SPI MISO line being held up

J

Jamie Morken

Guest
Hi all,

I am using a magnetoisolator (il717-3) on the SPI bus for one of the SPI
slave devices and I now realize that this is a bad idea as the isolator
has no tristate output on the MISO, master in slave out, line. So other
slave devices have no way to communicate back to the master SPI device..

I think I should use a tristate buffer chip to isolate the il717-3 MISO
output from the SPI bus, or a relay that is controlled by the SPI
master, any thoughts?

cheers,
Jamie
 
On Mon, 08 Nov 2004 08:31:18 GMT, Jamie Morken <jmorken@shaw.ca>
wrote:

Hi all,

I am using a magnetoisolator (il717-3) on the SPI bus for one of the SPI
slave devices and I now realize that this is a bad idea as the isolator
has no tristate output on the MISO, master in slave out, line. So other
slave devices have no way to communicate back to the master SPI device..

I think I should use a tristate buffer chip to isolate the il717-3 MISO
output from the SPI bus, or a relay that is controlled by the SPI
master, any thoughts?

cheers,
Jamie

You might get by with a series resistor on this pin.

John
 
John Larkin wrote:
On Mon, 08 Nov 2004 08:31:18 GMT, Jamie Morken <jmorken@shaw.ca
wrote:


Hi all,

I am using a magnetoisolator (il717-3) on the SPI bus for one of the SPI
slave devices and I now realize that this is a bad idea as the isolator
has no tristate output on the MISO, master in slave out, line. So other
slave devices have no way to communicate back to the master SPI device..

I think I should use a tristate buffer chip to isolate the il717-3 MISO
output from the SPI bus, or a relay that is controlled by the SPI
master, any thoughts?

cheers,
Jamie



You might get by with a series resistor on this pin.
Of course!! geez I should have thought of that :) Does 2Kohms sound
about right? I think that would mean that a slave device would need to
drive about 5V/2k=2.5mA

If I was to use a tristate buffer does anyone know of a single channel
buffer IC in a small smt package?

cheers,
Jamie


 
On Mon, 08 Nov 2004 19:38:50 GMT, Jamie Morken <jmorken@shaw.ca>
wrote:

John Larkin wrote:
On Mon, 08 Nov 2004 08:31:18 GMT, Jamie Morken <jmorken@shaw.ca
wrote:


Hi all,

I am using a magnetoisolator (il717-3) on the SPI bus for one of the SPI
slave devices and I now realize that this is a bad idea as the isolator
has no tristate output on the MISO, master in slave out, line. So other
slave devices have no way to communicate back to the master SPI device..

I think I should use a tristate buffer chip to isolate the il717-3 MISO
output from the SPI bus, or a relay that is controlled by the SPI
master, any thoughts?

cheers,
Jamie



You might get by with a series resistor on this pin.

Of course!! geez I should have thought of that :) Does 2Kohms sound
about right? I think that would mean that a slave device would need to
drive about 5V/2k=2.5mA
OK, if speed isn't a problem. Maybe a little lower, 1K maybe. The R
will limit clocking speed, depending on how much capacitance there is
on the data line.

If I was to use a tristate buffer does anyone know of a single channel
buffer IC in a small smt package?
Fairchild NC7SZ125, TinyLogic tristate buffer.

John
 

Welcome to EDABoard.com

Sponsor

Back
Top