TPS3700EVM-114: Battery undervoltage protection massive voltage drop when connecting a load

Hi Abe,

Can you draw out your schematic?  On the EVM board for the TPS3700, can you please tell me what are the jumper settings?  Have you checked to see if the boost regulator may be the issue?  Try loading the boost regular by adding a resistor (maybe 10K) from Vout to GND and measuring the voltage.  For the TPS3700, can you use a power supply and make sure the part is properly working and that you are getting the right window voltage detection.

Ben

Hello Ben,

Here is a schematic:

I am sure that my converter is working as I have verified it works through testing. For the evaluation module: J10 shorts pins 2 and 3 while J11 shorts pins 1 and 2.

You will notice that I altered the OUTA output to trigger the enable pin on my TPS61030 as I had no success by connecting it to VBAT directly. Please, can you verify whether I am using the EVM board correctly??

Hi Abe,

What is the voltage of VBAT?  

For the TPS3700, can you use a power supply and make sure the part is properly working and that you are getting the right window voltage detection?

According to my calculations, TPS3700 Vmon(UV) = 3.25V and Vmon(OV) = 3.94V.  Are you able to see the threshold on the bench/scope?

My suggestion is to evaluate the TPS3700 by using the oscilloscope and a power supply to ensure the TPS3700 is properly set up for your system.  From your schematic, I see that you are using the TPS3700 to drive the EN pin.  Please capture some scope photos (ina+, inb-, and OUTA) so I can understand if the TPS3700 is functioning correctly and a few photos of the output of the boost regulator.

Is the output of the boost regulator 5V?

Ben

Ben,

My VBat can vary between 2.5V and 4.2V.

I have used the power supply and the part is working properly.

I unfortunately do not have access to an oscilloscope at the moment due to COVID measures but here is how I require my system to work.

The battery should be allowed to go as low as 3.2 V and as high as 4.2 V and not flicker on and off at the undervoltage threshold. I am not sure how you got your Vmon values although I did take into account the resistor divider for ina+ where R1 = 604kOhm and R2 = 83.5kOhm [(69.8 + 13.7)kOhm] where if I provide 3.2 V at the input the voltage at OUTA would be 0.389V < 0.4 V threshold and 3.3 v at the input would give OUTB 0.401 > 0.4V threshold. All my measurements are theoretical and any measurements I have mentioned on previous forums are with a digital multimeter.

Can you please verify whether my measurements are correct??

Yes, the output of the boost regulator is 5 V when the input is above 3.3 V.

Thanks,

Abe   

Hi Abe,

Since you are only using OUTA, the calculations are as follows:

400mV/(83.5k + 687.5k) = 3.29V (rising threshold)

394.5mV/(83.5k + 687.5k) = 3.248V (falling threshold)

It seems your calculations are correct.

Ben   

Hello Ben,

I have done a little bit of testing with the current configuration of the board and I have noticed a few things. When I attach a 1 A load to my boost converter, the voltage drop at the OUTA triggers the Enable pin on my TPS61030 earlier. This makes my actual thresholds when my loads are attached as the following:

- Rising threshold = 3.47 V

- Falling threshold = 3.36 V

This explains why my comparator never reaches the falling threshold when my initial battery voltage is around (Vmon =) 3.3V. It seems that whenever the enable pin is lower than 2.7 V (ie the OUTA pin) the enable disables the load with the output of the converter giving 0.416V and as we get closer to the preset rising threshold (3.29V) the OUTA = 15mV. Is there a way to ensure that my comparator would still allow for the load to continue drawing from my battery till it reaches my preset falling threshold 3.248 V???

Thanks,

Ibrahim

Hi Abe,

Something doesnt make sense.  Why would the rising and falling threshold go up with you place a load on the boost output?  Is the ground on TPS3700 being "lifted" when 1A is being sourced from the boost regulator?  From your schematic, you are using PGND for the TPS3700.  I am not sure how are you setting up the GND connection on your eval board.

Ben 

Ben,

I have provided the values of the battery voltage for when the output disconnects and reconnects while the load is connected.  I am not sure why the Enable pin of the regulator is being triggered as early as that but the load is connected to +5V and PGND while the battery is connected to VBAT and PGND (VDD and GND on the eval) which is also what the positive and negative terminals of the 5v regulator input is connected to.

I suspect that the instantaneous voltage drop caused by connecting the load is setting VDD to go low which causes OUTA to go low enough for the enable to be set to LOW and triggers the load to disconnect and remains this way as long as the load is connected. If I disconnect my load, it functions appropriately. I require the device to function the same way as it would with or without the load while the enable is connected. 

The culprit seems to be the TPS61030 chip or it may be the voltage drop at OUTA of the TPS3700EVM-114 discussed in the initial post. Any ideas on how to fix this??

Abe