• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 16 subscribers
  • Views 200 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TMUX7208: Turn off TMUX for low power usage

Johan Christiansen
Prodigy 10 points
Part Number: TMUX7208
Other Parts Discussed in Thread: OPA991, TMUX8108, TMUX7308F

We are using two TMUX7208 as the cell selector mux in conjunction with an OPA991 for measurement of cell voltages in a battery-string. We use the single supply mode of the TMUX and have it hooked up to the top cell-voltage (+33.6V) and ground (0V). The OPA uses the same supply.

Due to the rather high quescent current of all three devices, we want to be able to shut them down along with the MCU power domain on the battery pack.

We have looked at two solutions.

A) A high side switch: This removes +33 V from the TMUX VIN. But according to the datasheet, S1-S8 must not exceed VIN+0.5V. There is nowhere i can figure out how the protection input works. So i have actually fitted diodes from each S1-S8 to VIN to ensure this is never exceeded. Therefore the high-side switch will only work, if the S1-S8 can be allowed to exceed VIN when the device is not powered. (And i remove the diodes)

B) A low side switch. Cutting the GND off using a low side FET. This has the simple advantage that we can use a singe low-side TTL level N-FET easy to operate from the MCU. Hoever again this is questionable, if it's allowed.

I hope you can help us with some technical knowledge on how to disable these devices.

  • 0
    TI__Mastermind 23593 points

    Hello Johan,

    Sounds like you're trying to remove the power supply (VDD) but still have the voltage inputs at S1-S8. More specifically it sounds like the two options you're presenting is to float the VDD or the GND of the device and still apply inputs at the I/O pins. I wouldn't recommend floating either. As you mentioned, there shouldn't be any signals exceeding VDD + 0.5V on the S1-S8 I/O pins. There is an ESD diode internally and by exceeding these levels, you'll forward bias it and can actually end up back powering your device and other devices downstream. This can potentially cause harm to the device and other devices that may not be able to support I/O's that high when they are powered down. The protection feature you're looking for that would help mitigate this is powered-off protection.
    Here are some useful resources with information regarding this : 

    https://www.ti.com/lit/an/scda015c/scda015c.pdf?ts=1632930770466&ref_url=https%253A%252F%252Fwww.google.com%252F#:~:text=Protected%20Switches&text=Switches%20with%20powered%2D%20off%20protection,even%20when%20VDD%20%3D%200%20V.

    https://training.ti.com/ti-precision-labs-switches-and-muxes-simplify-power-sequencing-powered-protection

    Powered-off protection will have internal circuitry that will allow for the device to support inputs on the I/O lines when the power supply is 0V or floating. In the VDD = 0V or floating condition, the source pins will remain in a High-Z state. 
    We do have some new 8:1 devices that will support the voltage it looks like you're working with that have this build in feature. Admittedly the supply current will be higher when powered on, but this will allow for them to powered off when not in use.
    The TMUX7308F and TMUX8108 will be 8:1 devices that will support this. TMUX7308F will come with fault protected features as well, which cause the Sx pins to become high-impedance as well if there is an even in which the I/O exceeds the VDD level (plus some Vt threshold of 0.7V for this device).

    Let me know if this helps or if you need any more help here.

    Thanks,
    Ramiy

  • 0
    Prodigy 10 points

    Thank you very much for a quick response.

    We actually have already purchased TMUX7308, and was going to use it for this design. As it's better on all accounts, except for one little detail. It's Ron is 300 Ohm, whereas the TMUX7208 is only 4. This is important for us since we are planning on using it for an active battery balancing system moving energy from one cell onto a big capacitor and then using it to charge another cell. The ESR of the battery is 0.08 Ohm and the ceramic capacitors almost nothing. So a huge portion of the energy loss would come from the MUX.

    I have not yet seen the TMUX8108 but it seems to be placed on the middle. It has the turned off protection, and yet only 35-50 ohms compared to the 2-300 ohm of the TMUX7208. Maybe it's the winner here.

    One final note: if we had a series diode on the VDD input, and were to float the ground. I don't se any reason this shouldn't be safe. As the chip shouldn't draw any power and the VDD cannot back-power anything. I don't have a schematic of all the protection diodes so it's a difficult to guess what would happen.

    Also, can you comment on the OPA911 capabilities and turning that off? Or is that a new question under another forum?

  • +1 in reply to Johan Christiansen
    TI__Mastermind 23593 points

    Yes, that was my thoughts exactly when I mentioned these devices as well actually. TMUX8108 sort of is a middle ground that offers benefits of both devices. You'll lose the fault protection aspect of the 7308F but will still maintain powered off protection.

    The external diode should work for preventing the power supply back powering. But you would run into some issues with regards to the ESD. All I/O's have protection diodes to VDD and to GND. So by placing a diode at VDD, you would be blocking off the path during an over voltage event. 

    That being said, there is an app note that discusses how to externally protect the device from ESD events :

    System-Level Protection for High-Voltage Analog Multiplexers

    This method involves an additional series current limiting resistor plus some extra zener diodes. Drawback here is of course the extra price and complexity with adding these extra components. 


    I won't be able to provide much input on the OPA991 and I could loop in the appropriate people but I think it may be better to put that question within the Amplifiers Forum just for future searchability. Across TI our forum response times are fairly quick (We aim for 24hour responses), so I would expect a response shortly. 

    Thanks,
    Rami