TPS61253AEVM-803: noise is emitted from the circuit at 5 ohm load.

Hi Tristan,

Did you test the 5 ohm load on TPS61253AEVM or with a board designed by youself?

I tested the 5-ohm load on the TPS61253AEVM. When I do the tests I'm outputting around 200mA and around 1-2V. I have connected the board correctly by using the user guide. I am not sure why this would be happening, I wanted to see the 1A output with the 5-Ohm load but the board emits a high frequency and seems to not be able to handle it. 

Hi Tristan,

This device has softstart current limit which is lower than normal boost mode current limit. I would suggest you apply the 5-ohm load after the softstart finished. 

Do you know what the softstart current limit is? The Lipo battery is 2900/3000mAh, is that fine? Also, could you elaborate on applying the 5-ohm load after the softstart is finished?

Hi Tristan,

If the Lipo battery voltage is higher than TPS61253A UVLO 2.3V during startup, it should be OK.

Below is the explanation of startup.

I've read this and have a better understanding of this, I still have a few questions though. Why does this only happen to lower loads rather than larger loads? When I applied 200 ohms to the load it worked well. Is this because at larger loads there is less current? Also, I still am uncertain what you mean by applying the 5 ohm load after the soft start has finished. 

Hi Tristan,

With lower loads, such as 200ohms, it means the output current (Iout) is smaller compared to 5 ohm load. During startup, the device needs to charge the output capacitor to the setting voltage. The load resistor will also flow some current during this phase. That' why we don't hope the load resistor draw too much current.

I mean after you see the output voltage reaches 5V, then you connect the 5ohm load to EVM output.

Zack,

I attempted what you asked me to do and that did not resolve issue. I let the output voltage become 5V flat and as soon as I applied a 5-ohm load the voltage dropped to 2.2V and noise was emitted from the circuit. I am still confused on why this is happening because in the datasheet for this chip it says this specific circuit can output up to 1500mA. There has to be a deeper issue here. The issue is not the startup at all, it is something completely different. 

Hi Tristan,

Could you measure the input voltage with DMM or capture the input voltage waveform, SW pin, Vout pin waveform with oscilloscope? Please make sure the input Lipo battery has enough current capability. For this application, to output 5W power, the input average current is around 5W/0.9/3.7V=1.5A. In some cases, if the Lipo battery inner resistance is too large or power cable is too long and thin, the EVM input voltage will decrease to a very low level. So please check the EVM input voltage first.

If the EVM input voltage is correct, can you take a picture of the whole bench? 

Zack,

I can try and get those images for you in the morning. On the other hand, when I measure the input voltage, the original value is 3.9 but when applied to the circuit it drops to about 3.3V. The internal resistance is around 60-80 ohms and I do believe the battery has enough current capability. I do think it's possible that the wires are too long/thing causing this issue. I believe the internal resistance is related to the sectional area and length of the wires. 

Hi Tristan,

Have you got the waveforms?

Zack,

I am currently out of town and do not have access to my lab. I should be back by next week, if you could keep this case open for me I'd appreciate it.

Best,

Neima

Hi Tristan,

Sure. I'll wait for you feedback.

Zack, 

I should be back later this week and get those scope shots for you, but I wanted to ask you a question. Before I left I was able to get the cirucit to not have any noise. My output load is a 25W 5.6 ohm resistor. The output voltage was around 4.5V and the output current was around 667mA. This is a step in the right direction but the resistor was becoming very hot. I believe the load is the issue causing the output voltage/current to not be what it is intended to be with that load (5V & 1A). It is surpring that the resistor is getting that hot and causing this issue because it has a power limit of 25W. If you have any insight I would appreacite it. 

Best,

Neima

Hi Neima,

Firstly please check the voltage and current is measured correctly. Per your description, the output voltage is 4.5V. The output load resistor should consume 4.5V/5.6ohm=803mA, not 667mA. I would suggest you take a picture of the whole bench hardware setup.

Always use short, thick power cables to connect power supply, load to the EVM. Use DMM to measure the board input and output voltage & current please. 

Zach,

I will have images of my multimeter for you tomorrow sometime in the morning. I know I measured it correctly. The output current was 660mA and the output voltage was 4.5V. That was why I was confused because the output current should have been higher and the resistor should not have been heating up. When I measured the input current it was around 1.1A and the input voltage was 3.8V. Instead of a power supply, I am using a lipo battery with short 20 gauge wires.

Best,

Neima

Zach,

I have been able to pretty much fix the issue at hand. With a load of 5.6ohms, I have been able to get up to 826mA. I believe it is not around 892mA due to the setup I have (breadboard/wires). The load resistor does heat up quite a bit but I do not think this will happen If I were to reconstruct this on a PCB. 

Best,

Neima 

Zach,

Quick question. Is the switching valley current limit for this part 4.55A? 

Best,

Neima

Zach,

Thank you. I wanted to ask one more thing. After doing calculations I came up with solutions that at 3.7V input, the output current limit should be around 2.81A (Duty cycle of 33.5%). At 3V input, the output current limit should be around 1.375A (Duty cycle if 69%). With this design for the evm board I am using, it says that the max load current is 1500mA. What I am asking is, is this possible for the output current limit for 3.7V input to be 2.81A when the max load current is 1500mA? 

Best,

Neima

Hi Neima,

You can apply more load current than 1.5A. But since this part switching frequency is 3.5MHz, the efficiency will be really low at higher load current conditions.

And I'm afraid the IC temperature will be really high. 

Zach,

Thanks for your help. Out of curiosity, do you have a booster chip that can output 18W or more? A Vout of 5V and current output of around 3A?

Best,

Neima 

15W I mean, 15W to 18W or more....

Hi Neima,

For 3.7V Vin, TPS61022 could output 5V/3A 15W power.

But as your input is a Lipo battery, what's your target minimum input voltage requirement of output 15W or 18W?

Zach,

What do you mean? Do you mean what's the lowest I would like to go on the input voltage to still output what I need? And thank you for showing me this chip, it is exactly what I was looking for. 

Best,

Neima

Hi Neima,

Yes. Since the boost converter is limiting the input current, with lower Vin, the maximum input power will be lower. So output power is also lower. You have to consider the worst case (minimum Vin) if still want to output 15W or 18W.

Zach,

What is the lowest I can go before I lose that power? From reading the datasheet the lowest would be 2.75V? And once it got lower I could still output over an amp of current correct?

Best,

Neima

Hi Neima,

For TPS61022, the switch valley current limit is 6.5A minimum. Assume maximum input average current is 6.5A and efficiency is 90%, the lowest input voltage would be 18W/0.9/6.5=3.07V. It is really low for your Lipo battery, right?

The lowest I can go before not outputting 15W or 18W is 3.07V? I would not say very low. The max on my Lipo battery is 3.7V, 3000mAh, or 4000mAh. Would you recommend I start with a higher input voltage? Also, I appreciate your help. 

Hi Zach,

I was only confused because you said that 3.07V might be to low for the lipo. What did you mean by this?

Best,

Neima

Hi 

Jusr let you know taht Zack is out of office. Please allow us reply next week. Thanks for your kind understanding.