Hi 454Casull,
Of course the comp driver has much less distortion at high levels (and it gets to those high levels through less compression- a bit of an oxymoron
)
It's the low level signals that I am concerned about. Not just the Esotar, but the good Scanspeak, Seas e.t.c. dome tweeters produce that shimmer and DECAY on a stuck jazz cymbal, the air of an orchestral triangle e.t.c that are missing in action on a DE250.
I go to concerts. I know what the real thing sounds like. I'm talking about maybe 1/4 sec, 1/2 sec of triangle (or something) that just isn't there on the DE250 (but it is there on my recordings).
Norris,
Thanks for the tip. I'll do a bit of research
David
Of course the comp driver has much less distortion at high levels (and it gets to those high levels through less compression- a bit of an oxymoron
)It's the low level signals that I am concerned about. Not just the Esotar, but the good Scanspeak, Seas e.t.c. dome tweeters produce that shimmer and DECAY on a stuck jazz cymbal, the air of an orchestral triangle e.t.c that are missing in action on a DE250.
I go to concerts. I know what the real thing sounds like. I'm talking about maybe 1/4 sec, 1/2 sec of triangle (or something) that just isn't there on the DE250 (but it is there on my recordings).
Norris,
Thanks for the tip. I'll do a bit of research
David
"I've tried the waveguide (horn?) with and without the open-cell foam. Better without the foam IMO in every way."
No "horn sound" without the foam?
That would be good news, saving me of cutting up layers of the foam I got for the XT1086's.
I have BMS 4552ND to use with them. They're highly regarded, but I don't know if they're better than the DE250.
No "horn sound" without the foam?
That would be good news, saving me of cutting up layers of the foam I got for the XT1086's.
I have BMS 4552ND to use with them. They're highly regarded, but I don't know if they're better than the DE250.
"If I can hear fine detail on the Esotar and not on the DE 250, which is the more accurate (at least at lowish levels)? i.e. which is really distorting?"
If the fine detail includes better spatial presentation, it is probably an improvement. Some apparent detail by itself may be due to midrange frequency anomalies.
If the fine detail includes better spatial presentation, it is probably an improvement. Some apparent detail by itself may be due to midrange frequency anomalies.
Hi --
I don't have experience with compressiondrivers and horns, but want to mention that according to others, who tried "waveguides" on dome-speakers(!), the "shimmer" you mention tends to "get lost" WITH the guide ... But it's supposed to be not a real "loss" but rather the absence/reduction of room-reflections that accounts for the difference.
So in reality there IS NO "shimmer", just "detail" added by the room ... (?)
But I have to admit, I don't know if that's really all there is to this "phenomenon" of "micro-detail".
Regards
Bernd
I don't have experience with compressiondrivers and horns, but want to mention that according to others, who tried "waveguides" on dome-speakers(!), the "shimmer" you mention tends to "get lost" WITH the guide ... But it's supposed to be not a real "loss" but rather the absence/reduction of room-reflections that accounts for the difference.
So in reality there IS NO "shimmer", just "detail" added by the room ... (?)
But I have to admit, I don't know if that's really all there is to this "phenomenon" of "micro-detail".
Regards
Bernd
Hi noah,
I look forward very eagerly to your results. (I'm working on a direct comparison between the DE250 and the 18sound NSD1095N!).
With the foam they sounded like they had a sock in them- the dynamics sounded just that bit compressed (supposed to be no attenuation, and I've got to admit that I didn't measure any either).
thoriated,
That's a good way of putting it- better spatial presentation.
I look forward very eagerly to your results. (I'm working on a direct comparison between the DE250 and the 18sound NSD1095N!).
With the foam they sounded like they had a sock in them- the dynamics sounded just that bit compressed (supposed to be no attenuation, and I've got to admit that I didn't measure any either).
thoriated,
That's a good way of putting it- better spatial presentation.
D OB G said:
I don't think that another driver is going to solve your problem. The differences in compression drivers are not that great, especially compared to the difference between a direct radiator and a CD waveguide.
Make no mistake about it there is a "sizzling" quality about a direct radiator that is beaming HFs right into your ear. If you like this sound then stay away from CD waveguides and try an exponential horn or something like that which has a much narrower HF directivity. Personally, when I listen to dome tweeter now, this same "sizzle" drives me up a wall after a few minutes (many others have made this same comment). Initially it sounds like "detail" later on it sounds like ...
