I've yet to find a manufacturer showing real driver performance. They all show sugarcoated data on the verge of fraud. Quite frustrating.
I'd say they differ in important aspects:
Ture, nevertheless depending on the room they will sound different than a less distorted driver.
Attachments
I don't feel that this will change results dramatically. The off axis peak around 10kHz won't go away.
When listening to music, no. When listening to a reproduction device like a speaker, yes. You're confusing art and engineering.
That's just a red herring. Don't turn this into another religious war. Instead we could simply agree to disagree.
I have 1 of the 1st and 3 of the 2nd. I don't trust any of them above 10k. They are a tool to help build a better speaker. You have to keep in mind that the curves we see are what is happening on the surface... they tell us nothing about the performance 10, 20, 30, 40dB down (in the presence of that surface signal). It is what is happening down there that makes the difference between a good speaker and a great speaker. If you are evaluating the speaker soley on what the mic/measuring system tells you about the surface then you are not practising very good engineering. It is nice if it is mirror smooth surface but it is far from the whole story.
If the only tool you have is a hammer, then everything gets treated like a nail.
dave
That's why mine are calibrated. They show exactly what is happening.
What are you talking about? Non-linear distortion?
Another red herring. My measurements show the frequency response on and off axis. It's not good. If you think that frequency response is meaningless then so be it. You're free to believe whatever you want.
"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
Last edited:
Hmmm I think that is a tad extreme
I felt that SB accoustics specs for the SB12MNRX25-4 that I bought (a 4" but not really relevant to this thread) were quite good.
See attached SB's measurement (on IEC baffle at 36CM) and my measurement taken outside free feild in a small vented box with baffle only slightly larger than the driver at distance 1M. Mine is 12th octave smoothed. Also same measurement with distortion data. Note that the baffle geometry filled in some of the dip that occurs at around 1.5Khz with the IEC measurement, (due to the baffle step hump).
When I measured T/S parameters they were well within the variations I would expect when home measuring compared to factory measuring.
Measurements of my Morel DMS37 tweeters are *very* similar to published SPL curves as well (though FS seems way off).
Tony.
Attachments
Last edited:
Markus as I said comparing IEC baffle measurement to free field on a small baffle box I don't have the facilities to replicate a factory measurement. The main thing is that they aren't hiding anything in their measurement... I think 1/12th octave smoothing is perfectly fine.
Maybe if I have time I will SPL trace the factory measurement and apply baffle geometry of the cabinet and then compare. But probably not tonight
What did you use to overlay my measurement on the factory one? I'd like the ability to do that!
edit: The attached measurement was done outside at a distance of about 30cm from memory... I've detached this from a different thread as I don't have the measurement handy so don't know the exact details, but it was close enough to not be showing so much baffle step.
Tony.
Maybe if I have time I will SPL trace the factory measurement and apply baffle geometry of the cabinet and then compare. But probably not tonight
What did you use to overlay my measurement on the factory one? I'd like the ability to do that!
edit: The attached measurement was done outside at a distance of about 30cm from memory... I've detached this from a different thread as I don't have the measurement handy so don't know the exact details, but it was close enough to not be showing so much baffle step.
Tony.
Attachments
Last edited:
Markus as I said comparing IEC baffle measurement to free field on a small baffle box I don't have the facilities to replicate a factory measurement. The main thing is that they aren't hiding anything in their measurement... I think 1/12th octave smoothing is perfectly fine.
Maybe if I have time I will SPL trace the factory measurement and apply baffle geometry of the cabinet and then compare. But probably not tonight
Thanks for the offer but you don't need to do it for me.
If asked what manufacturers should show I'd like to see unsmoothed data 0°-90°.
What did you use to overlay my measurement on the factory one? I'd like the ability to do that!
Tony.
Did it in Photoshop.
Unsmoothed 0 - 90 I would think that for almost all drivers that would be unintelligible
Here is the above measurement RAW (ie no smoothing), but since this is taken outside and there are definitely interacting objects (like a wall about 3M behind the mic, trees, roof, garden etc it has to be taken with a large grain of salt as to what is the driver and what is the environment.
Tony.
I would think that for almost all drivers that would be unintelligible Here is the above measurement RAW (ie no smoothing), but since this is taken outside and there are definitely interacting objects (like a wall about 3M behind the mic, trees, roof, garden etc it has to be taken with a large grain of salt as to what is the driver and what is the environment.
Tony.
Attachments
Last edited:
Here you go Markus. Post 27 and 32
10G40 MLTL, Transducer Labs N26Cx-A/ Pellegrene waveguide - Page 2
2-way with Beyma 10g40 and TL Labs N26C in a wave-guide. Measures pretty flat with no smoothing applied. Doesn't touch the MA7.3 in sound quality. My opinion is that the woofer is not a good match for the tweeter/wave-guide combo, as neither are bad performers on their own. Unfortunately, the graph doesn't tell you any of this and you would have to listen for the explanation to make since. Distortion is excellent throughout the range. Measurements are useful tool in speaker design and can help tremendously with the final outcome, but they really say nothing about how it will sound. The 7.3 is really the ark I am trying to hit with a multiway, adding bottom end and improving dispersion high up. Yet to do so. Next is C Quenze 18h with TL Labs Tweeter.
Wintermute,
Check out the new MCF offerings. Excellent extended response, with good distortion numbers that rival the satori except for the LF area.
10G40 MLTL, Transducer Labs N26Cx-A/ Pellegrene waveguide - Page 2
2-way with Beyma 10g40 and TL Labs N26C in a wave-guide. Measures pretty flat with no smoothing applied. Doesn't touch the MA7.3 in sound quality. My opinion is that the woofer is not a good match for the tweeter/wave-guide combo, as neither are bad performers on their own. Unfortunately, the graph doesn't tell you any of this and you would have to listen for the explanation to make since. Distortion is excellent throughout the range. Measurements are useful tool in speaker design and can help tremendously with the final outcome, but they really say nothing about how it will sound. The 7.3 is really the ark I am trying to hit with a multiway, adding bottom end and improving dispersion high up. Yet to do so. Next is C Quenze 18h with TL Labs Tweeter.
Wintermute,
Check out the new MCF offerings. Excellent extended response, with good distortion numbers that rival the satori except for the LF area.
Last edited:
- Status
- Not open for further replies.