Yes, sure John, i aleady noted that a softdome tweeter breaks up in bending waves say over 8kHz and there is no rise in distortion. I think i have now swallowed the linearity pill.
At least we may agree that it is extremely hard to construct a situation happening on the radiating area of a dynamic speakers in the univers known to man that is not perfectly predictable and coherent. Even when there is such a situation it is a second order effect that does not teach us anything new, at least not anything we can influence and is not worthwhile to spend considerable time on. OK, so what´s next Dr.Audios.
At least we may agree that it is extremely hard to construct a situation happening on the radiating area of a dynamic speakers in the univers known to man that is not perfectly predictable and coherent. Even when there is such a situation it is a second order effect that does not teach us anything new, at least not anything we can influence and is not worthwhile to spend considerable time on. OK, so what´s next Dr.Audios.
I did a quick and dirty EQ on axis,
Results measured at 0 deg and 30 deg
An externally hosted image should be here but it was not working when we last tested it.
Not sure why the link does no show the pic any more.
Attachments
Taiwanese is a dialect with no characters of it's own. Note sure how you're going to read that.
By anyway, Earl mentioned that EQ for one point in space might screw up other areas. I'm sure that can happen. My point is that if one understands what's going on, it will not happen. Additionally, if it were wrong at other locations, it can be heard. At least I can most of the time. But if we can kill two birds with one stone, might as well do it.
I think Earl's concern is that his wave guides have a hole in the on axis response, so it makes no sense to EQ on axis. Personally I'd be interested in seeing what those waveguides would measure like. There are so many design tradeoffs, and it really takes a long time to go through all of them and quantify them. Would be nice if the big guys can share some real data to help out.
Heh, there is a TV series on the story of Bruce Lee playing here. I'm finding some similarity in my exploration into audio technology.
By anyway, Earl mentioned that EQ for one point in space might screw up other areas. I'm sure that can happen. My point is that if one understands what's going on, it will not happen. Additionally, if it were wrong at other locations, it can be heard. At least I can most of the time. But if we can kill two birds with one stone, might as well do it.
I think Earl's concern is that his wave guides have a hole in the on axis response, so it makes no sense to EQ on axis. Personally I'd be interested in seeing what those waveguides would measure like. There are so many design tradeoffs, and it really takes a long time to go through all of them and quantify them. Would be nice if the big guys can share some real data to help out.
Heh, there is a TV series on the story of Bruce Lee playing here. I'm finding some similarity in my exploration into audio technology.
Last edited:
Just yesterday, I went through the process of applying driver modifications and EQ. The process involved a step by step application and measurement. After a certain level of change, listening is compared using two channels with different application. Then when no further measurable improvements can be seen, both channels were made the same and EQ applied. With each stage, measurable and audible results can be detected. EQ is very important, and as I have shown, with a dynamic driver, 30deg off axis can match on-axis quite well in terms of directivity. However, that is still not the whole story. I wait for the famous people to show more beef.
A few weeks ago, I had a nice steak with my daughter. The quality was good and the price was good. It reminded me of the $3.99 prime rib I had very often at Steak and Ale. Makes my mouth water just to think of it.
I wait for the famous people to show more beef.A few weeks ago, I had a nice steak with my daughter. The quality was good and the price was good. It reminded me of the $3.99 prime rib I had very often at Steak and Ale. Makes my mouth water just to think of it.
Last edited:
Hello all,
From all the ire and personal messages you would think I poured gasoline on the fire, talk about flames. When I posted regarding reverse acting controls I was at work , in a hurry and had my ME hat on. For you electronic EE types reverse acting control is mechanical / Control Theory speak for negative feedback.
You say what negative feedback? The signal or impulse stops the driver / voice coil continues to move and generates Back EMF which in my case is feed back to the cathode of my Single End Triode. The amplifier actively applies the brakes to the motion of the voice coil. BTW this may explain in part the reduced current at the voice coil at resonance.
For the history buffs Harry Nyquist is one the grand fathers of Control Theory look up Nyquist stability criterion.
Yes dampening is putting on the brakes. In this case reverse acting control is active dampening.
For the DSP Nyquist fans I am sure the UE is all about active negative feedback.
This mechanical geek is not so far off base after all.
DT
All just for fun!
From all the ire and personal messages you would think I poured gasoline on the fire, talk about flames. When I posted regarding reverse acting controls I was at work , in a hurry and had my ME hat on. For you electronic EE types reverse acting control is mechanical / Control Theory speak for negative feedback.
You say what negative feedback? The signal or impulse stops the driver / voice coil continues to move and generates Back EMF which in my case is feed back to the cathode of my Single End Triode. The amplifier actively applies the brakes to the motion of the voice coil. BTW this may explain in part the reduced current at the voice coil at resonance.
For the history buffs Harry Nyquist is one the grand fathers of Control Theory look up Nyquist stability criterion.
Yes dampening is putting on the brakes. In this case reverse acting control is active dampening.
For the DSP Nyquist fans I am sure the UE is all about active negative feedback.
This mechanical geek is not so far off base after all.
DT
All just for fun!
Think about it Earl, if you disturb one mass it will start out looking orderly, but as reflections from the boundaries enter the picture, the reflected waves start to interfere and the masses will all appear to be moving rather chaotically. You can add nonlinearity if you like, like nonlinear springs, it will still remain predictable, though perhaps unstable.
No George, its always going to happen wheather you understand it or not.
I've done that and I have posted most of it as well. You must not have been listening.
I show lots and lots of data of my stuff and even some for others - I don't understand what you concern is. That others do not show much data is not something that I can correct. I am trying to show more and more competitive data, but you have to have the competitive products to do that. Its not like anybody is going to supply me with the data.
Oh, so he's just saying something like that can happen. I guess I misunderstood his intention. Too used to his "you are wrong" type of remarks I guess. I think we all know EQ can go either way, solve many problems at once, or solve one problem but creating another. My point is cleaning up the CSD with EQ seems to have positive effect even if you only EQ at one point. But how much cleaning can be acheived depends on where you take that point.
Last edited:
There are actually two types of doubling. One is due to cone buckling which results in motion at 1/2 the driven frequency and the other is due to nonlinearities that cause the systems response to excitation to move towards infinity, i.e. no periodicity. The first type can occur instantaneously but the second takes time.
Most acoustical problems differ at various angular points, while virtually all mechanical and electrical resonances do not. The later is correctable with EQ, but the former is not. In Georges case it sounds like the mechanical problems dominate. This is why I like polar maps because these things are so clearly easy to see (but ONLY IF you DO NOT normalize them). The correctable problems are vertical ridges (which disappear if normalized just as EQ would do) while the uncorrectable problems are arcs.
fntn, now me being a half satisfied member of the linearity camp i must disappoint you again. I have put the idea aside to put the Scan wideband in a transmission line tube for now. I have your tapered tube idea firmly in my mind as an alternative and i will leave the linearity crowd as soon as convincing prove comes about that an unpredictable unlinearity effect exists that is audible.
What i found is that the Scan does not neead a big volume to work satisfactory down to 250Hz with resonable sensitivity. I simply put it in a small glas bowl and made a first attempt at a crossover. I think an LR4 is posible at 250Hz and a first order crossover is posible at 2.5kHz.
What i found is that the Scan does not neead a big volume to work satisfactory down to 250Hz with resonable sensitivity. I simply put it in a small glas bowl and made a first attempt at a crossover. I think an LR4 is posible at 250Hz and a first order crossover is posible at 2.5kHz.
I think the conclusion is the level of distortion is the key issue. To what level it can be identified as audible depends on design. However, the lower the level of distortion is audible at, indicates that other design aspects are closer to ideal.
I think Joachim may have mentioned it before. But it's my experience that if we compare distortion levels in relative to playback level, there are a few things that come into play in terms of it's audibility.
1. The absolute level of the distortion.
2. The frequency content of the distortion.
3. Human ear sensitivity curves.
4. Modulation of distortion induced harmonics with the content played.
I think Joachim may have mentioned it before. But it's my experience that if we compare distortion levels in relative to playback level, there are a few things that come into play in terms of it's audibility.
1. The absolute level of the distortion.
2. The frequency content of the distortion.
3. Human ear sensitivity curves.
4. Modulation of distortion induced harmonics with the content played.
Jamikl, i am happy that the members of your famiy survived.
Thorsten, you may know Martina Schöner. We listen together a lot and she has the oposite problem. When i play her ribbon tweeters or AMT´s she hears a disturbing noise in one of here ears. She definetly likes softdomes.
To all, the idea of the ZDL is to avoid as much as posible distructive interference so that when the speaker is flat on axis, the off axis performance is good as well. Digital EQ put into action on a ZDL will have beneficial effects on the off axis performance too.
Thorsten, you may know Martina Schöner. We listen together a lot and she has the oposite problem. When i play her ribbon tweeters or AMT´s she hears a disturbing noise in one of here ears. She definetly likes softdomes.
To all, the idea of the ZDL is to avoid as much as posible distructive interference so that when the speaker is flat on axis, the off axis performance is good as well. Digital EQ put into action on a ZDL will have beneficial effects on the off axis performance too.
Michal, i build my first Isobaric in 1980. I know about the problem that the volume between the drivers create. That is the reason i plan to fill the volume between the drivers with a solid material. Crossover will be at around 250Hz. In case of an LR4, amplitude will be down 6 dB already at this point. I do not claim that all problems are circumvented that way but i try to mimise the deficiencies.
Hello Joachim,
The stated design goals include the speaker becoming lost in the space and as wide a dispersion pattern as possible. Even with the drivers performing as perfect pistons the radiation patterns will narrow as the frequency rises.
What is your think about maximizing dispersion and crossover frequencies?
Others,
Crossover distortion was mentioned. In this context does that refer to the capacitors, inductors and the like in the filters or notch / crossover distortion in push pull amplifiers?
DT
All just for fun!
The stated design goals include the speaker becoming lost in the space and as wide a dispersion pattern as possible. Even with the drivers performing as perfect pistons the radiation patterns will narrow as the frequency rises.
What is your think about maximizing dispersion and crossover frequencies?
Others,
Crossover distortion was mentioned. In this context does that refer to the capacitors, inductors and the like in the filters or notch / crossover distortion in push pull amplifiers?
DT
All just for fun!
I recently received a pair from zaph for testing. Eventually I'll dissect one of them. They look exactly like the dome in the Vifa NE19VTS that I have on-hand. Aside from the lower Fs shown by Vifa (I have not measured the OX yet), the primary difference seems to be faceplate (or lack thereof) and chamber size. The lower Fs is logical. The NE19 and I suspect the OX have eight openings underneath the surround in the fashion of the XT drivers for venting into the chamber.I got me a pair of the ox tweeters and I'll be trying to keep Joachim honest.
The FR curves shown by Vifa also show the influence of the two, small plastic "protectors" on the NE19. I cut these off. You'd be surprised at the difference. I've done this with other small and apparently cheap tweeters such as one would find on Sony or other volume brands. When these are cut off or the removed from the casing and mounted flush, even the cheap ones are at times reasonably flat up to 20K without these.
The OX20 looks to be a very good tweeter if mounted well. It may be a while, but when I open one up I'll post comparison photos.
Dave
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.