If you give a look onto the frequency response shown in the datasheet, you will see the significant peaking at 4 and 6 kHz. This is caused by resonance of the voice coil neck, mainly by tension pulling (tensile strength and the mass of the voice coil including the part of the diaphragm nearest to the joint of the diaphragm and coil and similar part of the spider). These resonances amplify the distortion components, if the driver goes into nonlinear behavior. Besides them, the said resonances can be nonlinear themselves, too. Unfortunately, this can't be cured any simple way, and the distortion components appear even if the signal is cut out in this region by the crossover. You can try to damp the voice coil neck solely. The neck is easy accessible at this driver and you can cover it by some coating, lacquer or dispersion paint (or glue) based on acrylic or PVAc dispersion. The layer of the coating must not be too thin, because its tensile strength need to be comparable with the strength of the neck themselves. In fact, you need to eliminate the flaw of the design of the driver, neglected by the manufacturer.
Forgot any reflections on crossover resonance, sound reflections, and so on - these are a kind of typical acoustic superstition (I call it "asphalt" btw). In fact, any reasonable transfer function of the crossover can't cause any audible "ringing", particularly if the crossover is of fourth or lower order. In contrary, the steep xover removes the critical components of the usable signal better (the 6dB xover steepness is insufficient any case), although the separation of the distortion components said above isn't affected. I spent much time at theoretical as well as practical work around the crossovers in the past 30 years, so I know very well what I say. Note the speech is about the resulting transfer referred to acoustical output, i.e. assuming the transfer of the drivers themselvels, which has the form of bandpass of second order at least. Btw. the high pass part of the xover filter having 6 dB slope can't reduce the excursion of the voice coil around and above the fundamental resonance of the driver sufficiently; for this purpose is the transfer function of the order no less than the second one. Much of this topic has processed by Hawksford, btw.
Forgot any reflections on crossover resonance, sound reflections, and so on - these are a kind of typical acoustic superstition (I call it "asphalt" btw). In fact, any reasonable transfer function of the crossover can't cause any audible "ringing", particularly if the crossover is of fourth or lower order. In contrary, the steep xover removes the critical components of the usable signal better (the 6dB xover steepness is insufficient any case), although the separation of the distortion components said above isn't affected. I spent much time at theoretical as well as practical work around the crossovers in the past 30 years, so I know very well what I say. Note the speech is about the resulting transfer referred to acoustical output, i.e. assuming the transfer of the drivers themselvels, which has the form of bandpass of second order at least. Btw. the high pass part of the xover filter having 6 dB slope can't reduce the excursion of the voice coil around and above the fundamental resonance of the driver sufficiently; for this purpose is the transfer function of the order no less than the second one. Much of this topic has processed by Hawksford, btw.
Last edited:
This mid rises about 6dB from 2-6kHz:
http://zaphaudio.com/temp/Scanspeak-12MU8731T-00-FR.gif
You'll need to compensate for that in your crossover. Otherwise, it looks well behaved. There is nothing wrong with the arrangement of the frame or spider for the frequencies being used.
If you have another mid or fullrange speaker, connect that to your setup to see if it has the same problem so you can rule out amp/dsp/etc as being the culprit.
edit: ancient thread ladies and gentlemen.
http://zaphaudio.com/temp/Scanspeak-12MU8731T-00-FR.gif
You'll need to compensate for that in your crossover. Otherwise, it looks well behaved. There is nothing wrong with the arrangement of the frame or spider for the frequencies being used.
If you have another mid or fullrange speaker, connect that to your setup to see if it has the same problem so you can rule out amp/dsp/etc as being the culprit.
edit: ancient thread ladies and gentlemen.
My memory is now a little foggy, but as Troels and others here and I have discovered, the Revelator spec sheets are much better than spec sheet, and I believe better than Zaph's measurements. It leads me to wonder if there weren't engineering changes between then and now.
I really really encourage anyone using them to measure what they have in terms of Revelator and Illuminator drivers.
Best,
E
I really really encourage anyone using them to measure what they have in terms of Revelator and Illuminator drivers.
Best,
E
I really like this type of baffles, when combined with WG there is nothing to complaint about.
The quarter-elliptic shape of the corners is really good, IMHO better than the usual quarter-circle one. They are another solutions, too, even better maybe, but they are harder to define them for the numeric controlled machinery (they can be handcrafted, of course).
I suspect lower resolution measurements than Zaph or certainly SS do would appear smoother. But it's just beer goggles.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.