Hello Ian
Basically wondering if he goes out flat or uses a flat shelf with no tilt above 1K.
Rob 🙂
Really? Older systems and XPL series nice drivers have a couple pairs
Rob 🙂
https://www.lansingheritage.org/images/jbl/specs/home-speakers/1988-cascade/page2.jpg
Your linked page says the XO is 400Hz (2/3 down). I haven't seen any mid dome being capable of the midrange and going down to 100Hz at the same time.
Now I get it. That's the Harmon Target curve not a dome driver curve, was asking profiguy what he didn't like about the Target curve above 1K go back and re read 839 Just a misunderstanding about the post.
That Cascade system is the lowest I know about from JBL and that never made it to market was just at CES.
Rob 🙂
I'd regard the so called Harman curve "pleasant" but not accurate by any stretch. It depends on your listening purpose and how loud the overall playback level is. The listening distance also matters. Add to this the typical EQ and compression used in mastering most modern recordings, it can end up sounding very thick, muddy and congested when you apply a down tilted EQ curve. At a minimum it will probably be too bass heavy and that's not pleasant with all types of music to begin with.
For people who prefer accuracy and transparent sound, the treble you end up with using the Harman curve can be too laid back, especially comparing the upper midrange around 4k not being compensated for at specific volume levels ie. perceived loudness, requiring a few dB dip at lower Q centered around that 3.5 - 4k area.
For people who prefer accuracy and transparent sound, the treble you end up with using the Harman curve can be too laid back, especially comparing the upper midrange around 4k not being compensated for at specific volume levels ie. perceived loudness, requiring a few dB dip at lower Q centered around that 3.5 - 4k area.
Attachments
Note that the Harman curve is not a room target curve, but a curve that shows what listeners preferred, subjectively. I'd rather use their predicted curve as a room target, to avoid long-term listener fatigue.
I don’t think anyone can predict what someone should or shouldn’t expect from those Harman - Floyd Toole figures findings. There are numerous dimensions or variables that come into play.
Largely at a technical level the tilt is used where wide angle CD horns are used as in some of JBL Synthesis SOA systems. According to Greg Timbers (ex chief engineer at JBL) a slight linear tilt can be of benefit depending on the room acoustics.
Unless you understand what is going on it’s best left alone according to Greg Timbers.
But with Dirac and the Lyndorf room correction people can make informed decisions and save their preferences.
Greg Timbers is currently a beta tester for the new DEQX platform that will be released soon we hope. Greg has over four decades of experience and expertise in bringing some of the most iconic loudspeaker systems to the market.
Because you nor me or anyone can predict the directivity or constant directivity of any one loudspeaker or a particular room acoustics making a generalisation is not appropriate.
Within those curves it is not just a snapshot in time but how long in the time domain the reverberant sound field takes to attenuate to a known level. It’s the overall balance of the direct versus indirect sound field in and around the listener position that we hear as normal, dull, brilliant or presence in tonal balance in the time domain.
The ear listens continuously after all. When the percentage of indirect sound exceeds the direct sound by a sufficient margin and doesn’t attenuate quickly enough it will sound too brilliant, bright or have too much presence.
The ear/ brain can to some extent compensate just like we do when to listen to someone in a crowded restaurant. But there are thresholds.
Home theatre
What we do know from experienced professional installers- integrators is that the customer will choose his / her own EQ preferences after listening to the Harman Curve. Invariably they like more bass.
It simply comes down to what you listen to, how you like to listen to it and through what equipment. Thats all that matters in the end. It’s entertainment after all and no one can dictate right from wrong as far as someone else’s situation is concerned. 🙁
Program material
Back in the day (1970’s) most integrated amps had tone controls and for good reason. There are significant differences in the low and high frequency tonal balance on vinyl recordings. So common sense prevails and you adjust it until it sounds balanced to you. I did this all the time.
Peter Comeau points this out in his podcast with John Darko. He says some remasters are good and some are not. This is where people reach different points of view.
Modern mastering and re masters are done somewhat differently pending on the media according to Comeau. Given he has actually worked in virtually every facet of the hifi industry in nearly the past five decades l tend to trust his judgements. Peter brings clarity to the hobby in a way that very few people can.
He is currently the director of audio at lAG which owns a number of brands.
The most significant variable is in the tonal balance of recordings. Should you leave it flat like the original?
Tonal adjustments
No not necessarily says Douglas Self who has recently published a very high quality Blaxandall tone control in the AudioExpress publication. The more accurate your system the more marked these variations will be from one recording to another.
Largely at a technical level the tilt is used where wide angle CD horns are used as in some of JBL Synthesis SOA systems. According to Greg Timbers (ex chief engineer at JBL) a slight linear tilt can be of benefit depending on the room acoustics.
