Differential line driver unused outputs

Go To Last Post
13 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm using three of the four channels available on this chip: http://focus.ti.com/docs/prod/folders/print/mc3487.html

The enable line controls two channels, so I'm left with one enabled that I'm not using. In my design I have the input (A) for the channel tied to ground, and a 100k across the Y/Z outputs. Is this the right way to handle the unused outputs?

Thanks

I have too many hobbies.
s-conductor.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Unused outputs may be left floating. Not much difference from what You did, so nothing to worry about.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Great thanks!

I have too many hobbies.
s-conductor.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kas wrote:
Unused outputs may be left floating. Not much difference from what You did, so nothing to worry about.

Bullshit. In a disabled state the unused outputs should be handled exactly as unused inputs, i.e. should be never left floating. Moreover, adding a 100k resistor across the outputs doesn't do anything useful - should use pullups/pulldowns instead.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MBedder wrote:
Kas wrote:
Unused outputs may be left floating. Not much difference from what You did, so nothing to worry about.

Bullshit. In a disabled state the unused outputs should be handled exactly as unused inputs, i.e. should be never left floating. Moreover, adding a 100k resistor across the outputs doesn't do anything useful - should use pullups/pulldowns instead.

The outputs are enabled, according to the OP. I see no reason to put pull ups/pull downs on them since they're being actively driven.

Drive the input, float the outputs.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Bullshit. In a disabled state the unused outputs should be handled exactly as unused inputs, i.e. should be never left floating.

Tell me why oh masterfull MBedder?

I think 'floating' is the wrong term for a driven but not connected output.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If the driver is permanently enabled - of course this is not necessary (as well as a 100 k or whatever other resistor between the outputs). I wrote my above note for a case when the unused driver is being enabled/disabled along with a used one, as can bee seen from the word "disabled" in my post above.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So why would you not leave the disabled outputs floating?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Because they are internally connected to the chip's MOSFET structures which are in no way ESD tolerant from any terminal (G, D or S).

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I also consider leaving disabled three-state outputs floating as general practice, unless there is something that proves against this, and I can't find anything in the datasheet about this.

But of course you must pull inputs (data and enable) to a defined state, whether you choose to disable or enable the transmitter. In enabled state it just wastes power a little bit even if not connected to anything.

Edit:

Really, a RS-422 driver with no ESD protection? Since they are drivers, they surely are robust enough. Even product page says it handles 2 kilovolts of ESD..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Tri-stated outputs are MADE to be left floating...

In the case of unipolar outputs with no internal protection however, they should be pulled down through a resistor to prevent damage to chip from ESD.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

In the case of the MC3487, there doesn't seem to be any mosfets. So where's the bullshit?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I had on my mind TTL driver chip outputs as fellows referred to.
Should have mentioned that in original answer, indeed.