What I am saying it appears you have over tightened the back screws on the drivers (or the factory has) or the Be diaphragms are not performing like all the BE diaphragms I have used do. They should be wider bandwidth then the aluminum phrams. The Radian drivers I have act like that when I muscled too much on the back screws. You really need to measure them and adjust the tightness of the clamp to get them working properly.
My system is designed for my expectations and room
I have not studied the microstructure of Be, because a friend found it a bit the the brittle side and had a few damaged under certain circumstances. Basically you want material to have fine and uniform structure otherwise they will cause some sound that impresses you in the beginning, but long term live level listening can be annoying. I fact this is important in almost any kind of passive and active devices.
Looking at the graphs g3dahl posted, I would try active equalizing the top end just to see how it will sound. If there is no breakup in the diaphragm, it should sound pretty good.
Looking at the graphs g3dahl posted, I would try active equalizing the top end just to see how it will sound. If there is no breakup in the diaphragm, it should sound pretty good.
Been thinking about this post a bit; hoped responses would show a little more pulse in the audience.
With Vandersteens, was impression that they needed turning up/down?
Is it your perception that at same relative levels that Vandersteens are easier to talk over than Summas?
Perhaps this is more about room response. Direct response of Vandersteens may be such that sound is perceived as 'correct', and is easy to pay attention to, but also not intrusive in sense of cocktail party effect of being able to tune in and out of multiple speech streams.
Summas likely have more uniform response within room providing greater sense of needing to overcome background.
As a cafeteria fills, sound of ones own voice seems to diminish, and tendency is to compensate by speaking louder. Damping of own ears to own voice comes into effect, and remains so do to overall sound intensity in room.
Have you ever noticed with Summas, that once aware of elevated conversation level, it is possible to reduce speaking level instead of speakers, and still retain conversational ability?
Hearing is actively adaptive. Listening to simple two way direct radiator system referenced for close listening, moving from sweet spot to peripheral room location makes both level drop, and timbrel perception awareness change. If listening persists for a few minutes from peripheral position, sound perception often changes to accept new timbrel balance as norm, and initial response on returning to originally referenced sweet spot results in temporary perceptions of sound being too bright.
It mostly is concerned with higher levels of non-linear distortion from the Vandersteens at high spl's.
That added distortion makes it sound *louder*, and when that happens you don't tend to increase the "volume". Alternativly, when you have really low levels of non-linear distortion (as in the Summa's) you are more likely to increase the "volume" - a bit less aware of how loud it really is. Once you do that you have to "shout".
This may occur for a conventional level of system optimisation, but is not the case when everything is working correctly. Both for two ways, and highly conventional single full-range drivers, a properly working system produces a sound which subjectively does not change in volume, or timbre, no matter where you move in the room.
Last edited:
Distortion perceived only as loudness is non factor. If it is factor, then measurements of two speakers at comparable levels at which raised voice situation occurs is worthy of investigation via auralization with headphones with normalization of response levels.
Power response in room above 3kHz is likely factor worthy of positive or negative identification through appropriately designed and executed experiment.
In quiet room, use of earplugs tends to raise level at which person speaks. Source with uniform power response in room leads people to naturally raise voices, even when not needed in directed conversation in both terms of speaking direction and listening focus. Human voice content has very non uniform radiation.
That's why I used the term "subjective" - I wasn't referring to the fact that "technically" the FR was highly constant over a certain coverage area, rather that the ear/brain was receiving sound of sufficiently good quality for the mind to automatically balance the sound, without conscious effort. I've heard plenty of horn sound over the years, I sure most of it had pretty good FR, but 98% of it sounded pretty, or totally, crappy because it was riddled with distortion - the 'good' directivity couldn't start to compensate for the other problems ...
The typical non-linear distortion most audio systems project, the very thing Lynn was indirectly referring to a couple of pages ago, http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-971.html#post3701138.
Most PA systems are much worse, largely because the driving electronics work so poorly as a system, and, they're badly set up. I have heard a very small number of outstanding counter-examples, which only serves to demonstrate that the technology is available to do it right, and has been for many years, but overall it appears that very few sound engineers are capable of 'hearing' what is right, and what is not right ...
