Hi all.
if any are familiar with my posts, you might equate me to a measurement guy, and a FIR processing guy.
Rightfully so, cause i am, to the bone. Most of the time, i think i can make anything sound very close to anything.
I thought i should post a case where despite the fact the only change in the speaker was a pair of drivers that were swapped with pretty close equivalents,
and then tuned to exactly the same on-axis mag and phase traces,
.......and i do mean exactly the same mag and phase traces....
well, the sound is oh so different.
I've always attributed a sound difference when on-axis traces are identical, to off-axis differences, and how that creates a different power response interacting with the room.
But in the case i'm writing about, it was simply two different 10" drivers mounted in the same synergy horn, firing through the same ports.
And on-axis traces for the two sets were exactly the same....EXACTLY 🙂
Yet, what a diff...listening both indoors and out.
10 Fe200's, and 10PR300's , in the synergy.
The 10FE200's are loose and soft sounding compared to,
the 10 PR300's sounding tighter and righter.
I'm guessing the motor strength difference between the two is the thing, firing into a port that constitutes a 10:1 compression ratio.
Dunno....???
Can't be a power response difference it seems, given the same horn, same ports, same measurements....
Measurements were made at about 90dB @ 1m, so can't see it as a low level vs high level issue. Multiple measurements...pretty sure no error....
Ideas?
So the point here, of this post, is the 10FE200 and 10PR300 have identical looking transfer functions and impulses, pretty much textbook perfect, along with the whole speaker response.
And again, can there be a power response difference?
I don't often encounter a clear situation where it measures the same, but sounds so different ...
Thought in fairness to all my posts about measurements, FIR, and replication, i should throw this confusing experience out there.....
if any are familiar with my posts, you might equate me to a measurement guy, and a FIR processing guy.
Rightfully so, cause i am, to the bone. Most of the time, i think i can make anything sound very close to anything.
I thought i should post a case where despite the fact the only change in the speaker was a pair of drivers that were swapped with pretty close equivalents,
and then tuned to exactly the same on-axis mag and phase traces,
.......and i do mean exactly the same mag and phase traces....
well, the sound is oh so different.
I've always attributed a sound difference when on-axis traces are identical, to off-axis differences, and how that creates a different power response interacting with the room.
But in the case i'm writing about, it was simply two different 10" drivers mounted in the same synergy horn, firing through the same ports.
And on-axis traces for the two sets were exactly the same....EXACTLY 🙂
Yet, what a diff...listening both indoors and out.
10 Fe200's, and 10PR300's , in the synergy.
The 10FE200's are loose and soft sounding compared to,
the 10 PR300's sounding tighter and righter.
I'm guessing the motor strength difference between the two is the thing, firing into a port that constitutes a 10:1 compression ratio.
Dunno....???
Can't be a power response difference it seems, given the same horn, same ports, same measurements....
Measurements were made at about 90dB @ 1m, so can't see it as a low level vs high level issue. Multiple measurements...pretty sure no error....
Ideas?
So the point here, of this post, is the 10FE200 and 10PR300 have identical looking transfer functions and impulses, pretty much textbook perfect, along with the whole speaker response.
And again, can there be a power response difference?
I don't often encounter a clear situation where it measures the same, but sounds so different ...
Thought in fairness to all my posts about measurements, FIR, and replication, i should throw this confusing experience out there.....
Interesting, I can imagine the sound difference.
Did you tried to measure the harmonic distortions?
Anyway, a Qes 0.74 vs 0.29 is a huge difference.
Did you tried to measure the harmonic distortions?
Anyway, a Qes 0.74 vs 0.29 is a huge difference.
Wait.
You mean to say that what you measured ISN'T what you heard ?
No no no no noooo.
You obviously measured it wrong.
Oh no, next you'll be upsetting the cable and capacitor people that insist if you can't measure a difference then a difference doesn't exist.
🙂
There is a joke in quantum theory that says..
You altered the outcome by looking at it.
You mean to say that what you measured ISN'T what you heard ?
No no no no noooo.
You obviously measured it wrong.
Oh no, next you'll be upsetting the cable and capacitor people that insist if you can't measure a difference then a difference doesn't exist.
🙂
There is a joke in quantum theory that says..
You altered the outcome by looking at it.
Different harmonics ? Have you tried running single tones through the drivers bandpass range and looking at the harmonic spikes on an rta ?
