz so we really shouldn't look at the plot below that frequency. The speaker appears to be about +/- 2.5 dB from 250Hz to 13-14KHz. For the era that is pretty good. I assume that is not a ground plane measurement of the woofer.

The tweeter is obviously crap, but then other than compression drivers I don't think JBL made any decent tweeters before they bought Audax.

 

[Dennis Murphy]

Their box tuning was intentional. JBL used to always put in that bump. Not my idea of a good design, but they sold a pile of these things!

Well, based on your comments the measurement is anechoic above 250Hz so we really shouldn't look at the plot below that frequency. The speaker appears to be about +/- 2.5 dB from 250Hz to 13-14KHz. For the era that is pretty good. I assume that is not a ground plane measurement of the woofer.

The tweeter is obviously crap, but then other than compression drivers I don't think JBL made any decent tweeters before they bought Audax.

I don't think you're reading my posts. That might be a good idea in general, but it is causing confusion here, assuming anyone else is interested in all of this. The plot I just posted has nothing to do with the original design. That's the plot for my revised crossover. This is how the original 100 measured, and it's not +/- 2.5 dB.

 

[Mr. Widget]

I don't think you're reading my posts. That might be a good idea in general, but it is causing confusion here, assuming anyone else is interested in all of this. The plot I just posted has nothing to do with the original design. That's the plot for my revised crossover. This is how the original 100 measured, and it's not +/- 2.5 dB.

Apparently I have missed your post where you explained what you are demonstrating. My apologies for adding to the confusion.

That said, I think we both agree that the stock 4310/4311 and L100 are not an ideal solution.

 

[gino1961]

Their box tuning was intentional. JBL used to always put in that bump. Not my idea of a good design, but they sold a pile of these things!

Well, based on your comments the measurement is anechoic above 250Hz so we really shouldn't look at the plot below that frequency. The speaker appears to be about +/- 2.5 dB from 250Hz to 13-14KHz. For the era that is pretty good. I assume that is not a ground plane measurement of the woofer.

The tweeter is obviously crap, but then other than compression drivers I don't think JBL made any decent tweeters before they bought Audax.

it's not only the bump but also the following dump around 150 Hz that shocks me deeply
The frequency response in the woofer range varies +/-10dB Never seen something like that
this has been a bad surprise for me
the woofer looks of tremendous quality
i would try closing the reflex first just to see if it is the problem

 

[Dennis Murphy]

it's not only the bump but also the following dump around 150 Hz that shocks me deeply
The frequency response in the woofer range varies +/-10dB Never seen something like that
this has been a bad surprise for me
the woofer looks of tremendous quality
i would try closing the reflex first just to see if it is the problem

As I explained earlier, the dip at 150 Hz is floor bounce cancellation. Its exact location and magnitude is a matter of the distance to the floor, not the inherent quality of the woofer. Any woofer would do that if placed in the 100 cabinet. That's why a few companies like to place the woofer at floor level, which avoids the issue but can cause other problems. What's unusual about the JBL plot is the height of the room mode at 70 Hz. That's a matter of woofer selection and box tuning. When designing a crossover, I like to work from room plots in the bass region because it helps me determine the right amount of baffle step compensation. You can't do that with a ground plane measurement or by splicing a nearfield plot onto the anechoic portion of the measurement. In a speaker like the JBL, a full 6 dB of baffle step compensation would produce unbearably heavy bass. Of course, the stock JBL "crossover" not only doesn't provide any bsc, it basically doesn't compensate for anything.

 

[gino1961]

Thank you very much for again for all your great explanation
This also tells me how important is the speaker.room interaction if a floor can cause that bump
But that maybe can be cured with a thick carpet A wrong design implies mods on the speakers
Room measurements seem mandatory The fact that people liked this speaker anyway is puzzling
Human senses are not that accurate i am afraid

 

[Dennis Murphy]

Thank you very much for again for all your great explanation
This also tells me how important is the speaker.room interaction if a floor can cause that bump
But that maybe can be cured with a thick carpet A wrong design implies mods on the speakers
Room measurements seem mandatory The fact that people liked this speaker anyway is puzzling
Human senses are not that accurate i am afraid

Bass frequencies aren't much phased by carpeting. It's still controversial whether floor bounce cancellation is audible enough to be worth fixing. I honestly don't know.

 

[gino1961]

Bass frequencies aren't much phased by carpeting. It's still controversial whether floor bounce cancellation is audible enough to be worth fixing. I honestly don't know.

Thanks again Maybe some simulation would provide some answers about the box and the reflex
I am really shocked that the performance of that amazing woofer is so poor
I thought it were state of the art
I think that the one in the l166 is a different beast

 

[Dennis Murphy]

Thanks again Maybe some simulation would provide some answers about the box and the reflex
I am really shocked that the performance of that amazing woofer is so poor
I thought it were state of the art
I think that the one in the l166 is a different beast

It's not the woofer's fault. It's the undersized cabinet.

 

[gino1961]

It's not the woofer's fault. It's the undersized cabinet.

Thanks again I am afraid i lack some essential basic knowledge clearly
I will try to use one of those sw for cabinet sizing to try to understand which woofer parameters count for designing a closed cabinet
i don't like bass reflex a lot
I have a little project in mind I have already mids and tweeters to be mounted like in the famous dq10 on a separate baffle
The woofer should cover the range up to about 6-700Hz
I need to study a lot more

 

[Pete Basel]

The primary problem with the L100 bass is that its tuned too high, bring the tuning down to the
mid 30s or low 40s and that will help a lot. But, yes the box is too small so the low tuning will
minimize excursion at those frequencies allowing for at least 6 dB of boost on the woofer that
could be provided by a miniDSP or something similar, or an analog solution.

 

[gino1961]

The primary problem with the L100 bass is that its tuned too high, bring the tuning down to the
mid 30s or low 40s and that will help a lot. But, yes the box is too small so the low tuning will
minimize excursion at those frequencies allowing for at least 6 dB of boost on the woofer that
could be provided by a miniDSP or something similar, or an analog solution.

hi thanks a lot for the very kind and helpful advice
i wonder how they could miscalculate the speaker weird
if size were a constraint they should have used another woofer

 

[Pete Basel]

@gino1961 I'm not a JBL fan so I don't know the details as far as if the L100 woofer was made specifically
for the speaker or if they just grabbed something from their standard woofers. It is really more of an
acoustic suspension woofer given the very low Fs. I think that they just chose the high tuning based on
getting the impressive boom that they wanted. The low tuning that I suggested will only provide good
bass with some EQ boost due to the small box size. They did make a larger tower version of the L100, I
think the L100T.

 

[Pete Basel]

It would be nice if a mod could split off all this L100 talk to another thread with a proper title.