Also, never use a DEQ to EQ a waveguide because it can screw it up badly. A waveguide should be EQ for flat power response not flat axial response which a single point microphone can't do.
From your discussion I don't think that you are setting up the waveguide/horn properly which is the problem, not the driver. Direct radiators are easy to set up and so they are more difficult to screw up. Its very easy to mess up a waveguide.
You opinion of foam, for instance, is completely the opposite of anyone else that I have talked to who has tried it. I would strongly suggest NOT using the DEQ and EQing in a more conventional way.
You re-EQ'd each trial?
Basically, I don't EQ to any single axis, but to the average of the forward sound. The whole concept of EQing precisely to a single point leaves me cold. Does the EQ gate the reference? and if so what is the window? There are so many ways for an "automatic" process to screw up that I don't think that I could enumerate them all.
Your experience with the foam appears to be unique and I might be concerned about that if I was in your position.
It could also be just what I said, that you are looking for that pin-point sound that a direct radiator has at HFs. People have come to expect it and when its not there, as with a well done CD device, the sound is a little strange at first. This has been a big hump to get over in auditioning my speakers. This is why I prefer to do auditions here - I don't let people leave in 60 seconds, they have to stay an hour or more. By this time they have acclamated to the different sound and find it far more appealing.
But the bottom line here is that changing drivers is not likely to change your perceptions. I've measured dozens of drivers and they are all very very similar if they are of the same size throat and diaphragm. In fact many of these devices you can interchange the diaphragms, which are more than likely all made in the same place (ATON in Thailand).
You might try the De500 as the titanium diaphragm will go a little higher. I don't expect that you'll find much difference, but you can try it. As to other manufacturers, they are all very competitive with each other. I once tested a Chinese knock-off of the DE250 - the diaphragms were interchangable. It was $17, did not quite measure as good as the DE250, but I'm guessing that in the system you wouldn't know the difference.
Basically, I don't EQ to any single axis, but to the average of the forward sound. The whole concept of EQing precisely to a single point leaves me cold. Does the EQ gate the reference? and if so what is the window? There are so many ways for an "automatic" process to screw up that I don't think that I could enumerate them all.
Your experience with the foam appears to be unique and I might be concerned about that if I was in your position.
It could also be just what I said, that you are looking for that pin-point sound that a direct radiator has at HFs. People have come to expect it and when its not there, as with a well done CD device, the sound is a little strange at first. This has been a big hump to get over in auditioning my speakers. This is why I prefer to do auditions here - I don't let people leave in 60 seconds, they have to stay an hour or more. By this time they have acclamated to the different sound and find it far more appealing.
But the bottom line here is that changing drivers is not likely to change your perceptions. I've measured dozens of drivers and they are all very very similar if they are of the same size throat and diaphragm. In fact many of these devices you can interchange the diaphragms, which are more than likely all made in the same place (ATON in Thailand).
You might try the De500 as the titanium diaphragm will go a little higher. I don't expect that you'll find much difference, but you can try it. As to other manufacturers, they are all very competitive with each other. I once tested a Chinese knock-off of the DE250 - the diaphragms were interchangable. It was $17, did not quite measure as good as the DE250, but I'm guessing that in the system you wouldn't know the difference.
Earl, how far off axis do you measure to?
My technique of installing the foam would inevitably have been different with the 18sound XT1086. I cut the pieces to be perpendicular to the surface i.e. they were "laminated" from left to right, but they weren't glued together, just to the horn. Would these interfaces have caused a problem?
As far as averaging is concerned, I can easily get Praxis out, average over the range of angles you recommend, and then apply that to the DEQX response.
Do we both want a flat on-listening axis response? Because I find it hard to understand how that will be achieved by averaging reponses that will fall more at higher frequencies the further off axis we are?
My technique of installing the foam would inevitably have been different with the 18sound XT1086. I cut the pieces to be perpendicular to the surface i.e. they were "laminated" from left to right, but they weren't glued together, just to the horn. Would these interfaces have caused a problem?
As far as averaging is concerned, I can easily get Praxis out, average over the range of angles you recommend, and then apply that to the DEQX response.
Do we both want a flat on-listening axis response? Because I find it hard to understand how that will be achieved by averaging reponses that will fall more at higher frequencies the further off axis we are?