Maybe you are EQing some of the different harmonics into the final response (a bit like when people EQ speakers indoors for flat at listening and are ending up with a flat response made up of direct sound plus the room reflections)
Rob. Here's a pic of my old Edgar midrange horn with a series of tones being played, and overlayed.
Maybe you are EQing some of the different harmonics into the final response (a bit like when people EQ speakers indoors for flat at listening and are ending up with a flat response made up of direct sound plus the room reflections)
Rob. Here's a pic of my old Edgar midrange horn with a series of tones being played, and overlayed.
Attachments
Popping your cherry into subjectivism hmm? 😀
There is far greater motion control/position over the VC with the MUCH lower Qe (higher Bl) with the 300's (particularly with lower excursions).
Harmonics might be it yeah, very different cone profile with these drivers => different bandpass filter with the enclosure? Also, not sure your measurement setup but could there be some variance in microphone positioning? I mean did you measure them the same with exact same mic and dut position? Have you measured the power response? this is interesting thread 🙂
Last edited:
Hi guys, thanks for the comments.
YSDR & RobWells, the only times i looked at harmonics were quick peeks with REW just to see if anything unusual popped out.
Both drivers 80-90dB 1meter looked close enough not to peer deeper. Pretty much around -60dBr throughout.
10pr300 below. In room, got some reflections in the 200-300Hz range for sure. (spl is correct)
ScottG, as I don't really understand T/S paramenters very well, I've just been looking at BL as an indication to drive thru the ports. As you and YDSR are saying, I guess Qes is a big factor in BL...
tmuikku & Charles, bandwidth for the 10" mids is 100 Hz to 500Hz
(LR72 dB/oct hpf & LR 96 dB/oct lpf)
I've found exact same mic and dut position aren't really as important as exact same tuning methodology....which is speaker closely in same spot and mic within same sphere of maybe a 4" radius.
And then I build a FIR file to where ever things exactly are for each case. Only real variable is what degree of correction is applied. Generally i use correction based on 1/6 oct smoothing. If a driver, or driver and horn, is very well behaved, I can go up to 1/12th based smoothing for more detailed correction.
If behavior sucks, i move down to 1/3.
I define behavior as how well corrections hold up, how well response stays smooth, off axis.
I didn't measure power response yet.
My thinking has been, given the same synergy horn with the same ports, same coax CD...
IOW, all things identical other than the mids...10FE200's vs 10PR300's...that power response would pretty much be the same, especially since the drivers are firing though a port.
Maybe, perhaps evidently not???
Anyway, the sound difference comes from the lower end of the mids range...100-150Hz i think.
The 10FE200's have kinda a cardboard drone line softness to them...for lack of better words. Heard inside and out. Thought for a while it might be the foamboard secondary flares.
I kept fiddling trying to pinpoint what continued to sound off....so finally sprung for a pair of the 10PR's to try.
(I'm loving the sound of syn7 with the 10PR's....second pair arrive today 😀)
Anyway again, this is the most clear case of sounds different, measures the same for all practical purposes, I've yet encountered...
I really don't pay much attention to minor differences in sound...i'm convinced our hearing changes as much as our vision does, but that we can't tell it like we can with vision.
IMO, if things sound close enough to need double blind to verify there's a real difference....well...they sound the SAME !!! (pragmatically anyway😉
YSDR & RobWells, the only times i looked at harmonics were quick peeks with REW just to see if anything unusual popped out.
Both drivers 80-90dB 1meter looked close enough not to peer deeper. Pretty much around -60dBr throughout.
10pr300 below. In room, got some reflections in the 200-300Hz range for sure. (spl is correct)
ScottG, as I don't really understand T/S paramenters very well, I've just been looking at BL as an indication to drive thru the ports. As you and YDSR are saying, I guess Qes is a big factor in BL...
tmuikku & Charles, bandwidth for the 10" mids is 100 Hz to 500Hz
(LR72 dB/oct hpf & LR 96 dB/oct lpf)
I've found exact same mic and dut position aren't really as important as exact same tuning methodology....which is speaker closely in same spot and mic within same sphere of maybe a 4" radius.
And then I build a FIR file to where ever things exactly are for each case. Only real variable is what degree of correction is applied. Generally i use correction based on 1/6 oct smoothing. If a driver, or driver and horn, is very well behaved, I can go up to 1/12th based smoothing for more detailed correction.
If behavior sucks, i move down to 1/3.
I define behavior as how well corrections hold up, how well response stays smooth, off axis.
I didn't measure power response yet.
My thinking has been, given the same synergy horn with the same ports, same coax CD...
IOW, all things identical other than the mids...10FE200's vs 10PR300's...that power response would pretty much be the same, especially since the drivers are firing though a port.
Maybe, perhaps evidently not???
Anyway, the sound difference comes from the lower end of the mids range...100-150Hz i think.
The 10FE200's have kinda a cardboard drone line softness to them...for lack of better words. Heard inside and out. Thought for a while it might be the foamboard secondary flares.
I kept fiddling trying to pinpoint what continued to sound off....so finally sprung for a pair of the 10PR's to try.
(I'm loving the sound of syn7 with the 10PR's....second pair arrive today 😀)
Anyway again, this is the most clear case of sounds different, measures the same for all practical purposes, I've yet encountered...
I really don't pay much attention to minor differences in sound...i'm convinced our hearing changes as much as our vision does, but that we can't tell it like we can with vision.
IMO, if things sound close enough to need double blind to verify there's a real difference....well...they sound the SAME !!! (pragmatically anyway😉
Attachments
Mark,
Sounds like you are hearing a difference in transient response between the two low mid drivers.
Perhaps that could be measured with square waves in the passband?
Looking at the cone profiles of the two drivers, it appears the lower BL 10FE200 would have a greater VTC (volume of throat chamber) than the 10PR300, making the acoustical band pass lower for it, more stored energy to "overcome" with less available BL.
Art
Attachments
Thx Art,
Very admittedly I don't much understand many of the finer points of acoustic construction.
I did play around with reducing volume under the cone to see how it effected the acoustic lpf, but since i was already getting response well past the upper frequency of 500 Hz that i wanted to cross to CD, i didn't spend much time experimenting there.
So right now, with both of the 10"s, they simply sit flush on baffle, with the same routed recesses for excursion. No apparent need to reduce volume on either to get to 500Hz.
I just keep thinking the 10FE200's don't have the torque (BL) to drive a 10:1 compression ratio.....like when a car is hung in too high a gear going uphill....
Even sounds like chugging...lol
One thing i need to ask...cause i don't get what square waves would show me... what/how are you saying i should test/look for?
Very admittedly I don't much understand many of the finer points of acoustic construction.
I did play around with reducing volume under the cone to see how it effected the acoustic lpf, but since i was already getting response well past the upper frequency of 500 Hz that i wanted to cross to CD, i didn't spend much time experimenting there.
So right now, with both of the 10"s, they simply sit flush on baffle, with the same routed recesses for excursion. No apparent need to reduce volume on either to get to 500Hz.
I just keep thinking the 10FE200's don't have the torque (BL) to drive a 10:1 compression ratio.....like when a car is hung in too high a gear going uphill....
Even sounds like chugging...lol
One thing i need to ask...cause i don't get what square waves would show me... what/how are you saying i should test/look for?
Maybe waterfall plots will show something going on?
The cheaper drivers definitely aren't made for horn-loading. Actually quantifying it is elusive... even experts have a hard time trying to explain it. It looks like that cone folds under pressure and the motor doesn't exert enough control either after a short sound event has passed. Still doesn't fit with the science. Who knows.
The cheaper drivers definitely aren't made for horn-loading. Actually quantifying it is elusive... even experts have a hard time trying to explain it. It looks like that cone folds under pressure and the motor doesn't exert enough control either after a short sound event has passed. Still doesn't fit with the science. Who knows.
A lot of open baffle guys experience something similar. In theory the under-damped response of weak motors should make up for the loss of bass.
In practice many people feel that stronger motors with electrical bass boost sound cleaner, sharper, etc. I happen to feel the same way but can't give a good reason for it. Woofers with QTS between .20 and .35 seems to be the sweet spot in most designs.
In practice many people feel that stronger motors with electrical bass boost sound cleaner, sharper, etc. I happen to feel the same way but can't give a good reason for it. Woofers with QTS between .20 and .35 seems to be the sweet spot in most designs.
Last edited:
Have you investigated the modal behaviour of the cavity. An 8" space can potentially support an in band cancellation. When equalising one compared to the other, what were the differences?
Beneath the waterfall plot also the step response might show the differences in temporal response between these drivers.
Regards
Charles
Regards
Charles
I was thinking "old school", something that could show the stored energy difference between the drivers/enclosure.
