Below is display of sources of Non-linearity in drivers.
Note that (green) L(x) is a source of excursion modulation of inductance, while (brown) L(i) is a source of flux (current) modulation of inductance. The current variation is largely an effect in larger (and larger) woofers.
Also note that the display freq. range is where these problems typically occur for a driver.. BUT that cone vibration in particular (if it's not well controlled) will effect the driver throughout the freq. range by effecting coil position within the gap. Drivers with lighter compliant surrounds are particularly prone to this
The driver displayed appears to be a midrange/fullrange driver with an Fs likely near the cross-point of Bl and Kms.
Note that (green) L(x) is a source of excursion modulation of inductance, while (brown) L(i) is a source of flux (current) modulation of inductance. The current variation is largely an effect in larger (and larger) woofers.
Also note that the display freq. range is where these problems typically occur for a driver.. BUT that cone vibration in particular (if it's not well controlled) will effect the driver throughout the freq. range by effecting coil position within the gap. Drivers with lighter compliant surrounds are particularly prone to this
The driver displayed appears to be a midrange/fullrange driver with an Fs likely near the cross-point of Bl and Kms.
Attachments
Those up slopes too. All do it except H2, which seems to follow the beat of its own drummer. THD is just almost copy of H2, so it could be turned off.
Hey Dave, hope your doing better! 🙂
Noise is a "global" (multiple sources) thing in these cases, however..
JohnPM created REW (..which I believe was used for those harmonic distortion sweep plots), and this is what he has to say about (..and is much better at explaining it than I):
Speaker Distortion - REW Measurements - AVS Forum | Home Theater Discussions And Reviews
Speaker Distortion - REW Measurements - AVS Forum | Home Theater Discussions And Reviews
Last edited:
My guess is that 3rd is good to about 5 kHz in that graph.. and after that is pure crap. 😱
..and yeah, THD is pretty much useless.
Last edited:
I guess we could all discuss in vague terms.
But in response to your actual post: not really.
The choice with the calculations of each harmonic is: 1) calculate vs. the output at the fundamental frequency SPL (i.e., old school), and 2) calculate vs. the output at the harmonic frequency (when you think about it, this should be the way that it calculates). The choice that I used was "2)" for that plot. You can also select "1)" to see that plot, too:
Chris
Attachments
Last edited:
Perhaps a FR plot will illuminate a bit. This was 3 years ago when the Belle was set up as a center between two Jubilees using a sub only crossed at 40 Hz. The drop in LF output of the following plot was due to that condition at that time of using only one crossover frequency to the SPUDs: 40 Hz. I later changed it to 70 Hz for the Belle center channel only, and 40 Hz for the Jubs, and that FR drop out was bridged more successfully.
The 2-way JuBelle configuration (K-402 on top with B&C DE750) is much cleaner in terms of HD, even at 100 dB/1m, but 'll spare you...
The 2-way JuBelle configuration (K-402 on top with B&C DE750) is much cleaner in terms of HD, even at 100 dB/1m, but 'll spare you...
Attachments
Last edited:
I am thank you. Many fro yesterday’s event can attest to seeing me hobble around unaided on hard even ground.
Thanks for the links.
dave
LOL, thanks. I knew what it meant, and had wanted that option - but a lack of coffee forbid my brain from understanding it when I read it. 😀
"If the response of the system being measured is flat this makes no difference to the results, but when the response is not flat (as for most acoustic measurements) it can remove the influence of the loudspeaker's fundamental response from the distortion figures. "
That's the take away, right there.
OK, going back to the original post, Jon is asking two related questions.
Harmonic Waveform Generator :: Electronic Measurements
by playing with the harmonics and the waveforms. Try the same distortions, but at different fundamental frequencies. While this might not correlate directly to distorted music, it at least gives you a start to directly controlling it.
No, as least not as to their subjective effect. You can get an idea here
Harmonic Waveform Generator :: Electronic Measurements
by playing with the harmonics and the waveforms. Try the same distortions, but at different fundamental frequencies. While this might not correlate directly to distorted music, it at least gives you a start to directly controlling it.
How would you test that - either by ear or by measurement? How would you proceed? Perhaps take a good fullrange driver and measure and listen to it, then replace its bass and treble with a woofer and tweeter and find the differences.
Unless the surround does NOT stabilize/center the operation of the driver well, MOST of a driver's distortion centers around fundamental resonance. The exception is a surround "resonance" (which tends to occur somewhere between 900 Hz for large drivers up to 2.2 kHz for much smaller drivers).
A good way to select the pass-band for the driver is by looking at its impedance.
Here's the little 4 inch mid from SB:
http://www.zaphaudio.com/temp/SBAcoustics-SB12MNRX25-4-IMP-2.gif
A little above 200 Hz is a good high-pass area because it's right before the driver's impedance start's rising toward lower freq.s.
The driver has a surround "resonance" issue at around 1.6 kHz (..and it spans a wider freq. range than most drivers because of its very compliant surround). Still, it's unlikely you would limit a 4 inch mid to 1.6 kHz. 😱
Instead look to around 2.2-2.5 kHz (before the impedance starts rising much further from it's "resonant" range) as a more practical place for a low-pass area.
Analysis like this should give you a decent starting-point for the test for most cone drivers.
..IMD of course will be limited with a pass-band limiting filter.
Last edited:
- Status
- Not open for further replies.