Bernd, do you have an explanation how the SPL could be the same with 2.9V and 5.9V at the transducer, or did I misunderstand?
To exclude a difference in language, 'at the transducer terminals' I understand measured at the two connections at the speaker driver?
Edit: or was the 5.9V measured at the other side of the 20R, at the amplifier side, not at the transducer side?
Jan
No, Sir, no, that is wrong.
Total lack of basic knowledge. Refusing to take subtle hints (#53 [1] [2] [3] ). The measurements are trash. Trash is an engineering term. ;-)
You failed the challenge. Thank you for participating. That is personal.
Last edited:
Dear Jan,
Thank you very much for your bright attention! Much appreciated, thank you!
You successfully and correctly reverse engineered my silly cabling mistake. That is really bright.
I do not need to explain anything. My measurements were open peer reviewed and replicated already, just not enough to seep in.
Help >keantoken< to post a meaningful measurement chart.
Best regards
Bernd
Thanks Bernd. Would it be too much to ask you to re-do your measurement, now that we know that the measurements are incorrect?
As you have seen, there's several people who are willing to join you in the measurements, which is less usefull if your own measurements are in error.
You may also wish to look at your peer review and replication process - you seem to imply that they replicated your wrong measurements without catching the error.
Jan
As you have seen, there's several people who are willing to join you in the measurements, which is less usefull if your own measurements are in error.
You may also wish to look at your peer review and replication process - you seem to imply that they replicated your wrong measurements without catching the error.
Jan
Last edited:
Kindergarten
The culprit is relationship between the electrical phases (sic!) and the SPL phase.
Best regards
Bernd
Yes. But you are an intelligent guy.
I'm sure you will eventually grow up to be a reasonable and mature person.
Jan
The screenshot above pictures what is sufficient to summon the theoretical heroes.
Best regards
Bernd
The reduced harmonic distortion at higher frequencies when using a series inductor ties in well with reported subjective results, in the case of this forum's Manzanita OB design. As I've suggested (on several occasions) before, in fact.
However, against that, there may be a problem when it comes to intermodulation distortion. Unusually, it would appear that, for once, IMD performance may well diverge from HD performance. I've hinted at this before, in another thread, and the basis is the following driver model:
https://audiojudgement.com/speaker-equivalent-circuit/
This suggests that an amplifier with negative output impedance would be the most appropriate in reducing IMD, as it would negate the effects of varying Bl with excursion (which would be minimal without the accompanying low frequency signal):
And of course, voltage drive is much closer than current drive in terms of implementing a negative output impedance...
However, against that, there may be a problem when it comes to intermodulation distortion. Unusually, it would appear that, for once, IMD performance may well diverge from HD performance. I've hinted at this before, in another thread, and the basis is the following driver model:
https://audiojudgement.com/speaker-equivalent-circuit/
This suggests that an amplifier with negative output impedance would be the most appropriate in reducing IMD, as it would negate the effects of varying Bl with excursion (which would be minimal without the accompanying low frequency signal):
And of course, voltage drive is much closer than current drive in terms of implementing a negative output impedance...
I would say the negative output impedance helps mostly around resonance in order to reduce the effects of BL nonlinearity.
Further up we have the effects of Lc as a function of excursion. And here a high driving impedance effectively helps. And it should help quite a lot when it comes to IMD.
Just keep in mind that you modulate the pole frequency that is caused by Lc and the (now) resistive part of the driver terminal impedance by large excursions in the LF range. This leads to AM and PM in the higher ranges that the driver has to reproduce.
As soon as I have the time to do so I will test IMD on the woofers that I mostly listen to when driving them with different driving impedances.
Regards
Charles
Further up we have the effects of Lc as a function of excursion. And here a high driving impedance effectively helps. And it should help quite a lot when it comes to IMD.
Just keep in mind that you modulate the pole frequency that is caused by Lc and the (now) resistive part of the driver terminal impedance by large excursions in the LF range. This leads to AM and PM in the higher ranges that the driver has to reproduce.
As soon as I have the time to do so I will test IMD on the woofers that I mostly listen to when driving them with different driving impedances.
Regards
Charles
The point is that Le won't vary very much when there's little low frequency signal, because excursion will be small. It's when you add in the large excursions caused by the presence of significant low frequency signal, Le will be modulated, but Bl will be modulated even more - and that affects the parallel elements in the model (above).
Hence there may be potential for increased IMD with current drive.
Hence there may be potential for increased IMD with current drive.
There may also be another mechanism at play here, affecting HD. At higher frequencies, distortion products may well fall into cone breakup and resonances. In this case, the degree of electrical damping offered by the amplifier could be significant. Ideally, the total mechanical impedance seen by the cone where it meets the voice-coil would match the characteristic impedance of the cone as a transmission surface (carrying transverse waves).
Yes, agreed, but I have no idea if that is even a consideration with manufacturers?
Jan
Yes, but in practice we can't build a voicecoil with low enough resistance or a magnet with high enough flux to uncover the cone resonance peaks so they can be damped.
Furthermore those resonances are often only loosely coupled to the voicecoil and so could not be critically damped.
Most real drivers have slight bumps in the voicecoil inductance where the cone modes almost push through. In order for it to be possible for these to be critically damped, the peaks must be fully unconvered from the inductance/resistance. You can try to use negative resistance but then all the motor distortion mechanisms get magnified dramatically.
Also, doing this may even worsen response. For instance in the case of even cone modes, their positive and negative displacement areas across the cone cancel out and don't radiate. But if you damp them through the voicecoil, you are then causing the modes to affect the response.
A simple way to look at it is that in order for us to damp cone modes through the voicecoil, we must reduce it's impedance to the point where the modes show up as peaks and dips. We must then drive it with voltage drive to have any damping effect, and so the response of the speaker will follow the peaks and dips of the impedance. If we use current drive we get the original flat response but no damping effect.
Furthermore those resonances are often only loosely coupled to the voicecoil and so could not be critically damped.
Most real drivers have slight bumps in the voicecoil inductance where the cone modes almost push through. In order for it to be possible for these to be critically damped, the peaks must be fully unconvered from the inductance/resistance. You can try to use negative resistance but then all the motor distortion mechanisms get magnified dramatically.
Also, doing this may even worsen response. For instance in the case of even cone modes, their positive and negative displacement areas across the cone cancel out and don't radiate. But if you damp them through the voicecoil, you are then causing the modes to affect the response.
A simple way to look at it is that in order for us to damp cone modes through the voicecoil, we must reduce it's impedance to the point where the modes show up as peaks and dips. We must then drive it with voltage drive to have any damping effect, and so the response of the speaker will follow the peaks and dips of the impedance. If we use current drive we get the original flat response but no damping effect.
Last edited:
Coincidently, I am on a project doing measurements on a field coil driver.
The most outstanding parameter is the very low distortion, on a par with an ESL.
I wonder whether the mere replacement of the permanent magnet with an electromagnet could have that effect, or whether there are other mechanisms at work?
Jan
The most outstanding parameter is the very low distortion, on a par with an ESL.
I wonder whether the mere replacement of the permanent magnet with an electromagnet could have that effect, or whether there are other mechanisms at work?
Jan