LM25116EVAL: LM25116 reconfiguring the output when receiving different design values from the WebBench design tool and the LM(2)5116_quickstart spreadsheet.

Hello John,

The output capacitors on the EVM are only rated for 6.3V, which is likely why you are not seeing the correct output voltage when configuring the system for 15Vout.

Regards,

Hardit Singh

The output caps were changed to 35V parts with no effect on the output. With the new caps and a 28V input the output gets to 8.8V.

Hello John,

Thanks for the clarification. Do you have any waveforms from the operation that you can share? I am specifically interested in:

- Inductor current

- SW node voltage

- VCCX pin Voltage

- VCC pin Voltage

- FB pin Voltage

- Enable pin Voltage

- UVLO pin Voltage

Can you also clarify if any modification to C14 was made, and if the Vout you are seeing is at no load.

The component modifications you made seem ok.

Regards,

Hardit Singh

Hi Hardit,

This is a screen shot of the SW node waveform vs the Vout level (showing 8.859V). Cursors show 350KHz switching.

While I don't have an easy way to obtain an inductor current reading, the rest of the nodes show as follows:

Vccx pin voltage = 0V (C14 is as delivered, 1uF)  

VCC pin = 7.25V

FB pin = 850mV

Enable pin = 13.0V

UVLO pin = 1.43V

Let me know if yo'd like to see other screen shots, or multiple signal groupings.

Hello John,

Thanks for sharing the measurements and waveforms. It looks like the device is regulating the output voltage, but not to the expected level. Can you check if the output voltage moves at all with 100-200mA of load?

I would also be interested in knowing if adjusting the FB resistors will change the setpoint.  As a sanity check, can you measure the resistance of the populated top and bottom feedback resistors?

Best,

Hardit Singh

Hi Hardit,

Placing any amount of load crashes the output. Today, the circuit started with a no load output of 9.18V, (again this was set for 15V out). Still, I measured the output at 50mA, 100mA and 150mA and the voltages were 3.32V, 2.48V and 1.62V, respectively.

Here are the SW node and Vout waveforms for open circuit and 50mA captures:

Measuring the actual values of the top and bottom FB resistors gave 13.69K and 1.2K which, from the spreadsheet calculator are correct.

Do you have any ideas what could be happening? If you'd like, I can send you my eval board to check out.

Thanks,

John

Hello John,

Thanks for sending that additional information. Your circuit setup seems ok, but there are a few other possibilities I would like to explore:

1) Can you try powering the EVM from a benchtop DC DC supply? I am curious if this is a limitation from the source? Can you confirm that the input of the Vin connection to the EVM is seeing an expected voltage when powered from the battery pack?

2) Can you please share the part number for the output capacitors you are using? Vout falling with increased load is another indicator of not enough output capacitance, so I want to double check your components.

3) Can you provide a voltage reading at the Comp pin of the controller across load conditions? This will give me more insight into if the problem is coming from the stability of the control loop. Looking through the excel calculator you filled out, and some additional analysis on my end, this is unlikely, but I would like to double check.

Regards,

Hardit Singh

HI Hardit,

The eval board has only been supplied by a benchtop supply set to 30V with a 6A current limit - it shows no effect when the eval board starts up.

The output caps were replaced with two 330uF 35V caps (Panasonic EEU-FC1V331L). Since we are not reaching the programmed output with very light loads, it doesn't seem to be output cap related.

Here are the scope shots of the COMP pin and Vout for No Load and 50, 100 and 150mA Loads (Comp signal is offset by ~5V to get some resolution):

Thanks,

John

Hello John,

If you can try out a few more things for me, that would be greatly appreciated. I am also working on getting an EVM setup on my end to try and replicate the setup you have, but it will take me some time. Thanks in advance for your patience on this.

More things to check:

- What is the current limit set to on the benchtop supply you are using? I want to double check that you are not input current limited.

- Try removing the 464kΩ resistor in series on the Ramp pin.

- Please measure the value of the parallel 10mΩ resistors on the R11 footprint. I want to confirm that the device is not in current limit all the time.

Regards,

Hardit Singh

Hi Hardit,

The problem is the Rramp resistor, or rather the series placement of it. The LM25116-EVAL schematic shows no Rramp position, but the Excel sheet diagram shows Rramp should have been connected to VCC as a pull-up. I missed that. Still, simply removing the resistor I had in series with Cramp made the supply work.

Can you explain the note in the spreadsheet that explains "Rramp is only necessary when the output is >7.5V." Should I add the Rramp + Vcc connection to my original design ( it had only the Cramp Cap and didn't work at 24Vout)?