wrote:
On Wednesday, August 9, 2017 at 7:58:03 AM UTC-5, Scott Dorsey wrote:
In article ,
geoff wrote:
On 9/08/2017 9:49 PM, Scott Dorsey wrote:
Or a common-or-garden 5534/5534 . Dunno the LME49710NA, but a quick
glance looks like the dog's ********.
I'd do the 5532/5534 myself. It is a remarkably good-sounding part and
it measures well and is very stable. It gets an undeservedly bad rap with
audiophiles but it is a good choice.
Specifically designed to adequately drive a 600R load IIRC.
Yes, although just barely. It can benefit from a couple transistors on
the output if you are doing that. Still, it'll perform great into a
low-Z summing network... the busses on the Tridents were made comparatively
high-Z to keep power demands down, at some noise expense.
Scott Dorsey, how and where would someone wire in those transistors and which transistors?
I was referring to the idea of putting a push-pull pair on the output of an
opamp (but inside the feedback loop) to get more current drive capability than
is possible with just a monolithic chip.
Burr-Brown has a whole applications note on doing this, in
http://www.ti.com/lit/an/snoa600b/snoa600b.pdf and so does the competition
at
http://www.linear.com/docs/4114 as well.
If you look at a lot of consoles from that era, many of them use booster
circuits like this both on the outputs (to drive 600 ohm loads easily) and
on the buss amplifiers (in order to reduce the summing buss impedance so
lower resistance summing resistors could be used, for lower noise). But
doing this is not cheap and Trident was very much built to a price point.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."