Audio Technology 10C77-25-10-KAP
Copyright 2009 Troels Gravesen

Link to AudioTechnology website.

The bass drivers shown here are meant for future 3-way constructions. Cabinet sketch files are piling up and a classic 3-way, PRELUDE, will be the first one released. Before cutting sheets, some initial measurements on the bass driver and considerations regarding performance in a vented box. Click images to view large.

We need both hands to lift this driver safely! I never had a 10" bass driver in hand with a bigger magnet and my bathroom balance says 8.5 kgs - and for once I think it may be right... The center dome is made from the same paper material as the cone and we'll have a hard time finding a driver where the dome has been cut and glued to the cone with greater care. Next a rugged frame construction and machined magnet parts. The upper part of the kapton voice coil former is laminated with aluminium foil (KA) to reduce the impact of eddy currents and to strengthen the voice coil former in general. One of the prices to pay for modern highly ventilated baskets is a longer voice coil former, added weight and the risk of flexing at high excursions.

Initial measurements

Left: SPL/2.8V, 1 meter distance, 38 x 100 cm baffle. Final system sensitivity expected to be 90 dB/1 meter/2.8 volts.
Right: Impedance free air (red) and 33.4 litre closed test cab (blue).

The peak...

This driver is dedicated bass driver and unlikely to play anything above 500 Hz, but for the sake of curiosity, let's take a look at the 2.5 kHz peak.

Distortion measurements were done with mic some 10 cm in front of cone. Input signal equivalent to ~85 dB/1 meter.
Left: Blue = 2nd harm. Green = 3rd harm. Basically <40 dB over the entire range.
Right: 1/6 octave smoothing and 3rd harm. (yellow) and 4th harm. (purple) shown.

Left: CSD @ 40 dB scaling. Right: Same @ 20 dB scaling.

Modelling point of crossover around 800 Hz? No problem, even with a simple crossover.

TS data

Left: TS data, CLIO measurements, added volume method. Drivers not broken in. Right: AudioTechnology TS data.

After burn-in:

Having played the 10C77s for a month I measured this. Qt a bit down, Vas a bit up.
I expect the drivers to only change a little after this. Variation is remarkably low.

Comparing 10" bass drivers

Left: AudioTechnology 10C77-25-10-KAP. Middle: ScanSpeak 26W/8861T00. Right: SEAS W26FX002.

Based on experience from the Jensen 1071, SEAS Cyclop, Jenzen NEXT, PRELUDE and Jenzen Accu the 10C77 is a clear winner in terms of overall performance. I rate the SEAS a tiny bit higher than the ScanSpeak despite ScanSpeak's higher mechanical Q but time will tell when the 26W/8861T00 is tried in the Jenzen transmission line cabinets.

Box simulation 10C77-25-10-KAP

First of all I do not intend to make a vented system for this bass driver, rather a transmission line cabinet. But before doing so, I want to investigate the vented option to get a better feel of the driver and its performance from a typical bass reflex enclosure.

Getting any 10" driver in hand we usually start thinking of some 40-60 liter cabs. Most 10" drivers will produce decent bass from this size cabinet unless TS data are severely "out of range", i.e. having extremely low or high Qt. Ideally only drivers with a Qt close to 0.35 will perform optimally in vented enclosures. As few drivers are even close to this requirement, we tend to use too large cabinets for low-Qt drivers and too small cabinets for high-Qt drivers.
A textbook QB3 tuning for the 10C77 driver will dictate some 20-22 liters and obviously this is too small. It would seriously compromise low-end extension, thus 45-55 liter volume has been tried in LspCAD to predict performance and this is really where we're heading if a vented box was to be.
Below examples of 45 and 55 liter cabs from calculated from AT data (left row) and my data (right row) despite drivers not broken in. In fact, the lack of break-in does very little to the performance as can be seen.

Based on experience I like a bass driver for vented boxes with a Qt around 0.25-0.3 in a volume 1.5-2 x calculated optimum volume - and tuned a little lower than optimal to produce a slightly shelved response towards lower frequences. This will make a decent flat response taking into account the inevitable room-gain. I assume some 80-100 cm to the rear wall (behind driver; some call this the front wall... whatever), 50-70 cm distance to floor and 1.5-2 meters to side walls. Placing a bass driver in a corner is doomed to fail except for dedicated subwoofers taking advantage of some 6-8 dB roomgain.

- click images to view large -

Left: AT data and 45 liter, 30 Hz vent tuning. Right. My data, 45 liter, 30 Hz vent tuning.

Left: AT data and 55 liter, 30 Hz vent tuning. Right. My data, 55 liter, 30 Hz vent tuning.

Left: My data and 50 liter, 30 Hz vent tuning. Right. My data, 50 liter, 30 Hz vent tuning, predicted roomgain.
Vent diameter and length has been erased as LspCAD does not calculate this correctly.
For 50L/30Hz, use 80 mm (D) x 200 mm (L) vent or 100 mm x 320 mm.
I suggest a decent flared 100 mm diameter vent to reduce vent noise. This bass driver can move some air!

