[Coco] It's a small win, but a win nonetheless

[Gene Heskett]

On Saturday 11 March 2017 02:34:22 RETRO Innovations wrote:

> On 3/11/2017 1:22 AM, Dave Philipsen wrote:
> > Well if you can make it work without the cap then fine.  I guess I
> > don't know exactly what you're trying to build but if you are always
> > going to hook it up to a known entity that always has the same
> > characteristics you'll probably be ok.
>
> My goal is to read the data from the cassette port into an Atmel AVR
> MEGA168.
>
> > So you're saying that the waveform you're getting from the CoCo
> > swings between .6 and 1.3v referenced from ground?
>
> Yes, but you are correct, I should try on a few machines.
>
> > I guess I'd just want to be sure that that spec will always remain
> > the same without variations between different computers.
>
> I was able to clean up the jitter in the signal with the decoupling in
> place, but the end result (even with a 10M resistor to ground to help
> with the jitter that was left, the swing in the signal went to almost
> nothing (.4Vptp).  I am worried that does not give me much for noise
> margin.
>
> Jim

Jim, how big is the coupling capacitor? I'd use at least a .05 uf in a 
high impedance circuit, or a .5uf mylar in the lower impedance 
transistor circuit. The p-p measurement you started with, .7 volts, 
should not be reduced by the capacitor, only its dc operating point.  If 
its that reduced by the capacitor, the capacitor is too small. Multiply 
its capacitance by 5 to 10 in that case.

Also I don't recall if you specified the comparators jedec number.  There 
are comparators available whose inputs can include the -(ground) rail as 
opposed to needing a ttl level bias, or half the + rail bias at both 
inputs.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>