John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Jump to Latest
#13,321
Mark Johnson said:
Manly men who survived James Roberge's comprehensive opamp textbook, would prefer the LM101 / LM301 because it is externally compensated. Choose the gain-bandwidth tradeoff that YOU like, rather than mindlessly accepting the greatest-good-for-the-greatest-number compromise dreamed up by Fairchild.

And oh by the way, externally compensated opamps permit two-pole compensation for an additional performance boost, as explained in Linear Brief LB-4 (April 1969!)

I chose the externally compensated AD744 (JFET inputs) for the Norwood buffers, precisely because I wanted to (externally) set the phase margin to the number I wanted. Not to the number that the factory wanted.

_
Click to expand...


I remember reading that LB in about 1975/6 (as a very young man). I believe it was written by Robert Dobkin. A classic.
 
#13,322
PMA said:
I have a better one. Who knows what power amp it is?
.
.
Why do the manufacturers try to save every penny? Why do they use a non-latching XLR connectors in this amp, why not a Neutrik?? Why, at the same moment, they speak about careful part selection?? Your answer, R.S.?
Click to expand...

The 100k pot and the JRC4558 immediately put me off. The parts were of course carefully selected. For lowest cost, and that means even if your part is manifestly better but the other guy's is 0.1 penny lower, you lose the deal.

I lived and worked in Asia for 10 years flogging IC's into all sorts of markets (computing, mobile, consumer, auto) - I know how the consumer guys over there operate and it's ugly, ugly, ugly. That's all I'll say.
 
#13,323
Zung said:
Nope, they are potentiometers with continuous rotation, not switches; the smaller RK27 (Blue Velvet) have mechanical indents to mimic the feel of switches, but they're also potentiometers. And yes, they're rated for 15K cycles, so replacements are to be expected for pro use.
Click to expand...

hmm, I wonder what the number was of those attenuators, I have come across 3 of them over the years. All used in API consoles for the monitor volume. 2 of them were fixable, the other was worn out and replaced with an ELMA stepped attenuator. It was a little odd, they were a quad component wired in parallel for stereo. 4X10K

I wish the new audio company well in Indiana, looks like they are trying to utilize the human capital left behind by the abandonment of the Crown facilities in Indiana.

Cheers
Alan
 
#13,325
Bonsai said:
The 100k pot and the JRC4558 immediately put me off. The parts were of course carefully selected. For lowest cost, and that means even if your part is manifestly better but the other guy's is 0.1 penny lower, you lose the deal.

I lived and worked in Asia for 10 years flogging IC's into all sorts of markets (computing, mobile, consumer, auto) - I know how the consumer guys over there operate and it's ugly, ugly, ugly. That's all I'll say.
Click to expand...

The 4558 is far from the worst thing in that picture. I do not trust off-brand Chinese electrolytics.
 
#13,326
chris719 said:
Yeah, immersion silver looks awful after a very short amount of time if there are any bare pads. I'm not sure who uses it in mass production. Most consumer devices in high volume seem to be using OSP now.
Click to expand...

You're probably right, I haven't checked in a while. PCB isn't my main thing and I needed a modicum of corrosion resistance and solderability. Expensive, but probably best for my application (short of plating with platinum).
 
#13,327
Oh, definitely, ENIG and similar are very common. I think only the very high volume stuff like phones use OSP.

Seems like HASL is going away with everything moving to BGA and QFN packages.

I've ordered immersion silver before in protoypes from Sunstone but it was more of a question - I wonder who actually uses it.
 
#13,329
I'm working on a headphone with two 5mm X 5mm .4mm pitch 125 pin BGA's. It needs at least 6 layers and blind vias to connect the pins. The prototype PCB's were not cheap. Let alone parts and assembly. Troubleshooting can only be done via a microscope.
 
#13,337
Patrick/EUVL has asserted that the servo needs to be a supremely high quality, low distortion audio circuit. Because its output in the bass region is surprisingly large, and this is directly added into the feedback path (weighted sum). He dismisses the TL051 JFET opamp because its datasheet THD spec is only -90dB and that's just not good enough for a DC servo.

link 1

link 2
 
Status
Not open for further replies.
Top Bottom