Most PA systems are much worse, largely because the driving electronics work so poorly as a system, and, they're badly set up. I have heard a very small number of outstanding counter-examples, which only serves to demonstrate that the technology is available to do it right, and has been for many years, but overall it appears that very few sound engineers are capable of 'hearing' what is right, and what is not right ...
Indeed.
It happen at times that certain sound setups, or certain components, have a very impressing sound at the beginning, but are annoying on long term listening.
This is a general observation, I'm not referring here to the speakers in question, since I haven't heard them.
I liked the look of the frequency response plot from the Radian Be, Gary . It appears to have a graceful roll-off at the top end, not unlike the 288-C . I'm wondering if there's something fundamental that I don't like about the area from 8kHz upwards with the 288H , as it heads into break-up. The on-axis sound unfiltered/unmodified definitely has a unpleasant tendency to shriek on some material . I've been looking around (with no success) for a 288 bug-screen to see if it has any useful damping effect at the top-end . I know Romy and CV have both found the Vitavox S2 is very dependant upon this piece to sound correct . In the short-term I may employ the 'tissue-paper mod' in the throat to see if there's some milage in this approach . If I could get a design drawing of the bug-screen, I could get a local machinist to make one up for me very quickly, actually .
The other mod I'm going to try too is a small air-cored coil ( about 100 turns ) to roll-off the top, 1st-order . I have tried the Aurum G3 ribbon on top there, and despite the lower sensitivity ( on paper ) it actually adds something useful to the treble when run at the same voltage level as the 288H .
A few months ago I listened to three compression drivers that I have on the same horn. I noticed that the largest and most expensive compression driver that I have was noticeably 'smoother.'
But I think that the 'smooth' sound isn't just the diaphragm; it's also the motor.
For instance, if you take the same diaphragm and put it on two different compression drivers, the one with the larger motor will generally have a rolloff on the low end of it's bandwidth that's more gradual. This is because the qts of the system will fall as you increase the motor strength.
If my hypothesis is correct, then one might be able to make a compression driver sound 'smoother' by filtering it carefully. Something like minidsp would be handy for this.
I took the bug screens out of my 288s. Could not hear or measure any difference. Surprising, really.
I think you're on to something. There's a noticeable improvement in the Neo version of the Radian driver, even if both are using aluminum diaphragms. The Neo version sounds more effortless, smoother, less grain, etc. It has to be the magnet and the pole-piece design, since the diaphragms are very similar.
I also notice the same thing between ceramic and Alnico magnet woofers, although Alnico has more "sparkle" and tonal vividness to the sound (while having pretty much the same FR and impulse response).
Field-coil "magic"? I haven't heard it; what I have heard sounds like more BL in the gap; field-coils allow you to turn up the voltage until the gap is magnetically saturated, and that certainly changes the sound (usually better).
These differences may be washed out if the system has a Class AB or Class D high-feedback transistor amplifier. Speaking only for myself, I find amplifiers of this type grainier, more opaque, and tonally washed-out, and not musically "accurate", which is why I only use them for HT applications.
Last edited:
Hi Lyn,
How about Class A, low GNFB (Global Negative Feedback) or no GNFB, SS amplifiers?
I remember you wrote few days ago you don't use them, but how do they affect the sound?
The Pass Class A amplifiers I've had in my system were not to my taste; the sound on the Ariels was cold, flat-sounding and grainy. I've tried about 20 different transistor amps on the Ariels, and only liked the sound of the out-of-production LNPA-150.
There are probably are good-sounding transistor amps out there, but rather than find the one out of a hundred that matches what I like, I switched to direct-heated triode amps in 1995, and have stayed there since. The output tubes are good for 2000 to 3000 hours, which is plenty of listening time, and the small-signal tubes are good for 5000 hours or more.
I have no idea which speakers sound good with moderate power, Class A transistor amps; in market terms, that's a niche of a niche, much smaller than vacuum-tube amps, which are a traditional favorite in Asia, Germany, Italy, the UK, and some parts of the USA.