• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 634 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

TPS22810: Pull Down on ENABLE pin

TPS22810: Using a mechanical slide switch to control the Enable Line

Charles McClelland
Prodigy 250 points
Part Number: TPS22810
Other Parts Discussed in Thread: TPS2121, TPS2120

I would like to use a mechanical switch to enable the switch to turn on my micro controller based IOT project:

First Input - LiPO battery 4.2V Max, 1A max - bidirectional 

Second Input - DC Input 9v Max, 1.5A max - unidirectional

I have used these parts in previous designs and am confident about the load switch's capabilities.  However, in the past, I have controlled the Enable line using a logic gate.  In this application, I want to use a simple slide switch and have the following scenario:

- The switch will be SPDT so it will either be connected to ground or one of the power lines and will be break before make (non-shorting)

- I am looking for the minimum cost / components needed to debounce the switch during that brief period while between poles.  

Questions:

- Do I need to debounce the switch, can I use the TPS22810's slew rate as an alternative?

- If so, can I use a simple RC filter?

- If that is not enough, do I need to add a Schmitt trigger?

My device has a fair bit of input capacitance so a very short period of fluctuation is likely OK. But, your advice or tips would be appreciated.

Thanks, Chip

Thank you,

Chip

  • 0
    TI__Mastermind 32650 points

    Hi Chip,

    I'm not sure I understand what you mean by "can I use the TPS22810's slew rate as an alternative?" If the input goes low, the output discharge circuitry will be turned on. This would reset the output to 0V if it is being used.

    I would expect you to require some sort of filter or debouncing methodology lest the switch will incorrectly/inconsistently turn on.

    The amount of this filtering would be dependent on your system. I do not expect a Schmitt trigger be required but as I said, your system may be more susceptible than what I am familiar with.

    I also have a follow up question on your use case. Have you taken a look at TI's PowerMUX portfolio?

    http://www.ti.com/power-management/power-switches/power-muxes/overview.html

    I think a power mux might be a more integrated solution for your application. Is there a specific reason you are using 2xTPS22810s?

  • 0 in reply to Shreyas Dmello
    Prodigy 250 points

    Shreyas, 

    You had two questions:

    1) What did I mean by slew rate - I guess I was referring to the configurable rise time.  I was wondering if a right time longer than the short "floating" period as the switch went from one pole to the other might smooth the effect of the noise the switch would introduce during that period.  

    2) On your next question, you raise a good point.  I need to look at your MUX offerings but here is why I am a bit hesitant to make a change on this revision:

    - I have fielded a number of devices (~100) using the TPS22810 and am comfortable with the design.  

    - There are three distinct power scenarios for this device:

    1) LiPO battery and DC-IN which can carry the full load and charge the LiPO

    2) LiPO battery and a 3W solar panel.  The Panel may or may not be able to the carry the load / charge based on available sunlight and its voltage can drop quickly.

    3) DC-IN or LiPO only

    Unless there is a good application note on this, I would need some time to test / validate this solution works.  If so, please point me to it.

    3) In reading the datasheet on page 15, I found this text: "The internal de-glitch delay on EN/UVLO falling edge is intentionally kept low (2.5 μs typical) for quick detection of power failure. For applications where a higher de-glitch delay on EN/UVLO is desired, or when the supply is particularly noisy, it is recommended to use an external bypass capacitor from EN/UVLO to GND."  So, perhaps all I need to do is add a capacitor between EN and ground.  If so, any advice on the value (don't know the internal R to figure out the time constant).

    Thank you for your help.

  • 0 in reply to Charles McClelland
    TI__Mastermind 32650 points

    Chip,

    1. I have no data that relates to this specific use case but I doubt the rise time would help in this situation. I think going down that path may not be the best option.

    2. While I do not have any specific collateral on those topic, i would encourage you to take a look at TPS2120 or TPS2121 which are higher voltage, capable of driving 3.5A to 4.5A in a small form factor. While quick output discharge is not available on these devices, the slew rate is configurable and the user can set a current limit. TPS2121 is also capable of "fast switchover" times as low as 5us. *Please refer to the datasheet to see what TI defines as "switchover time"

    3.I think the capacitor on the EN pin is the best workaround with the TPS22810. The capacitance required would once again be dependent on how "bouncy" your mechanical switch is. On the other hand, because the bounces are transient I do not think this needs a heavy capacitance. It will require testing on your real system to correctly configure. Off the top of my head I would start with 10nF and work my way up.

    Thank you for choosing TI Power Switches in your designs.