Spectral DMA-50? with tube preamp?
Art Altman
An out of town friend is advising me to take advantage of the
availability of a used Spectral DMA-50 amp. This amp was well
received by TAS around 1986 and Spectral itself appears to be well
regarded.
Questions:
1. Can someone comment on the quality of this amp relative to modern
units?
2. Spectral swears that you absolutely have to use their amps
exclusively with their preamps and MIT cabling, otherwise the system
will be "unstable". Yet I have heard rumor of others using Spectral
amps successfully in mixed systems. Do any of you have any experience
trying to use Spectral gear with nonSpectral, especially tube, preamp?
non-MIT cabline? Does Spectral's claim make sense? Apparently it is
tied to the ultrawide bandwidth of their units.
Proposed system is Audible Illusions M3a, Thiel 1.5, Cal Icon II Boss,
with Cardas Golden Cross all the way around.
thanks,
Art
aal...@epri.com
Lyle Hobby
Sorry
The DMA-50 has an input imp. of about 10k. A lot of tube preamps will
not drive it well. Also it seams to need a high capacitance cable
like a MIT to not sound brittle. My spectral pair was transformed for
the better by MIT many years ago.
Lyle Hobby
In article <60t33f$8...@agate.berkeley.edu>, aal...@epri.com says...
uh OH!
On 5 Oct 1997 15:16:13 GMT, ta...@csa.bu.edu (Anastasios Kotsikonas)
wrote:
>About cabling... I use Transparent's speaker cables with no
>problems. To me it's still a mystery how an amp can oscillate with
>non-networked cables when the speaker load itself rolls off above
>20kHz or so.
Without a zobel network, the rising inductance of the tweeter makes
the load more reactive in the ultrasonic range, although a zobel
network in the speaker, to match the impedance of the tweeter would
make more sense I think than a generic one applied in a cable. One
might worry about the slight inductance in the capacitor, and try to
bypass that, and then bypass that one, and then that one, until the
load is compensated well beyond 2 Megahertz, which I believe is the
upper bandwidth limit of most of the Spectral amps, the bottom being
DC. I might speculate that instead of tapering off the ultrasonic
response, designer employed more negative feedback to gain bandwidth
but compromised on ultrasonic stability, relying on the MIT networks.
Anybody know for sure?
colin