I used the "engineering" term with reference to every aspect of a driver, including its performance. It was not meant to be restricted to the driver's robustness. I disagree completely with the notion of "engineering without compromise", since even this thread has highlighted several compromises that I will further highlight here...
ATC withdrew their 75mm dome from public sales when they employed non-conducting pole pieces in their own loudspeakers. To give some idea of the reduction in distortion that is apparent in their drivers today, you can employ current drive in the older drivers to much the same effect over the mid frequency range. The audible improvement is marked (and is so in the vast majority of moving coil drivers I have ever encountered).
At low frequencies, however, where displacement is the limiting factor, the ATC dome represents a specific compromise apparent with a short coil driver. The specification indicates a crest factor of just +6dB due to the short-coil moving out of the magnetic gap. There is then a compromise apparent with the ATC dome between the linear operation within the gap and its "peak" output rating.
Notably Volt developed a near aesthetically-identical driver but employing a long-coil in a short gap specifically to address the latter issue, and permit a lower frequency crossover point. This represents a different linearity-versus-displacement compromise to that opted for in the ATC design. The ATC dome midband performance appears to have been optimised after the low frequency crossover point had been accepted.
I believe the Volt 75mm dome is used by PMC to compliment their transmission line designs (and possibly developed for this purpose?). Here (I suggest) a lower midband crossover might be preferable to better suppress resonances in the transmission line. For this application, quite possibly the Volt dome is indeed optimal. Therefore any worthwhile comparison between these two look-alike drivers and their compromises should take their application into consideration.
The doubling you mention is exactly two on account of the mechanism I just described. I would suggest the use of the term "typical crest factor" invalidates matters, however. The term "typical" lacks any appropriate reference to be useful here.
Drums, for example, can exhibit a crest factor in excess of 100dB. Reproducing such peak SPLs is inevitably compromised at realistic SPLs, but our non-linear hearing rescues us here to some extent. Estimates of transient audible distortion requires different analysing tools than provided by conventional continuous THD measures. (Many years ago on this forum, I tried unsuccessfully to promote bispectral analysis for this purpose). But using THD measures to compare peak level transient performance leaves us again not comparing "apples with apples".
So in conclusion, driver engineering (and subsequent manufacturing too) is all about compromises (despite what sales people might say). I would venture the opinion that the ATC 75mm dome is just about optimal in the systems for which it was intended, and not necessarily for those for which it was not (especially given that ATC produce entire active loudspeakers systems).
https://www.diyaudio.com/community/...00-1-fusion-f253-project.408815/#post-7591053
I've nearly finished the project above, the ATC is stunning, such clear detailed sound, 16ohm version
I've nearly finished the project above, the ATC is stunning, such clear detailed sound, 16ohm version
Anyone familiar with this?
Revival Audio RASC™ 75mm MIDRANGE
I couldn't find any data or price. I don't even know if the driver is available for DIY purchase.
Revival Audio RASC™ 75mm MIDRANGE
I couldn't find any data or price. I don't even know if the driver is available for DIY purchase.
Engineering is definitely an exercise in compromise! For instance, the shallow waveguide in which 3" dome is mounted extends the low end response, but also narrows the high end response, which may be a benefit, or a hindrance, depending of the design target.
I, for one, would like to see directivity measurements of a 3" dome in shallow waveguide, like that of the Volt, ATC (or Revival) in a typical baffle (35+ cm wide)
@mbrennwa
In your comprehensive write-up of the Open Source Monkey Box, did I miss some polar maps of the Volt driver?
If so, would you please let me know?
I, for one, would like to see directivity measurements of a 3" dome in shallow waveguide, like that of the Volt, ATC (or Revival) in a typical baffle (35+ cm wide)
@mbrennwa
In your comprehensive write-up of the Open Source Monkey Box, did I miss some polar maps of the Volt driver?
If so, would you please let me know?
Last edited:
@tktran303 I was wondering the same thing about the Volt. I recall some sort of off axis data being shared but don't remember where I saw it. That kink around 4.5k looks like a phase cancelation issue combined with the effects of the WG. I wish they made a neo version of that driver with a smaller flange. Its just such a massive chunk of a magnet not allowing a reasonably tight baffle arrangement. 1/2 WL spacing at xover is a pipe dream.
@profiguy
re "Sometimes DATS doesn't detect Qtc correctly and reads the impedance peak wrong if it isn't that symmetrical. This almost always throws off TSP measurements with higher frequency peaks, mostly mids and tweeters that have larger diameter VCs with slight double hump peaks. The presence of another smaller peak can caused by insufficient VC former venting (airflow restriction) coupled with the air cavity under the VC winding. Its very common with larger VC dome drivers, especially if they employ ferrofluid."
I checked the original DATS files (attached graphs) & in my case it looks like Zmax peak detection for Fs was performed correctly. (As mentioned these were free air measurements on new/standard drivers.)
I'll try to isolate & remeasure D7608s in enclosures tomorrow & see if that sheds any light on low Qts values.
re "Sometimes DATS doesn't detect Qtc correctly and reads the impedance peak wrong if it isn't that symmetrical. This almost always throws off TSP measurements with higher frequency peaks, mostly mids and tweeters that have larger diameter VCs with slight double hump peaks. The presence of another smaller peak can caused by insufficient VC former venting (airflow restriction) coupled with the air cavity under the VC winding. Its very common with larger VC dome drivers, especially if they employ ferrofluid."
I checked the original DATS files (attached graphs) & in my case it looks like Zmax peak detection for Fs was performed correctly. (As mentioned these were free air measurements on new/standard drivers.)
