TPS61022: Reverse current limit

Tony Grubman

Part Number: TPS61022

When operating in forced PWM mode, how much current can flow in the reverse direction (VOUT to VIN)? Is the reverse current limit a peak or a valley limit?

over 6 years ago

0 Zack Liu over 6 years ago

TI__Mastermind 34231 points

Hi Tony,

Can you specify the situation, like what's the Vout and where does it come from? What's the Vin voltage, is ta a power supply or battery?

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Hi Zack,

Vin will be a battery, most likely a single cell LiFePO4 (2.5-3.6V). Normally, the boost converter should be boosting up to around 4V on Vout, and the maximum draw from Vout will be about 3-4A.

I would like to charge the battery by applying a 5V power supply to the output and enabling the forced PWM mode -- the feedback pin will be well above the reference, so the boost converter should try to sink current back into the battery.

Once the charge is complete, I can disable charging by switching back to PFM mode -- the boost converter should stop switching leaving Vout powered from the external power supply.

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

You cannot use TPS61022 as a battery charger. The reason is this is a boost converter, meaning stepping input voltage to a higher voltage. When working at prebias condition, FB pin voltage is higher than Vref, so the device will work at PFM mode.

The MODE pin doesn't mean you can control the device working status like you want. It just selects the operation mode at light load condition.

You need a battery charger.

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Hi Zack,

In the datasheet, section 7.4.1, it says "The high-side P-MOSFET is not turned off even if the current through the MOSFET is zero. Thus, the inductor current changes its direction after it runs to zero. The power flow is from output side to input side." Is this not the case?

Also, this thread seems to suggest that current can flow backwards through the device: https://e2e.ti.com/support/power-management/f/196/t/797046

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

I'll ask our apps to run a bench test to see what happens. Because we're on business travel now. So we can share you the bench result next week.

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Thanks Zack, I look forward to seeing the results.

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Hi Zack,

Do you have an update on this? Have you run the bench test yet?

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

Sorry for the late response. I was still on business travel these two weeks. I'll be back office this weekend.

Another applications engineer did a simple test for me. He uses a current source meter as the input and the circuit works fine. But if using a voltage power supply, the circuit is not working correctly. I need to run the test by myself to see more details.

I would say that even the device could work with reverse current, but TI cannot promise no issue. We design the device as a boost converter. No EVM is built for reverse current circuit and no tests were run to validate the circuit.

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Thanks for the update, Zack.

Yes, it makes sense that a voltage power supply at the input won't work -- it can only source current, and is not capable of sinking current. A four-quadrant source measure unit would probably be more appropriate.

Do you have an approximate number for how much reverse current was observed in the test with the current source meter?

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

I'll be back to office this weekend. I'll do the bench test next Monday myself.

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

The reverse current limit has big variations. The bench test result shows the maximum reverse current varies from 1.3A to 1.5A.

0 Tony Grubman over 6 years ago in reply to Zack Liu

Prodigy 70 points

Hi Zack,

Thanks for all your help. I will take this into consideration for my design.

Just to confirm, 1.3A to 1.5A was what you observed in the bench test, but there can be larger variation between devices, correct?

0 Zack Liu over 6 years ago in reply to Tony Grubman

TI__Mastermind 34231 points

Hi Tony,

Yes. This is based on bench test with only one unit.

For this case, external MCU is needed to monitor the battery voltage. Once voltage is close to maximum battery voltage, MCU needs to shutdown the charging circuit.

Power management

Power management forum

TPS61022: Reverse current limit