Unless you understand what is going on it’s best left alone according to Greg Timbers.
But with Dirac and the Lyndorf room correction people can make informed decisions and save their preferences.
Greg Timbers is currently a beta tester for the new DEQX platform that will be released soon we hope. Greg has over four decades of experience and expertise in bringing some of the most iconic loudspeaker systems to the market.
Because you nor me or anyone can predict the directivity or constant directivity of any one loudspeaker or a particular room acoustics making a generalisation is not appropriate.
Within those curves it is not just a snapshot in time but how long in the time domain the reverberant sound field takes to attenuate to a known level. It’s the overall balance of the direct versus indirect sound field in and around the listener position that we hear as normal, dull, brilliant or presence in tonal balance in the time domain.
The ear listens continuously after all. When the percentage of indirect sound exceeds the direct sound by a sufficient margin and doesn’t attenuate quickly enough it will sound too brilliant, bright or have too much presence.
The ear/ brain can to some extent compensate just like we do when to listen to someone in a crowded restaurant. But there are thresholds.
Home theatre
What we do know from experienced professional installers- integrators is that the customer will choose his / her own EQ preferences after listening to the Harman Curve. Invariably they like more bass.
It simply comes down to what you listen to, how you like to listen to it and through what equipment. Thats all that matters in the end. It’s entertainment after all and no one can dictate right from wrong as far as someone else’s situation is concerned. 🙁
Program material
Back in the day (1970’s) most integrated amps had tone controls and for good reason. There are significant differences in the low and high frequency tonal balance on vinyl recordings. So common sense prevails and you adjust it until it sounds balanced to you. I did this all the time.
Peter Comeau points this out in his podcast with John Darko. He says some remasters are good and some are not. This is where people reach different points of view.
Modern mastering and re masters are done somewhat differently pending on the media according to Comeau. Given he has actually worked in virtually every facet of the hifi industry in nearly the past five decades l tend to trust his judgements. Peter brings clarity to the hobby in a way that very few people can.
He is currently the director of audio at lAG which owns a number of brands.
The most significant variable is in the tonal balance of recordings. Should you leave it flat like the original?
Tonal adjustments
No not necessarily says Douglas Self who has recently published a very high quality Blaxandall tone control in the AudioExpress publication. The more accurate your system the more marked these variations will be from one recording to another.
Its quite interesting how much some of the highest regarded speaker designs can differ in tonal balance while actually being considered reference grade, ie useful as for monitoring recordings. What does become obvious is that relative phase response has to be smooth without abrupt swings along with fast, well controlled spectral decay / minimal stored energy and smooth directivity, all across the critical 300 hz to 10 kHz range. Anything with sharp, high Q resonances which decay slowly won't qualify. Many horn loaded compression drivers suffer from large amounts of energy storage problems which introduce too much of their own sound to an uncorrectable point. That's not to be confused with well designed, lower compression ratio wave guided drivers, which can exhibit better directivity across the lower midrange, where its much harder to obtain even in room response with most speakers in a typical listening environment.
One of the biggest issues using 2 - 3" midrange domes along with larger diameter cone drivers is the abrupt shift in radiation pattern from the wider radiating dome midrange to the narrow radiating cone driver at the crossover point. This causes an abrupt directivity change which can only be corrected using a waveguide on the midrange dome.
A speaker designed to sound flat on axis in an anechoic environment will sound thick and bloated in the upper bass / lower mids when the speaker is placed in a typically reflective listening environment. Rather than fixing the problem correctly, some designers work around this issue by compromising the lower midrange linearity and obtaining a happy medium between both the anechoic and in room FR. On a "better" design, this will also include compensating for speaker placement in regards to side and rear walls. The net result is a speaker that is overly sensitive to placement and tends to send very different in various listening locations throughout the room with a narrow sweet spot.
Getting the lower midrange to sound right is IMO the hardest aspect of designing an accurate sounding speaker with linear in room FR in a variety of listening environments.
A larger dome mid with a WG can be versatile for use in a 3 way and is IMO the more suited type of driver to use when trying to achieve better lower midrange directivity and overall linearity compared to a cone driver by itself, which often exhibits less than acceptable linearity and directivity in the upper mids. Some people are ok with crossing a larger tweeter lower to a larger cone mid, but then you risk disrupting that delicate 2 to 5k area with a crossover point. I'd honestly rather have a crossover point on the higher end of that range than on the lower end. Trying to reproduce that range using only one driver is IMO the key to designing an overall more accurate sounding speaker.
