Just after a long long Linesourse diy build, find out that some drivers must have to much undertones. "in them selfes"
Lika a Piano, but wrong....Hard to explain.
The measurement are really great, but high SPL not, something isent right!
It´s like drivers have to long undertones, that makes the sound "blurry/long" and makes sound unpleasant to human ears.
Driver MMS are 6,5 g and driver are peerless_fsl_0512r01_08
Still fantastic on voice-range, like "in a center"
Lika a Piano, but wrong....Hard to explain.
The measurement are really great, but high SPL not, something isent right!
It´s like drivers have to long undertones, that makes the sound "blurry/long" and makes sound unpleasant to human ears.
Driver MMS are 6,5 g and driver are peerless_fsl_0512r01_08
Still fantastic on voice-range, like "in a center"
Could be the XO alignment , when it crosses both drivers are playing and this introduces a lot of interferences and problems, in my 2dn order xo, I did listen tonight , it xo at 1800hz and from 1300 to 2300 it has this undertones, it is not as clear, it sounds like 2 sounds , 2 sources, but below it is clear.
Understand what you mean 👍
I have 8 peerless_fsl_0512r01_08 and a Audax tw025a28 in a WG per side.
4 peerless_fsl_0512r01_08 ported (tuned at 110 hz) from 100 to 1700 hz 2 order and 4 fsl_0512r01_08 with only LP 2 order at 3 K
Somehow the undertones from the fsl_0512r01_08 make sound "to long" at tones, like a piano but unneutral!
I have 8 peerless_fsl_0512r01_08 and a Audax tw025a28 in a WG per side.
4 peerless_fsl_0512r01_08 ported (tuned at 110 hz) from 100 to 1700 hz 2 order and 4 fsl_0512r01_08 with only LP 2 order at 3 K
Somehow the undertones from the fsl_0512r01_08 make sound "to long" at tones, like a piano but unneutral!
Attachments
Will try out 3 order tomorrow at the 4 midranges 👍 (at the woofers i mean)
Usher have allway suffers from their midranges, but beautyful cabinettes top notch!
Usher have allway suffers from their midranges, but beautyful cabinettes top notch!
its made with carbon fiber
a veiled and congested sound, but with some great bass details texture under 150hz
a veiled and congested sound, but with some great bass details texture under 150hz
I disconnected one of the woofer, and the port. I only use 1 woofer.
Doesn't play like the Harbeth (harbeths are wayyy harder to satisfy with amplification)
Doesn't play like the Harbeth (harbeths are wayyy harder to satisfy with amplification)
I am building a 3 way scan speak, I hope to get close or better than my objective.
My amps are kt120 (soon to be finalized), right now using a prototype kt88, and a hugely upgraded sugden amp. Becoming fussy 🙂
My amps are kt120 (soon to be finalized), right now using a prototype kt88, and a hugely upgraded sugden amp. Becoming fussy 🙂
Here is measurement from listeningpoint in my sofa.
Obviously measurements do not show how sound is experienced!
Shifted 15 times yesterday between
1) 3 way SS 9300 and SS 18M Relevator and 15 inch Dayton RRS390ho,
2) to linesource 2,5 way with 8 Peerlees fls 0512, and Audax TW025A28 and 15 inch Dayton RRS390ho
Nr 1 wins every time even if SPL capacity is enormous in Nr 2.
Calmer sound despite the same SPL, more punch, quiet between notes
Frustrated!
Obviously measurements do not show how sound is experienced!
Shifted 15 times yesterday between
1) 3 way SS 9300 and SS 18M Relevator and 15 inch Dayton RRS390ho,
2) to linesource 2,5 way with 8 Peerlees fls 0512, and Audax TW025A28 and 15 inch Dayton RRS390ho
Nr 1 wins every time even if SPL capacity is enormous in Nr 2.
Calmer sound despite the same SPL, more punch, quiet between notes
Frustrated!
Attachments
Cone noise or poor DDR could cause that.
Often heard as “missing” information, lack of definition, an indistness. Loider pushes everything closer to the edge. XO often contributes to loss of information.
As well box alignment could be contributing, as levels change so do T/S (with mnost drivers) meaning the box tuning may no longer be acting as designed.
And with louder, is amplifier clipping or otherwise straining?
dave
Last edited:
A couple more comments.
In a line array the combing increases information loss.
Crossing a LA to a single waveguided tweeter is automatically difficult with lotsn of isses.
dave
In a line array the combing increases information loss.
Crossing a LA to a single waveguided tweeter is automatically difficult with lotsn of isses.
dave
Show the distortion for that measurement? And while at it, hit the Spectrogram as well.
//
//
Your three way has some rather nicely specified drivers in individual cabinets, consequently I imagine their behaviour, and self noise and decay performance should be better.
Maybe the Peerless driver in isolation is fine as a single midrange in a conventional three way, but with 4 of them for bass and 4 for mid duties you maybe hearing their combined signature. Their cones maybe lightweight but multiply that by 8 and you can see that the cabinet will see some energy from these drivers. Also you may be getting a bit more reflected energy (floor and Ceiling) from the line array. Caveat, I know nothing about line array rule of thumbs and do's and don'ts.
Taken from TNT's response above distortion and Spectrogram/waterfall type measurements may be useful to help break this down and identify something.
Maybe measurements at 85 and 95dB for us mere mortals, and with your ear protectors on, one at your preferred listening level.
If you are concentrating on the line array, for your own interest, it maybe splitting out the two sections and having measurement's for Bass, and mid sections and . That's a fair few measurement's.
Maybe the Peerless driver in isolation is fine as a single midrange in a conventional three way, but with 4 of them for bass and 4 for mid duties you maybe hearing their combined signature. Their cones maybe lightweight but multiply that by 8 and you can see that the cabinet will see some energy from these drivers. Also you may be getting a bit more reflected energy (floor and Ceiling) from the line array. Caveat, I know nothing about line array rule of thumbs and do's and don'ts.
Taken from TNT's response above distortion and Spectrogram/waterfall type measurements may be useful to help break this down and identify something.
Maybe measurements at 85 and 95dB for us mere mortals, and with your ear protectors on, one at your preferred listening level.
If you are concentrating on the line array, for your own interest, it maybe splitting out the two sections and having measurement's for Bass, and mid sections and . That's a fair few measurement's.
Yeah, the larger the driver, the further away the listening distance ideally needs to be to sound as one for a given BW.
Think your on to something!
"poor DDR could", is that fabric in dustcup?
So hard to explain sound, but if you think of a piano the undertones are many and also long.
Take ½ % of that, is what i belive i hear.
The "calm" in the sound isent there, its like the cone are on steriods and thinner than "smoke"
Maby sound are a little little "analytic/metallic", or in that direction.
No using a big Krell amp
Can you do anything about combing?
Was also in the early stages thinking about using 4 smaller tweeters.
Last edited by a moderator:
You are right Ray, the combo SS 18m and 9300 is hard to beat and im impressed by them every time.
Absolutly!
A litte flaw get bigger with 16 units.
Here is also a "problem-point" for me.
The speakers will be mounted on the walls, on each side of my telly.
Mounted so the tweeter sits 3/4 up from tv ending.
Today with xover tests, i have the speakers 25-30 cm to high (but have no alternatives)
So maby much more reflections from the backwall now when testing.
I will try gather some strength, and do more measurement.
Thinking also i try out 18 dB LP at the woofer section, crossing into midrange section.
Cone self-noise certainly impacts DDR, but is not the whole story. The simpliest way to think about DDR is how well the driver (or other device) reproduces very mall detail — even in the face of much louder stuff playing,
An Analogy would be that (good) FR would be a smooth surface on a body of water, and DDR the depth of the water. Deep (and flatish) is preferred.
Tip: I find a thin coat of modPodge on paper cones decreases their cone self-noise and thus improve DDR.
dave