I'll try to isolate & remeasure D7608s in enclosures tomorrow & see if that sheds any light on low Qts values.
Attachments
Just take a look at the data at the GitHub repository at plots/raw_drivers.
Isn't the horn/waveguide on the ATC (and the Volts) too small to have much effect in the low midrange? I've wondered for a while if its primary function is actually just to provide some protection for the dome in the field, by recessing it behind the baffle surface/grille.
That is entirely correct! The morphing of the original straight-sided conical front plate into some curvilinear thing was a rather late modification with an unremarkable benefit. DIY constructors will also remember the odd requirement of having to mount the driver from the rear of the baffle.
Whilst there might well be some advantage in terms of reducing magnetic non-linearities, I would anticipate a lower power handling capacity due to the decrease in the encompassing metalwork.
I have not seen the Volt data either, but such an effect around 4.5kHz is common to 75mm domes in general with or without a waveguide, and provides a practical limit to their useful operating bandwidth.
From the descriptions here, it is not entirely clear to me where exactly that balsa ring is placed in the ATC.
@soundbloke: could you provide a bit more detail to clarify its construction and placement?
@soundbloke: could you provide a bit more detail to clarify its construction and placement?
The ring is glued to the inside edge of the former, around its top. If you cut away the dome, it would be the first thing you would see.
You could add some cooling fins at the back for more efficient heat transfer. Even a fan if you take it serious ...
Practical limit is way lower if you want good off axis behaviour. These domes normally don't radiate very wide.
A modern 3" Dome doesn't need the recessed mounting and has it's dome resonance >10kHz.
https://hificompass.com/en/speakers/measurements/bliesma/bliesma-m74a-6
Off axis beaming starts at 2kHz, crossover at 3kHz already means you would need a waveguide for the tweeter.
@profiguy
With Matthius' consent, I processed the off axis data for the Volt VM752 mounted on a 360mm W x 682mm H baffle.
The center of the dome is 260mm from the top edge, offset 43mm from center line.
This is part of @mbrennwa 's stellar Open Source Monkey Coffin loudspeaker-
Reference:
https://www.diyaudio.com/community/threads/open-source-monkey-box.327594/
Here is the frequency response of this 3" dome on a relatively wide baffle, without crossover:
You might notice a big dip ~2.5KHz.
The project documentation said "The on-axis response of the Volt VM752 midrange shows a dip at 2.3 kHz. This dip disappears in the off-axis curves, which indicates that this is an uncritical diffraction artifact related to the driver/waveguide geometry rather than a problematic resonant effect."
Moving forward, by massaging the on-axis perfectly flat, this is what you may get in the off axis:
Polar (heat) Map:
Directivity:
I recall the crossover was acoustic LR4, but not the exact crossover points.
Although the directivity is slightly asymmetric, the DI is impressively low and flat.
The -6dB beamwidth is about +80 degrees one side and -70 degrees on the other side, from 300Hz to 2.5KHz.
ATC rate their SCM50ASLPro as having a "Horizontal Dispersion: ±80°, Coherent". I guess there is a method to their madness of using a small dome on a wide baffle to control directivity.
With Matthius' consent, I processed the off axis data for the Volt VM752 mounted on a 360mm W x 682mm H baffle.
The center of the dome is 260mm from the top edge, offset 43mm from center line.
This is part of @mbrennwa 's stellar Open Source Monkey Coffin loudspeaker-
Reference:
https://www.diyaudio.com/community/threads/open-source-monkey-box.327594/
Here is the frequency response of this 3" dome on a relatively wide baffle, without crossover:
You might notice a big dip ~2.5KHz.
The project documentation said "The on-axis response of the Volt VM752 midrange shows a dip at 2.3 kHz. This dip disappears in the off-axis curves, which indicates that this is an uncritical diffraction artifact related to the driver/waveguide geometry rather than a problematic resonant effect."
Moving forward, by massaging the on-axis perfectly flat, this is what you may get in the off axis:
Polar (heat) Map:
Directivity:
I recall the crossover was acoustic LR4, but not the exact crossover points.
Although the directivity is slightly asymmetric, the DI is impressively low and flat.
The -6dB beamwidth is about +80 degrees one side and -70 degrees on the other side, from 300Hz to 2.5KHz.
ATC rate their SCM50ASLPro as having a "Horizontal Dispersion: ±80°, Coherent". I guess there is a method to their madness of using a small dome on a wide baffle to control directivity.
Last edited:
You can chop three 120° breathing holes when magnet blocks. Known tip for Dali 18 owners. I think that has 2 inch baffle.
No, it uses unusual 3rd order elliptic filters very steep, and upper XO at 3khz.
Attachments
Bliesma has removed the product description of the M74 from their website. Earlier they responded to an email where I asked about Beryllium availability so someone could ask them if the M74 is officially discontinued or just temporarily unavailable (perhaps due to Beryllium sheet supply chain issues).
To be clear, I'm wondering what the acoustic response is, after the electrical filters are applied.
I couldn't find it in the documentation.
I couldn't find it in the documentation.
That is another statement that I don't think is generally applicable when system design is taken into consideration, not to mention the listening environment and the application. The most important factor in terms of directivity is to achieve a monotonically decreasing sound power output response, hence the common adoption of crossing over a 75mm dome in the 3.5-4kHz region that avoids the 4.5kHz problems. (it is also worthy of note that most applications of a 75mm dome are in active systems where 4th order alignments are commonplace).