One of the biggest issues using 2 - 3" midrange domes along with larger diameter cone drivers is the abrupt shift in radiation pattern from the wider radiating dome midrange to the narrow radiating cone driver at the crossover point. This causes an abrupt directivity change which can only be corrected using a waveguide on the midrange dome.
A speaker designed to sound flat on axis in an anechoic environment will sound thick and bloated in the upper bass / lower mids when the speaker is placed in a typically reflective listening environment. Rather than fixing the problem correctly, some designers work around this issue by compromising the lower midrange linearity and obtaining a happy medium between both the anechoic and in room FR. On a "better" design, this will also include compensating for speaker placement in regards to side and rear walls. The net result is a speaker that is overly sensitive to placement and tends to send very different in various listening locations throughout the room with a narrow sweet spot.
Getting the lower midrange to sound right is IMO the hardest aspect of designing an accurate sounding speaker with linear in room FR in a variety of listening environments.
A larger dome mid with a WG can be versatile for use in a 3 way and is IMO the more suited type of driver to use when trying to achieve better lower midrange directivity and overall linearity compared to a cone driver by itself, which often exhibits less than acceptable linearity and directivity in the upper mids. Some people are ok with crossing a larger tweeter lower to a larger cone mid, but then you risk disrupting that delicate 2 to 5k area with a crossover point. I'd honestly rather have a crossover point on the higher end of that range than on the lower end. Trying to reproduce that range using only one driver is IMO the key to designing an overall more accurate sounding speaker.
Hello
Ok so you are talking below 300Hz? Am I understanding you correctly? That's room/schroder territory I would say placement is more of an issue than flat response. Unless you do a Roy Allison and design for specific room placement, It's hard to design for correct balance with the X/Room factor.
How would you propose to do that for a typical box design with unspecified room placement in any room?
I frankly wouldn't know where to begin?
Rob 🙂
Last edited:
😆 and more B6 as well as less alcohool my doctor said to me too !
@Robh3606 The speakers I usually build are very large, min 3 way to 4 way with multiple 12"+ LF drivers and 6 - 8" midbass. I attempt to control as much lower midrange as I can going off to the sides, usually with multiple drivers in unconventional arrangements ie circular, semi coax array. This is what i planned for with the 4 + 1 × 3" dome mids (M74A + SS D7608) described earlier in this thread. I'm planning to steer down to 500 hz with this arrangement, but not using a WG. Teton does something similar using shading. It can work well with large mids as well as tweeters.
Ok that makes sense thanks! So like the steerable LF arrays used in S/R. Food for thought if you have not seen this
Rob 🙂
Rob 🙂
@Robh3606 Ita similar to that principal, just with only a pair of drivers in one plane (both horizontally and vertically arranged) which is a bit of a compromise in control compared to that of several drivers.
The required phase offset to generate the lobes defining the directivity is created by offsetting the crossover phase response between the center midrange driver (M74A in my case) surrounded by 4 x D7608 mids encircling it in slightly offset, asymmetrical angles to diffuse the lobing across a wider frequency range. This reduces the deeper cancelation you'd get with perfect 90 deg circular arrangement.
The HP filter on the center mid brings in the lower midrange response at a softened 1st order rate above the net HP frequency chosen for the entire array of mids. The surrounding mids will be brought in at an LR2 HP slope, which delays the lower mids enough to narrow the dispersion angle at the beginning midrange passband and widening as the center driver catches up in phase. The surrounding drivers then become delayed by their dedicated LP filter, taking over and further rolling off their upper midrange at a steady rate before the group LP of the entire midrange section. There will be a slight amount of lobing associated with this type of d'appolito-ish arrangement, but it will have less combined ripple being asymmetrically arranged.
Hope all this makes sense. I'll still sim this to get a closer approximation of baffle spacing and filter application.
The required phase offset to generate the lobes defining the directivity is created by offsetting the crossover phase response between the center midrange driver (M74A in my case) surrounded by 4 x D7608 mids encircling it in slightly offset, asymmetrical angles to diffuse the lobing across a wider frequency range. This reduces the deeper cancelation you'd get with perfect 90 deg circular arrangement.
The HP filter on the center mid brings in the lower midrange response at a softened 1st order rate above the net HP frequency chosen for the entire array of mids. The surrounding mids will be brought in at an LR2 HP slope, which delays the lower mids enough to narrow the dispersion angle at the beginning midrange passband and widening as the center driver catches up in phase. The surrounding drivers then become delayed by their dedicated LP filter, taking over and further rolling off their upper midrange at a steady rate before the group LP of the entire midrange section. There will be a slight amount of lobing associated with this type of d'appolito-ish arrangement, but it will have less combined ripple being asymmetrically arranged.
Hope all this makes sense. I'll still sim this to get a closer approximation of baffle spacing and filter application.
Last edited: