Michele Carla` wrote:
On Wed, 2006-04-05 at 08:16 -0400, Timothy Miller wrote:
I'm more concerned about noise and oscillations. I think you're right
that a simple reflection over less than a centimeter won't be a
problem for us. If we can tri-state the transmitter, and the receiver
has sufficiently high impedence, I don't see why we couldn't connect
them in pairs to let software select the direction. Only the
transmitter needs to be switched off, while we'll just ignore the
receiver when we're transmitting.
How do you tri-state an analog signal without distorting it? Just
cutting power to the opamp probably isn't enough because of whatever
will happen to the analog signal when it goes into the output of an
opamp.
No this isn't a good idea (IMHO), what about a VCA ?
That isn't going to help. It is still going to have a very low output
impedance even when you set the gain to 0.
What you need is an Op Amp that has an output disable. If you are
building this, you have MOSFET push pull output and to disable the
output you just turn (force) both of them off. If such a chip doesn't
exist, your only choice is an OTA which should act as an open circuit
when the bias current goes to 0. So, if you use non-inverting
configuration, all you have is the feedback resistance (like 10K to the
inverting input). Still it would probably be best not to drive the
non-inverting input with any signal as there might be some feed through.
--
JRT
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
