Output Impedance for Digital Logic Gate Chips?

[fastcat95]

Hello!

I am trying to find out what a typical output impedance
(in ohms) woulf be for a 74VHCU04 inverter gate. I am
planning to make up an SPDIF output buffer using this
chip, and will be paralleling 5 of the inverters for the
output. The question is what the total output impedance
of the 5 gates will be when trying to set up for a 75 ohm
final output impedance.

Anyone have technical knowledge about these gates?

Many Thanks!

fastcat

[Jocko Homo]

Guaranteed to sound rotten, if they are all in the same chip.

I think that I measured it once, and it was somewhere in the 20-30 ohm range.

Jocko

[fastcat95]

Jocko:

Thanks for your reply!

Do you know of an alternate output gate
system that would sound better for making
a simple SPDIF output, rather than using
a series of paralleled 74VHCU04 inverters
to get a good current output capability?

I have seen the paralleled inverter gate
circuit type elsewhere, and assumed that
it was done to improve current output
capability.

Many Thanks!!!

Fastcat

[Jocko Homo]

Why do you need so much extra current drive???

The most important thing is to make the output as close to 75 ohms resistive as possible.

I know.......a lot of places use parallel inverters, but that method increases jitter.

A simple one does work, but most are not implemented properly. The layout and routing are more critical than most people would imagine to get the output impedance looking right.

What is your starting point?

Jocko

[fastcat95]

Jocko:

Thanks for your reply!

I have an NEC 602 SCSI CD-ROM transport that
currently just runs the digital signal from a chip
on the main board to an RCA without any quality
form of buffering or correct output impedance.
Most people on the forums have said that the
quality of this SPDIF signal is not too good on
the NEC. (Rounded off, malformed digital waveforms)

So I want to come up with some sort of simple,
but effective IC buffer for the main board out signal
that also has the correct 75 ohm output impedance.
I am basic electronics/ construction savvy, but too
green to know the specifics of a good output buffer
design, although I am aware of the need for low noise
supplies, good bypassing, low inductance ground re-
turns, etc.

I am aware that many people also say that using a high
performance pulse transformer after the buffer is great,
but I do not have access to these transformers (even if
I did know which one to use). Do you have any ideas for
a good, simple IC buffer circuit with output resistors for
the 75 ohm SPDIF that would work off of a +5 VDC power
source?

Many Thanks!!

Fastcat

[Jocko Homo]

Digi-key used to sell Schott transformers......until Schott went kaput-ski. They made very good ones.

An old trick..........330 ohms (or so) from the output of the chip that the SPDIF comes from, and 100 ohm to ground. New output is from the junction of the 2. Will have a DC component, and some RXs do not like DC, so you may need to put 0.1 uF or so in series. Mucks it up some with all the inductance in the cap, but not too bad.

Run a small piece of 75 ohm coax to a BNC jack, and that will make a better output than what you have.

All this assumes that your chip outputs 5 volts, and that you want 0.5 V pk-pk, as it is supposed to be.

Jocko

[fastcat95]

Jocko:

Thanks fro the idea on the SPDIF output!

As to the 0.1 uF cap on the output, does all this
mean that a HQ COG multilayer ceramic capacitor
would be a good choice due to the HF inductance
issue, or would some other type be a better choice?

Also, I assume that the 330 ohm resistor would be a bit
less in practice due to the NEC chip output impedance?

Many Thanks!

Fastcat

[Da5id4Vz]

I used to buy these things called "Chabin Cables" for doing video circuits. They were about twice the diameter of a piece of Kynar wire and kind of springy. They were available pre-terminated with BBC and push on terminals w/ .1 inch spacing. They made it nice and simple to get a decent 75-ohm connection on and off a prototype board. I did some googel searching and cant seem to find them anywhere.

I know they were real popular with VME chassis based processors. Two companies I can think of that used scads of them in their products were Chyron and DaVinci color correctors.

Do these things sound familiar to anyone?

Its not entirely relevant to the questions here, but somehow I felt the need to jump up and say something.

[Jocko Homo]

A source of good, and affordable, 75 ohm cable is always welcome.

But anyway.......yes, the output impedance of the gate will raise it some.....probably to around 78 ohms. But most coax is rated +/- 3 ohms anyway.........+/- 1.5 if it is good.......and +/- 0.75 for the really expensive stuff.

Then it has to go to a 75 ohm BNC........

IOW, you are going to be a lot closer than J. A. Pan, Inc. is right now.

A COG is always better than the nasty ceramic mixes found in cheap bypass caps. Some guys will say that they can still hear the effects of one, but we are making progess on your system.

Jocko

[fastcat95]

Jocko:

Thanks again for your input!

I have heard some people say that the quality of
loading resistors used in an SPDIF can be more critical
than an very close match to 75 ohms. Hence the use
of things like Vishay and Caddock film resistors.

Any comments there?

Many thanks!

Fastcat