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.

Need LOAD SWITCH with Low Temperature & Rise Time Requirements

Part Number: TPS22811
Other Parts Discussed in Thread: TPS7H2221-SEP, TPS22991, TPS22997, TPS22992, TPS22992EVM, TPS22991EVM



I have an application for a Load Switch, but have some requirements I need to meet listed below.

Please assist in providing a TI part number that best meets these requirements. Is a LOAD SWITCH even the best option?

Operate at ambient -54degC to 125degC (I see a lot of parts at -40DegC, can these be tested and delivered to meet -54degC)?

Operate at +3.3V input and maintain a +3.3V output.

On the Load switch parts, can the Bias voltage and the INPUT voltage be from the same source and be applied at the same time?

Drive a capacitive load up to 100uF.

Create a linear risetime that is no faster than 30 V/ms and no slower than 6.5 V/ms. Thus, to reach 3.3V, risetime needs to be greater than 110 us, but less than 500us.

Needs to provide up to 6 amps continuous.

If one part gets close to meeting these, I need to know that too. I may be able to adjust.

With any questions, please contact me at 256-658-5101

Thanks a bunch!

  • Hi Monty,

    Welcome to e2e!

    The only load switch we have that is tested down to -55C is TPS7H2221-SEP, but it is space grade/radiation hardened so it will be very expensive, especially since you will need 5 to drive a 6A load.

    Otherwise, I recommend TPS22992 for a cost-optimized load switch that has capacitor-programmable rise time, or TPS22997 if you need a higher current capability/lower on resistance that's also capacitor-programmable. However, these parts are only tested down to -40C.

    And yes, the bias and input can be tied to the same rail.



  • Patrick:

    Yes, the rad hard part wouldn't be an option.

    Some clarification on using this product and my requirements:

    a) I do realize that the datasheet says -40degC to 85 degC operate with some data at 125degC operate. What if I try and operate below the -40degC operate temperatures? Can you provide selective testing at -54 degC and ship tested parts that do meet this temp limit?


    b) It's good that the VCC Bias voltage and the VIN INPUT voltage can be from the same source, but can they be applied (switched on) at the same time? I may have to switch (using a relay) them both on at the same time. However, using your ENABLE pin may resolve this concern. 


    c) Your datasheet Figure 7-9 shows Rise Time vs CT Capacitance curve. If I'm at 3.3V on VBIAS and a higher CL (say 50 uf) can I achieve a 200-500us risetime from 0 volts to +3.3V (i.e. less than 30 V/ms)

    Thanks for the quick reply!


  • Hi Monty,

    Apologies for the delay on this.

    1. Unfortunately, we don't provide selective testing. However, we just put a new part into pre-release that is tested from -55C - 125C, TPS22991. It is rated for use up to 3A, but you can parallel two of them to handle 6A. Will this work? Please see the switching characteristics for rise time information, as it is fixed for this part.

    2. Yes, VBIAS and VIN can rise at the same time as well. Often times these pins are tied directly tied together.

    3. The below table is from the TPS22992 data sheet, there is a similar table in the TPS22997 data sheet. It looks like for TPS22992 a 220pF or 270pF capacitor on the CT pin should meet your requirement at 3.3V VIN. Rise time is nearly entirely based on CT capacitance and VIN.



  • Patrick:

    The TPS22991 looks like a perfect fit! Good timing on its release.

    On RISETIME I need to be between 110us and 500us with a 50-100uF capacitive load, then handle just over 1 amp. 

    Your "B" version of this part says risetime of 141us into a 0.1uF load. My Cload is much larger than that. How do I determine the fixed risetime with my load parameters?

    Also, a 10 to 1 CIN to Cload puts a lot of capacitance on my front end. Is that amount necessary?

    These packages are so small! It must help with heat dissipation?

    When could I get an EVAL board to explore the TPS22991's operation? Is there a SPICE model available? I just received an EVAL board on the TPS22992. Will it provide similar results to the TPS22991?

    I'm curious why the TPS22991 meets -54 degC, but the other load switches aren't rated for such. What's the difference? Materials or just testing?

    Are there 3rd parties that could provide a tested part to -54degC that you work with? If I can't make this new part work.

    Thanks for all your help!

  • Hi Monty,

    Great! Cload/Rload only minimally affects the slew rate, so that number is ok to use for your load characteristics.

    10:1 Cin:Cload is only needed if your supply will be strongly pulled to 0V during operation (if there's no input capacitance and a high output capacitance, substantial current will flow from VOUT to VIN through the body diode and may damage the device), and 1:1 Cin:Cload is only needed if your input supply can't handle the inrush current.

    Yes! This helps with thermal conductivity and better performance.

    There are TPS22991EVMs available here: Currently, there is no PSpice model. The TPS22991EVM will be similar to TPS22992EVM, but the part is much simpler and thus the EVM is as well.

    The tested Tamb temperature is determined at the definition process and throughout the design process - unfortunately, we don't work with external parties to test wider temperature ranges.

    Please let me know how your TPS22991 evaluation goes, and we can go from there.



  • Patrick:

    We have a DCtoDC converter which will be on the input of your load switch. Currently we have 100uF on its output, but it can handle much more with its soft start output circuitry. I'd say I'm good there.

    I'll get one of those EVAL boards and check it out. 

    I'm still poking at the low temp range on the other parts. In your experience, if we tested a lot of the -40degC datasheet parts, could we find any that would work at -54degC??

    Would reliability be so questioned that this is truly a non-starter? 

    Appreciate all the inputs.