soongsc said:
Hi,
The Faraday ring not only alters the inductance. It lowers harmonic and intermodulation distortion. It makes a better driver.
By the way, what about transformers or not transformers. And if none, which? Wasn't (sorry - Isn't) ARIEL the arcangle of light? Due to some wonderous miracle we came to the topic of electrons. The new ARIEL has to feature a plasma tweeter, won't You see?
kismet
There are a lot of differencies between magnet materials even from different ferro blends or different ALNICO blends.
So its kinda oversimplification to tell about on or the other.
*But* what I am interested in is the mere conductivity .
Scroll down to the end of the list and check out "Electrical Resistivity (Ωm)" for ALNICO and Ceramic:
http://www.intemag.com/magnetic_properties.html
This for sure *is* different - hence my bridge to Bud's findings in the light of my own experience.
Michael
So its kinda oversimplification to tell about on or the other.
*But* what I am interested in is the mere conductivity .
Scroll down to the end of the list and check out "Electrical Resistivity (Ωm)" for ALNICO and Ceramic:
http://www.intemag.com/magnetic_properties.html
This for sure *is* different - hence my bridge to Bud's findings in the light of my own experience.
Michael
dlr said:Reminds me of our early TVs with rabbit ears and distant stations fading in and out. Standing at the right spot or holding the antenna helped. Must be some esoteric non-scientific explanation for that. Maybe I was the electron pool.
Perhaps our bodies might also be antennas.
Here's a fun thing to try...
Unlock and lock your car with the remote and keep moving away from the car until the remote is out of range and the car won't lock or unlock.
Now move a further 10 metres away, hold the remote against your head and click the button...
Cheers,
Alex
To tell you the truth, I have not enough data to confirm how it effects distortion. From what I understand is that since it creats a better motor, the effects are in the piston operating range of the driver. I have seen lowered Q due to the Faraday ring in one application, but really need to study where the more significant distortion comes from throughout the frequency range. Source of distortion may not always be from the motor.xpert said:
Sure anything can act as areal.
Basically you only need *distantace* to form an areal.
If we look at a speaker cable the singnal and its return should be close enough (= no distance for frequencies of interest) to each other to not act as antenna.
This should be true for all amps that are not prone to selectively demodulate the tiny *common mode* signals from its output and send them back to the speaker amplified so that you can listen to BBC news.
Things get different when we consider the speaker's coil to be kind of rabbit's ear like the wound areals of AM receivers.
Here we get a *differential signal* that is fed back to the amp's output.
If we have an all conductive shielding around the VC - as is the case for ALNICO - we hardly get that *differential signals*.
The assumption here would be that some amps may be more prone to the presence of tiny differential signals at its out than to tiny common mode signals.
The next assumption we could make is that galvanic connecting the speakers plates might help for ceramic speakers.
Easy to try at no cost - compared to ALNICO-hype prices...
Michael
Basically you only need *distantace* to form an areal.
If we look at a speaker cable the singnal and its return should be close enough (= no distance for frequencies of interest) to each other to not act as antenna.
This should be true for all amps that are not prone to selectively demodulate the tiny *common mode* signals from its output and send them back to the speaker amplified so that you can listen to BBC news.
Things get different when we consider the speaker's coil to be kind of rabbit's ear like the wound areals of AM receivers.
Here we get a *differential signal* that is fed back to the amp's output.
If we have an all conductive shielding around the VC - as is the case for ALNICO - we hardly get that *differential signals*.
The assumption here would be that some amps may be more prone to the presence of tiny differential signals at its out than to tiny common mode signals.
The next assumption we could make is that galvanic connecting the speakers plates might help for ceramic speakers.
Easy to try at no cost - compared to ALNICO-hype prices...
Michael
Hi
I'd like to share some of my measurements of the Jantzen JA8008 in OB.
Actually its two of them, spaced by roughly 40cm / 16” in a 30cm / 12” wide baffle.
They are connected in split power arrangement as described several times. All measurements were taken in 1m distance at a SPL of 90dB.
– This can be seen as a backup to John Kreskovskys claim about minimum phase behaviour of speakers.
– Also it might show that a good speaker like the JA8008 isn't a limiting factor when treated kindly
– and last but not least it shows what powerful tool we have available with convolution
I used PC -XO , and as not too many might be familiar with that approach some words to the principle.
Basically we send our music through a process that's called convolution.
Convolution imprints “signal shaping” of whatever desire to the music. This is done by the use of so called impulse response files – you convolute the IR with the music signal.
The impulse response file does everything for us in one single step – it provides low cut, it provides high cut and it provides equalisation – it is “signal shaping” as JohnK coined it - at its best.
You output by a (pro) sound card that feeds your amps.
One side effect that is very important to note – you can't do “over processing” of the signal with convolution – meaning it sounds crisp and clear even when you chase out small irregularities - if you like (this is *not* true for passive and even most active equalisation as already mentioned here several times)
The art is in creating the “right” impulse response files.
One software that is capable to help you in this is Acourate from Ulrich Brüggemann. It provides a set of extremely precise and useful tools to manipulate phase and frequency to meet a acoustic target curve as close as you wish (and even more).
There is other software as well of course – if interested in details have a look at Shinobivans excellent thread.
Acourate has kinda steep learning curve but once you got it - its a breeze to work with.
The process of optimising a IR is done within about 15min including
- measurement (I do with ARTA)
- mporting into Acourate do all the manipulations necessary and create a new IR file
- restart your convolution host application
- play music or do another measurement
After this longish wind up – I document in this post my *ideal* (acoustic) XO target, in the next one the on axis measurements and in the third one measurements at different angles
First pix is my XO of desire – in this case a linear phase Butterworth 3rd order bandpass at 300Hz / 2500Hz
next pix is to show the CSD of that target curve.
Its an import of the ACOURATE impulse response file shown above into ARTE and proves what JohnK has shown several times here - even an ideal XO (not even an electric measurement) can show severe decay depending on the filters used
Michael
I'd like to share some of my measurements of the Jantzen JA8008 in OB.
Actually its two of them, spaced by roughly 40cm / 16” in a 30cm / 12” wide baffle.
They are connected in split power arrangement as described several times. All measurements were taken in 1m distance at a SPL of 90dB.
– This can be seen as a backup to John Kreskovskys claim about minimum phase behaviour of speakers.
– Also it might show that a good speaker like the JA8008 isn't a limiting factor when treated kindly
– and last but not least it shows what powerful tool we have available with convolution
I used PC -XO , and as not too many might be familiar with that approach some words to the principle.
Basically we send our music through a process that's called convolution.
Convolution imprints “signal shaping” of whatever desire to the music. This is done by the use of so called impulse response files – you convolute the IR with the music signal.
The impulse response file does everything for us in one single step – it provides low cut, it provides high cut and it provides equalisation – it is “signal shaping” as JohnK coined it - at its best.
You output by a (pro) sound card that feeds your amps.
One side effect that is very important to note – you can't do “over processing” of the signal with convolution – meaning it sounds crisp and clear even when you chase out small irregularities - if you like (this is *not* true for passive and even most active equalisation as already mentioned here several times)
The art is in creating the “right” impulse response files.
One software that is capable to help you in this is Acourate from Ulrich Brüggemann. It provides a set of extremely precise and useful tools to manipulate phase and frequency to meet a acoustic target curve as close as you wish (and even more).
There is other software as well of course – if interested in details have a look at Shinobivans excellent thread.
Acourate has kinda steep learning curve but once you got it - its a breeze to work with.
The process of optimising a IR is done within about 15min including
- measurement (I do with ARTA)
- mporting into Acourate do all the manipulations necessary and create a new IR file
- restart your convolution host application
- play music or do another measurement
After this longish wind up – I document in this post my *ideal* (acoustic) XO target, in the next one the on axis measurements and in the third one measurements at different angles
First pix is my XO of desire – in this case a linear phase Butterworth 3rd order bandpass at 300Hz / 2500Hz
An externally hosted image should be here but it was not working when we last tested it.
next pix is to show the CSD of that target curve.
Its an import of the ACOURATE impulse response file shown above into ARTE and proves what JohnK has shown several times here - even an ideal XO (not even an electric measurement) can show severe decay depending on the filters used
An externally hosted image should be here but it was not working when we last tested it.
Michael
Now here is what we can get at the sweet spot – meaning on axis measurement.
First pix shows the twin Jantzen JA8008 in the time domain shaped by the IR
next pix shows CSD of the Jantzen JA8008 shaped by the IR
I could have done equalisation even closer to the ideal curve but it might be pretty good as it is
Michael
First pix shows the twin Jantzen JA8008 in the time domain shaped by the IR
An externally hosted image should be here but it was not working when we last tested it.
next pix shows CSD of the Jantzen JA8008 shaped by the IR
An externally hosted image should be here but it was not working when we last tested it.
I could have done equalisation even closer to the ideal curve but it might be pretty good as it is
Michael
To conclude here are the pix for measurement at 15 deg angle
and the pix for measurement at 30 deg angle
and the pix for measurement at 45 deg angle
and last but not least an overlay of all measurements
The measurements at different angels are not very precise as I moved the mic around the speaker (too lazy for the lazy Susanne )
What is apparent is that the baffle is too wide for the frequency range of 300 to 2500Hz.
The peak of that kind of baffle is somewhere around 1kHz and above that the directivity begins to broaden before it would fall apart into lobes even higher.