As phase_accurate and ErnieM suggest, a waterfall chart would show that, like the one below, showing decay time vs frequency.
I'd bet the weaker motor/larger cavity would have longer delay times.
Art
Attachments
Hi guys, again thx to all for continuing comments/thoughts...
here's some more direct measurements, kinda due to Parts Express sending one 8 ohm 10PR300 and one 4 ohm.
I had installed the 8 ohm in the bottom of the syn7, and when i went to do the top and saw the second one was 4 ohm...I'm like dammit...oh well, stuff happens..poor PE guys are probably running ragged...
Anyway, decided it made a great opportunity to do some meticulous head to head, toe to toe, measurements, one of each driver in the syn7.
Below are the 10FE200 and 10PR300 drivers, measured one at a time.
Just the mid driver measurements.....without the coax CD in play.
Each was individually tuned for flat mag and phase from 100Hz to 500Hz, with the mic centered on the horn about 1m away.
The non-used driver was shorted (which was clearly very necessary)
Nothing moved during the raw measurements, then the tuning, then the processed measurements.
I can't see any differences worth noting, other than maybe a bit more 2nd and less 3rd harmonic out of the 10PR300 ...????
They look damn near identical to me.....
But I'm not all that versed in reading waterfalls...anybody see anything?
The tuning process was the same I always use, and what I've been listening to, when hearing the clear difference in sound between the drivers...
Maybe two drivers running together change things to make this single measurement experiment a bit invalid...dunno....
Just know all the measurements tend to look pretty dang alike after tuning, one driver vs two, this vs that, ... dunno...
But the idea the sound diff is about transient response still smells right i think..
here's some more direct measurements, kinda due to Parts Express sending one 8 ohm 10PR300 and one 4 ohm.
I had installed the 8 ohm in the bottom of the syn7, and when i went to do the top and saw the second one was 4 ohm...I'm like dammit...oh well, stuff happens..poor PE guys are probably running ragged...
Anyway, decided it made a great opportunity to do some meticulous head to head, toe to toe, measurements, one of each driver in the syn7.
Below are the 10FE200 and 10PR300 drivers, measured one at a time.
Just the mid driver measurements.....without the coax CD in play.
Each was individually tuned for flat mag and phase from 100Hz to 500Hz, with the mic centered on the horn about 1m away.
The non-used driver was shorted (which was clearly very necessary)
Nothing moved during the raw measurements, then the tuning, then the processed measurements.
I can't see any differences worth noting, other than maybe a bit more 2nd and less 3rd harmonic out of the 10PR300 ...????
They look damn near identical to me.....
But I'm not all that versed in reading waterfalls...anybody see anything?
The tuning process was the same I always use, and what I've been listening to, when hearing the clear difference in sound between the drivers...
Maybe two drivers running together change things to make this single measurement experiment a bit invalid...dunno....
Just know all the measurements tend to look pretty dang alike after tuning, one driver vs two, this vs that, ... dunno...
But the idea the sound diff is about transient response still smells right i think..
Attachments
Last edited:
Looks the same....
The only other way I can think of is to compare the actual music waveform to the speaker output and overlay them to see where the differences are. Would have to be zoomed in to see the details.
I believe Barleywater posted measurements of his Linkwitz Pluto clones (with FIR) using this method.
Have you done IMD measurements?
The only other way I can think of is to compare the actual music waveform to the speaker output and overlay them to see where the differences are. Would have to be zoomed in to see the details.
I believe Barleywater posted measurements of his Linkwitz Pluto clones (with FIR) using this method.
Have you done IMD measurements?
Last edited:
I'm definitely not a 'waterfall expert' but to me it looks like there's almost a 'pulsing' as the sound decays (drops a bit, then blips up a bit, then drops a bit more) (Are these reflections if measured indoors ?)
My own measurements seem to show a smooth decay until the sound levels meet the background noise level. I can pm you a couple of pics of my waterfalls if you like rather than have them interfere with this thread.
Rob.
What do the plots look like before EQ/processing ?
My own measurements seem to show a smooth decay until the sound levels meet the background noise level. I can pm you a couple of pics of my waterfalls if you like rather than have them interfere with this thread.
Rob.
What do the plots look like before EQ/processing ?
Can you filter those waterfall plots? This shows a way too long window to see what's going on. Drastically shorten the rise time and the window. Like 0,1 rise time, 8 ms window and 10 ms time range to start...