home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.audio
- Path: sparky!uunet!drd!apctrc!znpt01
- From: znpt01@trc.amoco.com (Norman P. Tracy)
- Subject: Re: Low Price DAC
- Message-ID: <1992Nov11.153614@trc.amoco.com>
- Originator: znpt01@gpss43
- Sender: usenet@trc.amoco.com
- Organization: Amoco Production Company, Tulsa Research
- References: <1992Nov10.094830.2078@leland.Stanford.EDU> <1992Nov10.111307.28653@doug.cae.wisc.edu>
- Date: Wed, 11 Nov 1992 21:36:14 GMT
- Lines: 50
-
-
- In article <1992Nov10.111307.28653@doug.cae.wisc.edu>, loj@cae.wisc.edu (Lo Jeffrey) writes:
- > In article <1992Nov10.094830.2078@leland.Stanford.EDU> kong@leland.Stanford.EDU (Kong Kritayakirana) writes:
- > >
- > >
- > >CAL Sigma D/A Tubes??? Are they reliable? More expensive than
- > > the DDE v2.0 and the cosmetics isn't attractive
- >
- > I have never opened up a CAL Sigma myself, but I have heard that it uses
- > the Crystal Semiconductor CS4328 D/A chip which has *line level outputs*
- > on it, instead of the traditional current output which goes to an op-amp.
- > This means that the tube output stage is nothing more than a unity gain
- > op-amp which does nothing more than chew up the sound with more analog
- > circuitry. It looks good in an advertisement, though.
- >
- > Jeff Lo
- > loj@cae.wisc.edu
-
- I have opened up the CAL Sigma and examined the circuit, and measured it on
- the Audio Precision One DSP, and listened to it.
-
- Jeff is correct that with the CS4328 you get a line level output. What you
- do not get is a de-emphasis filter. So for CD applications some form of output
- amp/buffer is needed as a place to implement the de-emphasis filter. CAL
- uses what appears to be a hybrid tube/transistor output stage with discreet
- transistors around a 12AU7A (I think that is the tube type). Looked like they
- are also using an op-amp as a DC servo. The entire curcuit inc. power supply
- is on a single circuit board. Good quality fiberglass double sided board
- populated with quality parts. Given the low parts count allowed by the
- Crystal chip set about the only compromise evident inside the unit is the
- smallish (by hi-end audio standards) power transformers.
-
- As for the output stage "chewing up the sound" on the Audio Precision One
- using the DSP to generate S/PDIF test signals the results were what I would
- call outstanding. Total harmonic distortation + noise was better than -80db
- down (< 0.01%) from 20 to 20 kHz. The output vs. input linearity @ 500 Hz
- showed 16+ bit linearity, at -100 db in the output was -100.86 db. The freq.
- response was perfectly flat up to 4 kHz where a gentle roll off starts which
- is -0.7 db at 20 kHz. Keep in mind when auditioning the Sigma its output is
- +3.7 db above the 2 volt standard.
-
- In my system using a Sony 507es as a transport the Sigma sounded great. Clean
- clear and detailed. No warm wooly sound, you would never guess a tube was
- part of the circuit. In common with a DIY DAC I built using the CS4328 the
- Sigma tends to slightly emphasize the overtones in relation to the notes
- fundamentals. This becomes a system matching issue, on a speaker like the
- Vandersteen 2ce its great. The outstanding things about the Sigma are its
- imaging, detail recovery, and clean hi-end.
-
- Norman Tracy (znpt01@trc.amoco.com)
-