Tri State Devices

Is a Tri State "driver"(I think thats what its called) basicaly a device which lets you set either Low Z or High Z?

something like

C | |------| A ---| |--- B |------|

Where, say, if C is high then A and B see a high Z(open circuit) and if C is low then it acts like a low Z? (a short)

or maybe they are only unidirectional?

But basicaly it is like a controled switch?

The reason I ask cause in a circuit they have something like

D2 >------------+ | | |------| D0 >---------| |---+----- Data |------| | | ACK >-------------------+

Where the device is 74LS244.

Only thing I can figure out is that D2 acts like a control that lets D0 pass through or not. If its off then ACK "senses" the Data...

i.e.,

So D2 off(or on if inverted) would allow D0 to send data through to Data... and ACK could "sense" this if it wanted... if D2 was on then D0 couldn't send anything and ACK could then be used "sense" output comming from Data.

So, in a sense it lets Data be used as an input and output port... but without the device then data could go to D0 and if D0 is only an output it will be recieving input which might screw it up?

Is that the basic idea?

Thanks, Jon

Reply to
Abstract Dissonance
Loading thread data ...

A tristate driver ( a registered trademark of National Semiconductor, incidentally) has three possible states (giving it the name).

High. Low These are the output states of normal logic

High-z. This means the output buffer is 'disconnected' from the output pin (and therefore any circuitry hanging off it).

In all cases, it may be looked at as a controlled switch, where the output may be disconnected.

Found in many an application (most commonly in bussed applications, but often elselwhere).

This permits mutiple *outputs* to be connected together, with the designer setting it up so only one can be active at any time. For lower speed to do the same thing, we used to use open collector / open drain (and for ECL open emitter) techniques.

Most memory devices (even those that are read only) have tristate outputs, which become active on some mixture of control signals (usually output enable + chip select). Buffers and transceivers also have this when designed for use in such a system.

Another very popular use is to buffer signals across power domains, such that if a particular domain is unpowered, the signals are buffered through a tristate device that is held in tristate, preventing power bleed across domains.

Without seeing the entire schematic you are referring to, it's hard to know just what the designer was trying to achieve.

Cheers

PeteS

Reply to
PeteS

They are unidirectional (A->B) in your case, and digital only. They shouldn't be thought of as a switch as they aren't. It's a digital buffer whose output drive can be shut off.

I think thats a bad description, since a pass-gate is much more similar to a "controlled switch".

The schematic you drew is essentiall a bidirectional I/O. "ACK" sees whatever is on the Data line. If D2 is asserted the tristatable buffer asserts D0 onto Data. The theory here being that if you want to read, you shut off D2, and Ack will have valid data being asserted from off the schematic. If you want to write a value, D2 is asserted and D0 is buffered through to Data. Ack will still be a valid read, but it will simply read back D0.

One of the more common applications for this is on bidirectional busses. For instance on the ISA bus, the CPU may want to pass data to the peripheral boards by asserting D2, and then it may want to allow the peripheral boards to pass data back to it over the same lines by de-asserting.

Reply to
blah

--
If it were to be done with relays, here\'s how it would work:


               +V      GND
DATA>------+   |        |
           |   O--> |--------+        |
           |        O
         [COIL]- - -|
           |   O--> |
          GND  |
               +------->OUT



  DATA  | OE | OUT
 -------+----+-----
   GND   GND   HI-Z
  
   +V    GND   HI-Z

   GND   +V    GND

   +V    +V    +V

HI-Z meaning that OUT is floating.
Reply to
John Fields

Ok, Makes sense. Basicaly thats what I was thinking but I couldn't say it properly.

Thanks, Jon

Reply to
Abstract Dissonance

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.