D OB G said:
But thats the point, if it CD then the HF response doesn't fall off axis. If it does then its NOT CD. My waveguides don't fall off axis at the HF. The response falls, but uniformly with frequency. My designs achieve a +-2 dB response from LFs -> >10 kHz from about 10 degrees to 25-30 degrees. Its important to find that EQ that achieves this as flattening any one angle will usually make the broader problem worse. There is usually some problem right on axis of the speaker that cause a problem with flatness at this location.
I must still be missing the point, because that's where I measure (produce) a flat response- somewhere between about 20 deg and 25 deg, NOT on the speaker axis, but rather the listening axis.
I did a rudimentary test. I placed an Esotar in one waveguide and EQd (on tweeter axis- and listened on axis)- obviously nasty diffraction effects - but ignore that for a moment, and I compared it to an unfilled XT1086 EQd at about 20 deg off axis- and listened to on that axis).
The point is that there was more detail through the Esotar (I must mention this was above 6 kHz- an OB PHY-HP below). Yes, I know the polar responses won't match, but what is revealing is DICTION (amongst other fine detail).
It's possible to distinguish difficult to make out words on the Esotar that aren't quite there on the waveguide. Maybe all I'm demonstrating is that it's the XT1086 that isn't good enough. But the detail difference is still there (even under sub-optimal conditions for the Esotar).
That said, I admit there are other characteristics of CD that best the Esotar- dynamics in particular.
I did a rudimentary test. I placed an Esotar in one waveguide and EQd (on tweeter axis- and listened on axis)- obviously nasty diffraction effects - but ignore that for a moment, and I compared it to an unfilled XT1086 EQd at about 20 deg off axis- and listened to on that axis).
The point is that there was more detail through the Esotar (I must mention this was above 6 kHz- an OB PHY-HP below). Yes, I know the polar responses won't match, but what is revealing is DICTION (amongst other fine detail).
It's possible to distinguish difficult to make out words on the Esotar that aren't quite there on the waveguide. Maybe all I'm demonstrating is that it's the XT1086 that isn't good enough. But the detail difference is still there (even under sub-optimal conditions for the Esotar).
That said, I admit there are other characteristics of CD that best the Esotar- dynamics in particular.
I don't think that there is any way to explain your subjective perceptions, or at any rate, I'm not going to try, but if there some objective data to deal with then maybe something could be resolved.
EQing along a single axis, any axis, is just not the way I do things. The FFT based convolvers do such precise things that its more than likely going to mess up all other directions trying to get that one direction exact. Sound in a room is simply not a single direction thing. Its an integration of many directions of reflections and reverberation. I've only listened to one of those units a few times but never found it attractive.
EQing along a single axis, any axis, is just not the way I do things. The FFT based convolvers do such precise things that its more than likely going to mess up all other directions trying to get that one direction exact. Sound in a room is simply not a single direction thing. Its an integration of many directions of reflections and reverberation. I've only listened to one of those units a few times but never found it attractive.
Esotar tweeters are said to be among the very best there is, and despite of what is said here about that, it may simply be what you experience
But reality could still be the fact that you may not be getting the most out of your CD, as it is at the present
Better crossover have been mentioned several times, as is a better waveguide
Cant remember whether a supertweeter have been mentioned, but some do think its vital in getting the last bit of information
Question is if any of it will ever be enough, to better the obvious special qualities of your Esotar
Maybe an unavoidable design tradeoff
Sorry, only thoughts and not much help, but its obvious that you are not among those who are satisfied with doing 15khz
But reality could still be the fact that you may not be getting the most out of your CD, as it is at the present
Better crossover have been mentioned several times, as is a better waveguide
Cant remember whether a supertweeter have been mentioned, but some do think its vital in getting the last bit of information
Question is if any of it will ever be enough, to better the obvious special qualities of your Esotar
Maybe an unavoidable design tradeoff
Sorry, only thoughts and not much help, but its obvious that you are not among those who are satisfied with doing 15khz
Sorry, a small point
What worries me about DCs, which I also hope to be using in my next design, is the brickfall shutdown when it reaches its limit
I have learned from my master, that such should always be avoided
To me it seems a supertweeter would be the only solution to that issue, despite the difficulties in implementing it properly, if at all possible...well, I do hope it is
What worries me about DCs, which I also hope to be using in my next design, is the brickfall shutdown when it reaches its limit
I have learned from my master, that such should always be avoided
To me it seems a supertweeter would be the only solution to that issue, despite the difficulties in implementing it properly, if at all possible...well, I do hope it is
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.