It's been 6 years since I had the first
production samples of the
JA8008 driver and
quite a range of speakers have been made around this high-efficiency
OBL-11 plus quite a few never published.
The new driver is JA8008-HMQ and the new main
speaker constructions are
as former mkII constructions were based on a modified crossover. In
addition to this, SEAS T35 tweeter options were made having the same
crossover as the Audax TW034 versions.
The old JA8008 driver will be
discontinued when sold out and should you by accident damage your
driver, there will be an up-dated crossover for those who want to stick
to the mkII design and not make the stepped baffle. To be launched as
soon as possible, but please do not ask for a launch date.
Should you want to up-grade your existing speakers with the new driver, look here for a kit allowing you to maintain a non-stepped front panel: http://www.troelsgravesen.dk/DTQWT-mkIII-FLAT.htm. A new crossover is required.
Standard tweeter is now the new SEAS T35C002, but AUDAX TW034 can be used with no changes to the crossovers. They will not sound exactly the same as the frequency response is not exactly the same, but this is not related to intrinsic quality, so please do not ask which one is best. The T35 tweeter was introduced because some people think old is bad and new is good and there's nothing I can do about it. There's another group of people who think old is good and new is bad, and I'm sure there will be arguments that the old JA8008 is better than the new JA8008-HMQ. Nothing I can do about this either.
Voice coil former: The voice coil available at that
time was wound on an aluminum voice coil former. Aluminum voice coil
formers introduce damping of cone movement by shorting eddy current
around the magnet gap. The result was low mechanical Q, actually 1.7, which is not too bad, but
one thing that can characterise vintage drivers is low
mechanical loss and a more vivid and dynamic presentation. I write
can characterise, because I've had vintage drivers that did not
have high Qm despite paper voice coil formers and "low-loss"
suspensions. Low-loss suspensions were sometimes added some viscous
resin to strengthen the suspension and ensure long-term stability - but
also had the impact of introducing a lot of damping to cone movement.
Not surprisingly the basic sound
of the new driver close to the old one (same cone), but due to new voice
coil former and low-loss supensions, you will notice an improvement in
transient response, transparency and low-level detail. Dust caps are
always a potential problem as they kind of live a life of their own and
often display some response in the treble range where they really
shouldn't. A very efficient way of dealing with this problem is not
having it at all and use a phase plug. The downside is loss of radiating
area (reduced efficiency) but the loss here is low as it is only a small
percentage of total radiating area and the added magnet size compensates
for the loss.
Above the two most important measurements: To
the left we see free air impedance plots of the old JA8008 (green) and
JA8008-HMQ (red). Due to the new voice coil former we see a huge peak at point
of resonance, which results in a whopping Qm = 10. This is what I wanted
- and even better than anticipated.
Above a CLIO screen shot from the 8008-HMQ mounted in the DTQWT mkIII cabinet. Red is without filter, green is 1.0 mH in series with the driver. No smoothing applied to reading.
Should you want to use the new driver for your own creations
you can obviously buy