Author Topic: Help deciphering ts5a23159 power off state  (Read 638 times)

0 Members and 1 Guest are viewing this topic.

Offline oliv3rTopic starter

  • Frequent Contributor
  • **
  • Posts: 283
  • Country: nl
    • Rigol related stuff!
Help deciphering ts5a23159 power off state
« on: July 16, 2024, 03:55:47 pm »
Hey all.

I'm trying trying to understand what the [TS5a23159](http://www.ti.com/lit/ds/symlink/ts5a23159.pdf) does when no voltage is present on VCC (VCC < 0.1).

The datasheet has only one image really talking about the 'PWROFF' state, Figure 15 (attached). Which somehow ironically seems to imply that when VCC is 0, we get NO closed, and NC open?!

What I'm after, is that I want to know what the default state of the switch is without power applied, or if it always needs power to function. If this where a relais the answer would be obvious, but this being solid-state territory; I don't know.

I currently toggle the power to this switch and some other control logic via a mosfet, and while the consumption is probably irrelevant, I do want to know if I should keep the switch always powered.

Offline xvr

  • Frequent Contributor
  • **
  • Posts: 384
  • Country: ie
    • LinkedIn
Re: Help deciphering ts5a23159 power off state
« Reply #1 on: July 18, 2024, 01:41:33 pm »
Datasheet paragraph 10 (Power Supply Recommendations):

Quote
Always sequence VCC on first, followed by NO, NC, or COM.

Also, all Electrical Characteristics tables contains row:
Quote
COM, Vno, Vnc: Analog signal range: 0 - VCC

So when Vcc is 0 all input/output should be zero.

Quote
What I'm after, is that I want to know what the default state of the switch is without power applied, or if it always needs power to function.
Default state doesn't matter - output always will be 0

As I can guess this chip (as any other CMSO) has protection diodes to power rails on input/output, so when you turn off power (to high Z state) it will power itself from all NC/NO inputs (or COM, if it will be used as input)
 

Offline oliv3rTopic starter

  • Frequent Contributor
  • **
  • Posts: 283
  • Country: nl
    • Rigol related stuff!
Re: Help deciphering ts5a23159 power off state
« Reply #2 on: July 21, 2024, 04:14:57 pm »
Datasheet paragraph 10 (Power Supply Recommendations):

Quote
Always sequence VCC on first, followed by NO, NC, or COM.

Also, all Electrical Characteristics tables contains row:
Quote
COM, Vno, Vnc: Analog signal range: 0 - VCC

So when Vcc is 0 all input/output should be zero.

Quote
What I'm after, is that I want to know what the default state of the switch is without power applied, or if it always needs power to function.
Default state doesn't matter - output always will be 0

As I can guess this chip (as any other CMSO) has protection diodes to power rails on input/output, so when you turn off power (to high Z state) it will power itself from all NC/NO inputs (or COM, if it will be used as input)

Thanks, so how was I supposed to read that line. I just read it as 'the analog input range will between between 0 and VCC.

So if they are 'forced' to 0, that would mean, removing power would be a 'third' state, disconnected.

So why not just draw the picture like that, if vcc is off, COM is not connected to either pins?

But regardless, you answered my question, always supply power, as removing power will 'disconnect' all inputs.

Offline MasterT

  • Frequent Contributor
  • **
  • Posts: 808
  • Country: ca
 

Offline oliv3rTopic starter

  • Frequent Contributor
  • **
  • Posts: 283
  • Country: nl
    • Rigol related stuff!
Re: Help deciphering ts5a23159 power off state
« Reply #4 on: July 23, 2024, 02:07:03 pm »
Why not to look at TI web site?
"The VI/O pins (COM, NC, NO) can be subjected to voltages from 0V to 5.5V when the supply Vcc = 0V. "
https://e2e.ti.com/support/switches-multiplexers-group/switches-multiplexers/f/switches-multiplexers-forum/1188100/ts5a23159-can-you-please-confirm-that-it-is-okay-to-apply-1-8v-to-nc-no-or-com-even-when-vcc-0v?tisearch=e2e-sitesearch&keymatch=TS5a23159#

What do you mean? Does the TI website list stuff that's not in the official TI datasheet? That forum post kind-of almost asks the same I suppose, but its still not clear to me what happens to the signal. Will NC still conduct yes/no was my confusion. As I'm not an EE, the datasheet is not clear to me what happens when no power is supplied to the chip. Will it still pass voltage (the signal) or not on the NC pins. The picture on Figure 15 would suggest it does pass the signal, as the switch is closed. But this I've learned is not entirely true, due to the semiconductors of things ...

So to me, the switch has 3 states, A -> COM; B -> COM; nothing -> COM, which would be achieved by removing power on the VCC pin.

Offline MasterT

  • Frequent Contributor
  • **
  • Posts: 808
  • Country: ca
Re: Help deciphering ts5a23159 power off state
« Reply #5 on: July 23, 2024, 02:32:59 pm »
I see. 99.99%  switch doesn't conduct in case of absent power. "Isolated" means all ports (NC/NO etc) are in "high impedance state".
I 'm thinking to get one soon for evaluation, if it's any better than ts5a23157 I used to build ultra linear pwm dac.
Will test and update this post
 

Offline MasterT

  • Frequent Contributor
  • **
  • Posts: 808
  • Country: ca
Re: Help deciphering ts5a23159 power off state
« Reply #6 on: July 25, 2024, 01:30:48 am »
Get one, overnight delivery.
Switch, as expected, in case of Vcc = 0 going into high impedance. If voltage present at digital inputs - no worry, seems like there is no ESD diodes at all, or ESD protection is disconected somehow and doesn't interfere .

 So, A and B both have no connection to COM.

I also bought ts5a3153 to play with, switch is very similar to 23159 but single and has EN-able pin. Case Vcc = 0  is the same as deactivate EN pin - everything become disconnected, inputs, outputs and logic control input.
 

Online ch_scr

  • Frequent Contributor
  • **
  • Posts: 861
  • Country: de
Re: Help deciphering ts5a23159 power off state
« Reply #7 on: July 25, 2024, 08:47:06 am »
There is classical ESD protection with diode(s) to gnd/power rails. More modern solution seems to behave more like a zener diode,
still providing ESD protection by conducting above certain (max input) voltage, but no longer related to power rail.
This design also eliminates "limiting resistor" when driving e.g. 3.3V gate from 5V logic.
Look at e.g. 74lv1t34 datasheet for more detail on that.
 

Offline oliv3rTopic starter

  • Frequent Contributor
  • **
  • Posts: 283
  • Country: nl
    • Rigol related stuff!
Re: Help deciphering ts5a23159 power off state
« Reply #8 on: August 04, 2024, 04:57:31 pm »
Thank you for helping me figure this out! So why did they not draw it as I've drawn in red here? Or even remove the entire red bit.



Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf