You're right of course.
It's been a while (10years+) since I looked at their spec.
I get the impression that ATC and Quested are being rattled by diy.
A few years ago both would list useful info like xover points and slopes but both have stopped doing that now.
Not that there is much to forget since both use 24dB LR everywhere.
However Quested have moved away from the 3" Volt. Their smaller 3ways use a cheap(er) scandinavian-made domes while the large 4ways replaced the 6 or 7" cone+3" dome midrange for 2" Volt and a 7" hard polyurethane dome I know nothing about.
Possibly made for them exclusively by Volt since they do a few specials for others which are not available to the general public.
dunno what you talkin 'bout mate
have you looked at ATCs homepage?
they clearly state which x-over frequency they use for every single of their speakers.. from small two-ways to the big three-way top range. It's all there.
and its broadly known that ATC use 24dB x-overs.
I don't get your point here
ATC have been clearly not happy with others using their drivers that I am surprised it took them this long to withdraw from the market or why they were ever in it in the first place if it weren't for the fact that they started out as a manufacturer of drivers and became a manufacturer of complete speakers.
The drivers they use in their current monitors, the 3" S and the SL woofers, have never been available for sale individually.
Either way they used to publish quite a bit more useful info years ago.
The drivers they use in their current monitors, the 3" S and the SL woofers, have never been available for sale individually.
Either way they used to publish quite a bit more useful info years ago.
Alright. Since there isn't a dome that can quite match what the ATC can do in its range lets look at other technologies.
The two most promising available or soon to be available drivers I can find are from BMS and Celestion.
The BMS has some impressive specs and its smaller overall diameter allows for closer center to center spacing.
Overview
Celestion Axi2050 - AxiPeriodic Compression Driver
I would rear mount these drivers directly to the baffle and apply some damping material around the perimeters of the mid exit and tweeter. I've had good results with hornless compression drivers where wide dispersion is wanted.
The two most promising available or soon to be available drivers I can find are from BMS and Celestion.
The BMS has some impressive specs and its smaller overall diameter allows for closer center to center spacing.
Overview
Celestion Axi2050 - AxiPeriodic Compression Driver
I would rear mount these drivers directly to the baffle and apply some damping material around the perimeters of the mid exit and tweeter. I've had good results with hornless compression drivers where wide dispersion is wanted.
Neumann also have their own developed driver ever since ATC stopped making it for them in the newer series of KH monitors.
I've got two of these domes (94mm dome including surround, 80mm dome only). I've been testing these for a while now.
First test setup:
Performance of raw driver, measured without any waveguide or baffle:
Red is 3rd.
Promising, so on to making waveguides.
Cut from raw PVC 20mm thick plates. 2" Philips AD02150/SQ dome for size comparison.
Painted waveguide, in baffle with Satori MW19P-8.
Back view. Not my project, but a friend of mine. I have the same domes and waveguides though.
Gated measurements in waveguide.
Low distortion is probably due to copper in motor:
An externally hosted image should be here but it was not working when we last tested it.
Excellent post, Robbintip!
I've been wondering how those Neumann/K&H 3" domes would stand up to the venerable ATCs. Your response plots look very good, and I can't see any obvious problems. The output you've measured looks very clean and flat in the entire 500-5000 Hz range, especially with the waveguides in place. It would be interesting to see if the waveguides also result in the same consistent off-axis behaviour seen with the ATCs! 2nd harmonic stays below -60 dB over the entire range (-70 dB above 1 kHz), while 3rd seems to average around -75 dB. Now, the ATCs have even lower distortion, especially below 1000 Hz, but most cone midranges would be left in the dust already. Are all your measurements taken at a 2,83V signal level?
I'm very curious as to how you got your hands on these domes. They look very similar to a model made by Peerless Fabrikkerne India LTD (an OEM-only manufacturing facility, originally set up in the 70's by Peerless of Denmark, but later operating independently). You haven't included any impedance charts or T/S parameters, but if you would like to do some more measurements, it would be interesting to compare the results to this:
http://www.peerlessaudio.com/uploadfiles/pdf/3in dome midrange.pdf
I notice that the screw pattern on the front plate is slightly different from your units, but it is possible that one model is just a minor update of the other. The Peerless India website understandably isn't very detailed when it comes to constructional details where things matter, so there's no mention of the copper in the magnet gap. I'm pretty sure it can only have a single suspension, though, given the low moving mass. By comparison, the ATC's dual suspension and longer coil former results in a moving system nearly 3x heavier, which explains why they need those freakin' huge magnets to reach useful sensitivity!
I've seen others laugh at the merits of dual suspensions on domes, but, given the ATC's rather crude-looking motor design (a T-shaped pole seems to be as far as it goes), one has to wonder if the dual suspension is what actually takes distortion levels that low. For such a size trade-off, it would have to be worthwile, right? I'm only speculating, of course, but the distortion results you've come up with - especially if your domes are single-suspension ones - are impressive nonetheless.
In terms of low-end extension, the Peerless India domes actually appear to have superior Xmax (3,3 mm coil in 6 mm gap vs. 3,5 mm coil in 5 mm gap), but would be limited downwards by their higher Fs (~400 Hz). The Peerless data sheet also shows a Qms value that's nowhere near as high as the ATCs, but I'm not too worried about that. Some say that a high Qms (= low mechanical damping) is a good indicator of low energy storage, although the numerical value itself is only valid at the resonance frequency, which would be filtered out with a midrange driver anyway. To reduce energy storage at higher frequencies, other variables are probably equally important, such as dome coating, effective venting/termination into rear chamber, absorption on pole top, etc.
The cheap and flimsy Vifa/Scan-Speak 3" dome units used by e.g. Quested have been mentioned earlier. There are no published distortion plots for these domes, but they are basically carbon copies of the older Seas H304, for which data does exist. The latter has 2nd and 3rd harmonics dipping below -55 dB (@ 3,5 V), but only in a narrow band between 800-1600 Hz. Outside this range, distortion quickly rises some 12-18 dB per octave, and we're then back into cone territory. While an 80W nominal power handling is promised with 2nd order crossover at 500 Hz (high damping = flat impedance, which makes crossover close to Fs somewhat less risky), linear Xmax as defined by coil/gap height is only 1 mm peak-to-peak, so these will never sing very loudly in their lower range. A variety of the H304 was used by B&O as a "Phase-Link" device (i.e. to fill in the 2nd order crossover dip between same-polarity woofers/tweeters), rather than as a regular midrange driver.
regards,
Geir
I've been wondering how those Neumann/K&H 3" domes would stand up to the venerable ATCs. Your response plots look very good, and I can't see any obvious problems. The output you've measured looks very clean and flat in the entire 500-5000 Hz range, especially with the waveguides in place. It would be interesting to see if the waveguides also result in the same consistent off-axis behaviour seen with the ATCs! 2nd harmonic stays below -60 dB over the entire range (-70 dB above 1 kHz), while 3rd seems to average around -75 dB. Now, the ATCs have even lower distortion, especially below 1000 Hz, but most cone midranges would be left in the dust already. Are all your measurements taken at a 2,83V signal level?
I'm very curious as to how you got your hands on these domes. They look very similar to a model made by Peerless Fabrikkerne India LTD (an OEM-only manufacturing facility, originally set up in the 70's by Peerless of Denmark, but later operating independently). You haven't included any impedance charts or T/S parameters, but if you would like to do some more measurements, it would be interesting to compare the results to this:
http://www.peerlessaudio.com/uploadfiles/pdf/3in dome midrange.pdf
I notice that the screw pattern on the front plate is slightly different from your units, but it is possible that one model is just a minor update of the other. The Peerless India website understandably isn't very detailed when it comes to constructional details where things matter, so there's no mention of the copper in the magnet gap. I'm pretty sure it can only have a single suspension, though, given the low moving mass. By comparison, the ATC's dual suspension and longer coil former results in a moving system nearly 3x heavier, which explains why they need those freakin' huge magnets to reach useful sensitivity!
I've seen others laugh at the merits of dual suspensions on domes, but, given the ATC's rather crude-looking motor design (a T-shaped pole seems to be as far as it goes), one has to wonder if the dual suspension is what actually takes distortion levels that low. For such a size trade-off, it would have to be worthwile, right? I'm only speculating, of course, but the distortion results you've come up with - especially if your domes are single-suspension ones - are impressive nonetheless.
In terms of low-end extension, the Peerless India domes actually appear to have superior Xmax (3,3 mm coil in 6 mm gap vs. 3,5 mm coil in 5 mm gap), but would be limited downwards by their higher Fs (~400 Hz). The Peerless data sheet also shows a Qms value that's nowhere near as high as the ATCs, but I'm not too worried about that. Some say that a high Qms (= low mechanical damping) is a good indicator of low energy storage, although the numerical value itself is only valid at the resonance frequency, which would be filtered out with a midrange driver anyway. To reduce energy storage at higher frequencies, other variables are probably equally important, such as dome coating, effective venting/termination into rear chamber, absorption on pole top, etc.
The cheap and flimsy Vifa/Scan-Speak 3" dome units used by e.g. Quested have been mentioned earlier. There are no published distortion plots for these domes, but they are basically carbon copies of the older Seas H304, for which data does exist. The latter has 2nd and 3rd harmonics dipping below -55 dB (@ 3,5 V), but only in a narrow band between 800-1600 Hz. Outside this range, distortion quickly rises some 12-18 dB per octave, and we're then back into cone territory. While an 80W nominal power handling is promised with 2nd order crossover at 500 Hz (high damping = flat impedance, which makes crossover close to Fs somewhat less risky), linear Xmax as defined by coil/gap height is only 1 mm peak-to-peak, so these will never sing very loudly in their lower range. A variety of the H304 was used by B&O as a "Phase-Link" device (i.e. to fill in the 2nd order crossover dip between same-polarity woofers/tweeters), rather than as a regular midrange driver.
regards,
Geir
It wouldn't surprise me of they share some of their DNA ;-)
I currently stopped working on the speakers because it was impossible to use them with 12dB/oct. crossovers.
Maybe a friend of mine cares to eleborate some more on this subject.
Distortion wasn't measured at 2.83v as far as I remember. In hindsight, distortion performance is not as good as one would think looking at my graphs.
I currently stopped working on the speakers because it was impossible to use them with 12dB/oct. crossovers.
Maybe a friend of mine cares to eleborate some more on this subject.
Distortion wasn't measured at 2.83v as far as I remember. In hindsight, distortion performance is not as good as one would think looking at my graphs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.