[Coco] serial port problem

[Gene Heskett]

On Friday 01 November 2013 07:42:27 Steven Hirsch did opine:

> On Thu, 31 Oct 2013, Gene Heskett wrote:
> > The 741 is the most sloppily specified op-amp ever.  I have seen some
> > that will make a slew rate limited sine wave out of a 1khz square
> > wave.
> 
> Indeed.  First generation monolithic op-amps received a well-deserved
> bad rap for terrible sounding audio electronics, largely due to horrid
> slew rates.
> 
> The later bi-fet input devices (LF357, etc.) were not too bad in the
> right circuit topology.

Even the legendary 5532 was in reality a POS, but every body thought that 
because it needed a heat sink clipon and a fan, it must have been pretty 
good.  But at 20khz, it slew rate limits at abut 3 volts p-p.
The LF357 is usable, but I actually made all the audio DA's at WDTV using 
TLO84's.  Each card had a stereo input buffer that drove 4 ea 20k pots for 
individual output gains, for 1 input, 4 gain adjustable outputs per 
channel, 2 channels per card.

Much of our old analog system was all mono though because we had never put 
a stereo kit in the modulator.  That old phaeostron (sp?) based modulator 
did not have the baseband bandwidth to pass it anyway.  That design was on 
its knees, begging for mercy at 15 kilohertz, and because it was basically 
velocity sensitive, fell off like a brick wall with very high distortion at 
about 50 hertz.  It could just barely make a proof of performance, but if 
the power supplies did their job, it was a hockey puck for 10 years at a 
time after I had replaced the wandering, $1500 oven stabilized crystals 
with a $750 Vectron stabilized oscillators. Those were so stable that once 
I had put then within 10hz of channel 5, they were still within 20hz when 
we shut it off in 2008.  For 20+ years thats not half bad at all.  FCC 
tolerance was 1khz, but they measured the intercarrier for the audio, so if 
the rocks varied with the heater thermostats, which they did by kilohertz 
amounts when I ordered the Vectron's, you could be out of specs if they 
went opposite directions by only 501 hz each.

The only problems I had with the TLO84's as audio DA's were EMP related, 
long cables on the outputs made excellent antenna's so I usually had 5 or 6 
cards to shotgun after a severe thunderstorm, easy since I had them 
socketed.  Kinda hard to avoid when you have a 255 foot tall lightning rod 
40 feet outside the back side door.

But those TLO84's could put 28 volts p-p at 20khz, into the source ends of 
the 300 ohm a side push-pull build out resistors in each channel, showing a 
just barely visible dead spot flat at the zero crossover.  The distortion 
analyzer, a panasonic that went down to .001%, showed that as about .1%.  
Drop it to 20 volts p-p and it was .025%, all random noise.  Source 
generator was inhouse made by one of my other engineers, from a state 
variable design, it was about as pure as you could buy for 4 or 5 thousand 
dollars commercially, but we had maybe a $40 bill in it.  Used a lot of 
surplus parts.

And hey, TLO84's were dirt cheap (under a buck ea.) in sticks of 25. ;-)  
5532's would have needed 20x the power supply capacity, noisy fan cooling 
and another kilowatt in HVAC for the control room.  And would have been 
about $175 a stick of 25.  And from my experience, they were even more EMP 
sensitive than the TLO84's, we had an audio console that used up a couple 
sticks of them in the 15 years we had it.

Cheers, Gene
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)

If an experiment works, something has gone wrong.
A pen in the hand of this president is far more
dangerous than 200 million guns in the hands of
         law-abiding citizens.