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.

TPS3606-33: switching threshold inquiry

Part Number: TPS3606-33
Other Parts Discussed in Thread: TPS3613-01,

Hello,

I am using the TSP3606-33 IC in a battery powered design. For my design the part's Vdd is connected to a 3.3V regulated supply that is derived from a main battery pack. My application has a secondary "backup battery pack" that is regulated down to 3.3V as well. This regulated battery backup is connected to the part's Vbat pin.  I want the TSP3606-33's Vout to switch to the regulated Vbat source when the Vdd pin falls just below 3.3V.

Page 5 of the datasheet has the following statement "Vbat only connects to Vout (through a 2ohm switch) when Vout falls below Vswn and Vbat is greater then Vdd." I read in the datasheet that Vswn is typically 2.93V + 2% which is ~2.98V. So this would seem to indicate that the part will not switch until Vout < ~3V. The behavior I am seeing with my design is that Vdd must fall to < 3.0 volts in order for the switch to occur which agrees with my understanding of the datasheet. This however is problematic for my design because Vout from the TSP3606-33 IC powers circuitry that requires ~3.3V.

So I want the TSP3606-33 IC to switch when Vdd falls just below 3.3V. I am not clear how this part can achieve this behavior however. I am happy to share my application's schematic if that would be helpful.

I appreciate any help possible.

thank you.

-Sean

  • Sean,

    Your understanding is correct. This is because the common undervoltage trip point for a 3.3V nominal rail is usually ~7% to ~11% below the nominal voltage rail thus a 2.93V threshold. The 2.93V threshold is fixed in this device variant. You will need to switch to the adjustable threshold version of this device called TPS3613-01 which uses a 1.15V threshold at the SENSE pin and you can use external resistor divider to set the threshold to be any higher voltage +- 1.7% due to device accuracy. For a tighter tolerance, you want to use more accurate resistors. So use SENSE to monitor the voltage at VDD with a resistor divider as shown on first page of the datasheet and when SENSE pin drops below 1.15V, the device will switch over to the battery back up.

    http://www.ti.com/lit/ds/symlink/tps3613-01.pdf

    Please let me know if this accomplishes what you want. Thanks!

  • Hi Michael,

    Thank you very much for the information and confirming that my understanding of the functionality of the TPS3606-33 is accurate. I will review the datasheet for the TPS3613-01 part and I appreciate you suggesting that alternative part for my application. It appears to be a more appropriate part for my application given the switching threshold I am trying to achieve.

    Would it be possible for me to use the manual switch control pin and external comparator circuitry with the TPS3606-33 part to achieve the functionality that I am looking for? The reason I ask is because my lab has nearly 60 of the boards that implement the TPS3606-33 part. This design is quite old and is something I have inherited and I have only recently discovered the issue with the 2.93V switch threshold. So if at all possible I would be willing to hack these boards to try and achieve the switching behavior that I am looking for. But I suspect that will not be possible and it will be necessary to re-design the board with the proper part. But I did want to ask if it might be possible to use the manual switch control pin on the TPS3606-33 part to achieve the functionality and switch threshold behavior that I am looking for in this design.

    Thank you again for your help and time.

    -Sean

  • Sean,

    That option would work if you can control the MSWITCH to connect to VDD precisely when you want, just below 3.3V, but this is not acheivable with the device by itself and will require additional components I believe. One solution I can think of is to utilize the PFI, PFO pins. The PFI pin connects to an internal comparator with reference voltage of 1.15V. Using resistor dividers, you can connect PFI to VDD to monitor VDD and trip PFO right when VDD drops below desired threshold of just under 3.3V. The PFO pin drops to logic low when PFI drops below 1.15V so the PFO pin would have to connect to a switch that pulls MSWITCH to VDD when this occurs. You will have to determine the better trade off between re-designing the boards, or adding the resistor divider to PFI and a FET plus pull-up resistor at the PFO output.

    The TPS3613-01 is the same package and number of pins with slight differences in pinout. Depending on the board, the board could maybe be re-worked for drop in replacement of TPS3613-01 as well.

    Please let me know if you need assistance with any of these options and I am happy to help

  • Hi Michael,

    I agree with your suggestion and usage of the PFI and PFO pins along with Mswitch to try and achieve a switch over voltage closer to ~3.2V for the TPS3606-33 part. Luckily the design I am working with does use the PFI and PFO pins along with a resistor divider network to set the trip voltage for the PFO output. There is also a mosfet on the board who's gate is driven my PFO from the TPS3606-33 part.

    I have experimented with connecting 'Mswitch' from the TPS3606-33 to the drain of a mosfet that is part of the design. The gate of this mosfet is connected to PFO and the source is connected to ground. The drain is pulled high to Vout from the TPS3606-33. The pull up resistance is provided by an external module and I don't know the exact value. So when PFO is high the drain of the mosfet and 'Mswitch' are both tied to ground. When PFO goes low 'Mswitch' should be pulled high. However I am noticing that 'Mswitch' is somehow loading down the drain of the mosfet and preventing it from raising to Vout. I checked in the datasheet but could not find the input impedance for 'Mswitch'? If I disconnect 'Mswitch' from the drain of this mosfet it functions properly and goes high when the gate (which is driven by PFO) is low. So I think the combination of the 'Mswitch' input impedance and the pull up resistance connected to the drain of the mosfet is creating a resistor divider network that prevents the drain voltage from rising to Vout.

    thank you

    -Sean

  • Sean,

    The pull-up resistance must be low as to provide a strong pull-up for MSWITCH. Also, the MSWITCH section in the datasheet on page 6 states to connect MSWITCH to VDD for switching to battery mode rather than connecting MSWITCH to VOUT. Can you confirm if MSWITCH is connected to VDD through the pull-up resistor and not VOUT?

  • Hi Michael,

    That is a very good point you make regarding the fact that MSWITCH must be pulled high through a strong pull up and that it also must be taken high to VDD. Unfortunately the design I am working with has the mosfet's drain connected to a pull-up that goes to VOUT. So I don't think it will work for me to hack this design to try and make the TPS3606-33 switch from Vdd to VBAT when Vdd dips just below 3.3V. I think I am going to have to re-design the board and use the correct part.

    Thank you very much for your help and time with answering my questions.

    cheers,

    -Sean

  • Sean,

    We are here to support you anytime. Good luck with your design and let us know if you need anything else!

  • Thank you Michael and I appreciate your help and support.

    -Sean