As can be seen from above illustrations, even 50 liters may produce decent bass down to 30 Hz when placed e.g. 55 cm above floor, 110 cm to rear wall and 180 cm to side wall. This room-gain calculation is a simplified model and doesn't take into account a lot of things, but based on experience it does give a decent hint of where we're heading.

First listening impressions:

Having a pair of 65 liter cabs at hand I quickly set up the bass drivers and placed the AT-SWs on top. An electronic crossover, 24 dB-LR, made it easy to find a proper point of crossover between the two units. Playing a clean fender bass from a test CD is great tool to determine the best crossover frequency and 130-150 Hz made an overall coherent sound although the AT-SWs may do even better from being e.g. 15-20 cm closer to the bass driver - at least when listening at very short distance as my workshop only allows.

While playing the Fender bass track and adjusting the point of crossover from below 100 Hz to above 200 Hz I placed my finger tips on the 18H52 membrane of the AT-SW and I'm always amazed how much energy comes from the 100-200 Hz range. This is an indeed important frequency range and having a driver not capable of moving sufficient air leaves the overall sound thin and anaemic. 135 cm^2 like the 18H52 here does well down to 140-150 Hz, where an e.g. 6A77 with its 175 cm^2 does well to around 100 Hz. Using a 15H52 (95 cm^2) for mid we may have to increase point of crossover up to 300 Hz depending on front panel dimensions. The wider the panel, the deeper we can go.

The result from these experiments left no doubt in my mind that a system like this is a go! These bass drivers are exceptionally good. Clean, punchy and dry bass is the result from these 65 liter cabs initially fitted with 2 Variovents. Now, Variovents may be good for drivers having a higher Qt compared to these 10C77s, thus new rear panels were made and fitted with two 68 mm ports of 22 cm length, providing Fb ~28 Hz.Thanks to the generous motor system, these 10C77s' are control-freaks. I'm sure the majority of speakers that will be made from these driver will be vented systems.

Control is indeed what comes to mind when listening to these 10C77s. Adding two 68 x 220 mm vents to the rear brought things into perspective. Variovents with these drivers is a miss-fit. I don't think I've ever heard better bass from a vented system.


So, if you can't wait, the current recipe is this: The cabs seen above are a bit too high and the distance between bass and mid should be reduced to ~15-18 cm. Thus, bass cabs should be 36 x 65 x 36 (W x H x D). Made from 22 mm MDF, this will make a net volume of 62.5 liter. Allow generous bracing to swallow some 2-3 liter to make a final net volume of 60 liter. Place 10C77 with 10 cm from frame to top edge. Add two 68 x 200 mm vents to make Fb = 32-33 Hz. Place AT-SW on top and use an electronic 24 dB-LR at 140 Hz.
Result: It's a knock-out. Deep and dry bass, a midrange to die for and shimmering highs from the SWANS planar.
I can only imagine what it will sound like in rock-solid cabs, better power amps, better cables, better room, etc. Time will tell.

A classic 3-way from AT 10C77 bass + new AT 15H52-SDKAM middriver + ScanSpeak D2905/9900 tweeter.


January 2010: Numerous crossovers are being tested for this PRELUDE speaker featuring temporary middrivers while waiting for some new 15H52 drivers. 1st order filters for mid and LR2 for tweeter and bass deliver an astonishing presence and transparency with my new JungSon class A amp. "Problem" with this speaker is how loud it goes with very low distortion. All of a sudden I find myself listening at too high levels for the room....and for my ears.
Comparing the 10C77 bass driver to the SEAS W26FX002 - at low levels! - leaves no doubt that all the talk on the Audio Technology website about low-level authority isn't just - - talk. Indeed the rigid cone and high mechanical Q does the trick, which we usually have to go to PA drivers to find, only this driver comes with a set of TS data allowing modest cabinet size - and still maintains 90 dB/2.8V system sensitivity.

May 2010: Prelude launched. Click heading or image above. 

A few in-room (workshop) measurements on current set-up (10C77 + AT-SW)

Left: What happens when point of crossover is set to 130-150-170 Hz. Disregard response above 1kHz. These are nearfield measurements.
Middle: Left speaker in-room @ 2 meter, 100 ms time window and 1/6 octave smooting. Workshop has problems in upper bass, making sound a bit forward.
Right: Same thing, here left and right speaker. Right speaker has a dip (room derived) at 100 Hz and a boom at 30 Hz; sounds nice though!

Left: Left speaker with same and opposite polarity of bass and mid. Fair integration between bass and mid despite measuring conditions.
Middle: Nearfield measurements of port, bass and mid showing point of crossover at 130 Hz and port tuning at ~32 Hz.
Right: Same thing, only 20 dB scaling to show no midrange rubbish coming from port(s).

Overall: This speaker goes deep!

Click image to view large