TPS27081A: Fails to turn off completely

Hi Brian,

The device is turning fully off, but there is no reason for the output to be pulled down, especially with a purely capacitive load.

To elaborate, the function of the switch is to disconnect the power supply from the load. Once the switch is turned off, the connection between these two is broken and the capacitance is left on its own. Without a load to discharge the capacitance, there is no reason for the capacitance voltage to lower at a fast rate.

If you would like a load switch with output discharge capabilities, you can take a look at the TPS22918. This load switch is not only a more integrated solution, but it also features a Quick Output Discharge (QOD) feature that pulls the output of the switch down to ground when it is turned off. That way, when there is a purely capacitive load on the output, it will get discharged down to 0V when the switch is turned off.

Thanks,

Alek Kaknevicius

Hi Alek,

I understand your explanation and agree that this would explain why the voltage doesn't drop down to 0 when switching from ON to OFF.

I may not have described the issue properly, the main problem I'm experiencing is on startup.

On startup, the load capacitor is completely discharged (voltage across it is 0V) and pin 5 (ON/OFF) on the TPS27081ADDCR is pulled low to 0V.

When the device turns on, 5V is applied to pin 4 (VIN) on the TPS27081ADDCR and pin 5 (ON/OFF) is still pulled low (it has stayed at 0V for the entire time), pins 2,3 (VOUT) then go up from 0V to about 2.2V.

Do you know what would cause this issue?

Thanks,

Brian

Hi Alek,

Thank you for your response, that explains my issue.

I will use the TPS22918 in future designs.

Thanks,

Brian

Hello Alek,

I don't know if this is the right place to ask, but do you have an alternative solution to using the TPS22918?
The only issue with using this part is that I need to switch up to 2.75A to the load and the TPS22918 is limited to 2A.

Do you know of any alternatives?

Thanks,

Brian

Hi Brian,

We have several alternatives depending on what you're looking for. A few questions for you:

1. What is your VIN voltage?
2. Can you accept CSP devices? Do you have a preferred package?

The TPS22975 is one of our popular devices, and it is actually smaller than the TPS22918. Do you think a device like this would work?

Thanks,

Alek Kaknevicius

Hi Alek,

My VIN voltage is 5V.

My preferred package would be anything with leads which is part of the reason why the TPS27081ADDCR in the SOT-23-6 package was selected in the first place.

Packages without leads are also acceptable, but I'd like to stay away from CSP/BGA packages for now.

A quick Digi-Key search shows that there are two versions of the TPS22975, the TPS22975DSGT and the TPS22975NDSGT. The Digi-Key parameters suggest that only the TPS22975DSGT has a load discharge feature. Is it correct then that only this version of the device would solve my issue?

Thanks,

Brian

Hi Brian,

You are correct, the TPS22975DSGT is the device with the QOD feature and will keep the output of the switch down to 0V when the device is turned off.

If leads are a strong preference for you, then we also have the TPS22958 in an 8 pin leaded package. Same as the TPS22975, you would want the version without the "N" at the end.

Thanks,

Alek Kaknevicius

Hi Alek,

Thank you very much for your help. I've decided on using the TPS22975DSGT in my next design.

I had a design question relating to this load switch that I was hoping you could help me with. I'm not sure if this is the right place to ask you. I don't know if it was possible to send you private messages.

If you have any tips on where the right place to ask this question would be, let me know and I'll ask it there.

For my previous design with the TPS27081ADDCR I had no input capacitor on the input to the load switch and I had a 100uF load cap on the output. The switch is used to switch power to a 5V USB controlled device. The TPS22975DSGT datasheet recommends a 10 to 1 ratio for input capacitor to load capacitor to prevent reverse current. That seems a bit excessive with a load capacitance of 100uF. I am not sure how the input capacitor to the switch affects the device compared to the output capacitor, whether it acts as bulk capacitance for the device or not.

Could you recommend values for an input capacitor and a load capacitor that would both provide similar performance  to a 100uF bulk capacitor and still prevent the reverse current issue?

Thanks,

Brian

Hi Brian,

I can answer this question here no problem. 

To be honest, the 10 to 1 input to output ratio in our datasheets is very conservative. The purpose of this recommendation is to ensure that the power supply can handle the inrush current transient caused by the output capacitance during turn on. However, with the controlled rise time of the TPS22975, the stress on the supply will be minimal and a 10 to 1 ratio is likely not needed.

To protect against reverse current, it is recommended to have COUT>CIN. Otherwise, if VIN suddenly drops then VOUT may take longer to discharge and under the brief VOUT>VIN condition reverse current will flow through the body diode of the device. If you do not expect VIN to drop suddenly during your application, then this would not be an issue.

Thanks,

Alek Kaknevicius

Hi Alek,

Did you mean it is recommended to have CIN>COUT? In the datasheet it is suggested that CIN should be greater than CL to prevent reverse current.

If I move the 100uF output capacitor from the output of the load switch to the input of the load switch, will it still act as bulk capacitance for the device that is getting power switched to it (the load)?

I just want to make sure I can prevent the reverse current issue while the load still has appropriate bulk capacitance.

Thanks,

Brian

Hi Brian,

Good catch, my mistake. It is CIN>COUT. Also, If the capacitance is on the input of the load switch and the switch is turned on, it will act as bulk capacitance for the downstream load. However, it would still be recommended to keep a small capacitance close to the input of the load.

Thanks,

Alek Kaknevicius

Hi Alek,

Thank you for the response, do you have any recommendation for the value for the small capacitance close to the load?

The original bulk capacitance was 100uF. If I move that to the input of the load switch, is 10uF a reasonable output capacitance? Or can it be even smaller?

Thanks,

Brian

Hi Brian,

10uF is a good value, but it would really be dependent on your system and the performance you get when switching on/off the load. If the performance at 10uF works for you, then you could always try 1uF as well.

Thanks,

Alek Kaknevicius

Hi Alek,

Yeah, I was just wondering if you had any tips on that. We know that with the old load switch 100uF of bulk capacitance works when switching on/off the load.

If I move that 100uF bulk capacitor to the input of the new load switch, can you think of any reason the load would have issues with a 10uF capacitor added to the output of the load switch?

Thanks,

Brian

Hi Brian,

Unfortunately this is highly dependent on the behavior of the load. If the load has several fast high current transients during operation, then more capacitance would be needed on the output of the switch. While it is small, the load switch still adds a slight amount of resistance in between the input and output capacitances, so the amount of capacitance you need on the output is really dependent on how the load is expected to operate. If the output does not have sudden, high current transients then a lower capacitance can be used.

Thanks,

Alek Kaknevicius

Hi Alek,

Thank you for the explanation and answering all my other questions.

I appreciate your help.

Thanks,

Brian