This means that I would have to select a slightly smaller OB and a low pass around 1,5kHz at best.
To this frequency the 8" JA8008 would be veeery close to perfect. Some of the artefacts seen in 15, 30 and 45deg might be from edge diffraction of the support anyway.
Michael
and the pix for measurement at 30 deg angle
and the pix for measurement at 45 deg angle
and last but not least an overlay of all measurements
The measurements at different angels are not very precise as I moved the mic around the speaker (too lazy for the lazy Susanne
) What is apparent is that the baffle is too wide for the frequency range of 300 to 2500Hz.
The peak of that kind of baffle is somewhere around 1kHz and above that the directivity begins to broaden before it would fall apart into lobes even higher.
This means that I would have to select a slightly smaller OB and a low pass around 1,5kHz at best.
To this frequency the 8" JA8008 would be veeery close to perfect. Some of the artefacts seen in 15, 30 and 45deg might be from edge diffraction of the support anyway.
Michael
The issue here is that OB width of 30cm / 12" is limiting constant directivity to roughly 1kHz (the first OB peak).
Above that we see in the in FR overlay that traces are no longer congruent (towards 45 deg measurement a severe peak develops at 2kHz).
As John likes to outline, minimum phase behaviour of speakers is a function of point in space and if CD isn't met anymore (as is above 1kHz in my example) you can't do "one equalisation fits all".
*If* I cut the baffle width *and* lower upper XO - then the superb equalisation seen on axis will apply to all other angels as well.
(BTW above are all indoor measurements - no big deal)
Michael
Above that we see in the in FR overlay that traces are no longer congruent (towards 45 deg measurement a severe peak develops at 2kHz).
As John likes to outline, minimum phase behaviour of speakers is a function of point in space and if CD isn't met anymore (as is above 1kHz in my example) you can't do "one equalisation fits all".
*If* I cut the baffle width *and* lower upper XO - then the superb equalisation seen on axis will apply to all other angels as well.
(BTW above are all indoor measurements - no big deal)
Michael
Just seen that the 45 deg measurements are the same as 30 deg (wrong link)
Here are the correct 45 deg measurements:
Here are the correct 45 deg measurements:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Uhhh very few comments (not often seen those measurements put together somewhere else I admit)...
What I find most interesting is the behaviour we see most clearly at 45deg measurement.
The hump is created by the baffle not by the speaker!
We see that the OB behaviour is overriding speaker behaviour by far – exactly as is the case for boxed speakers. Though decay still is pretty short with the < 2ms shown - compared to boxed speakers where decay is several orders of magnitude longer (if not stuffed).
What else is striking to me is that we see decay from mere interference (front and back wave) – which I thought (and have read about) isn't the case.
This would have the impact that every comb filtering effect is prone to decay as well and *not* only to FR deviations as it is exactly the same case here.
Michael
What I find most interesting is the behaviour we see most clearly at 45deg measurement.
The hump is created by the baffle not by the speaker!
We see that the OB behaviour is overriding speaker behaviour by far – exactly as is the case for boxed speakers. Though decay still is pretty short with the < 2ms shown - compared to boxed speakers where decay is several orders of magnitude longer (if not stuffed).
What else is striking to me is that we see decay from mere interference (front and back wave) – which I thought (and have read about) isn't the case.
This would have the impact that every comb filtering effect is prone to decay as well and *not* only to FR deviations as it is exactly the same case here.
Michael
Normally I would like at both MLS based measurements and gated swept sine whenever I see response like that. Hard to comment on something not personally experienced. Whenever waves are in a relation where they will cancel them out somewhat, it is certain that it will decay much faster because of subtraction itself. Since the front and back waves meet at different times around the edge of the baffle, the cancellation also occurs at continuously different times. It would be interesting of this could be modelled and investigated to see how driver location on the baffle effects the response.
Midge said:Hugh very few comments (not often seen those measurements put together somewhere else I admit)...
What I find most interesting is the behaviour we see most clearly at 45deg measurement.
The hump is created by the baffle not by the speaker!
What else is striking to me is that we see decay from mere interference (front and back wave) – which I thought (and have read about) isn't the case.
This would have the impact that every comb filtering effect is prone to decay as well and *not* only to FR deviations as it is exactly the same case here.
Michael
What you are seeing is typical of a dipole response when the frequency ranges extends up to and beyond the dipole peak. A dipole is only constant directivity starting about 1 octave below the dipole peak frequency. Above that frequency, but below the first dipole null, the polar pattern starts to broaden out (the figure 8 getts fatter) and there is an off axis peak around the frequency of the dipole peak. If you were to LP the response with corner frequency an octave below the peak (or even lower) you would see true CD response. provided the driver also radiated omni directionally (both from the front and rear side).