LM27762: Issue with LM27762 (wrong input and outpu voltage)

Hi Alassane,

I will check on your request and I will come back to you on Monday.

Regards,

Davor

Hi Alassane,

I have a few more questions about your design:

1. What is your load current Iout?

2. Are you using low ESR ceramic capacitors as suggested by the datasheet?

3. Please check the voltage on the FB pins and make sure it reads 1.2V. If not I would consider using lower value resistors for the feedback divider, in fact I would recommend trying R2 = R4 = 50k, of course using a lower value for R1 and R3 to keep the ratio the same. 

Regards,

Davor

Hi,

1/ On both of my designs, i currently don't have any load. I didn't mount the IC because i don't have the right output voltage.

2/ I'm only using X7R and X5R capacitor which are supposed to be low ESR

3/ About the FB pins: I'll check, but the resistor value were given to me using TI WEBENCH...

I see that there are many issues with this LM27762. Should i consider using another chip?

On one PCB i have : 

Vin = 5V

Vout+ = 5V (i'm expecting 3.3v)

Vout- = -3.36 V

FB+ = 1.16V

FB- = -0.94V

On the other PCB i have :

The voltage at C1 (Vin) doesn't go up 0.5V...

Hi Alassane,

In order to check if there is a issue with the 2nd PCB or perhaps the LM27762 was damaged, please remove the LM27762 and measure the voltage on Cin with out it. If it reads still lower than 5V then there is a issue with the 2nd PCB somewhere else.

For the 1st one, did you perform your previous measurements for the same resistances as in your schematic? If yes, please try to change the resistances for lower values, such as R2 = R4 = 50k and suitable R1 and R3. I read that WEBENCH has given you these resistances in the design, but sometimes having large resistances helps noise to couple to the FB pins and this causes poor regulation.

Also please send the Vin voltage waveforms for both PCBs, so that we may debug further.

Regards,

Davor

Hi,

I'll try removing the chip later on the second PCB.

When you say R2=R4, you talk about R2 and R4 from my schematic or R2 and R4 from the datasheet?

Hi Alassane,

I am referring to R1 through R4 as shown in the datasheet.

Regards,

Davor

I replaced the resistors : 

R1 = 91.5 kOhm
R3 = 89.2 kOhm
R2 = R4 = 52.3 kOhm

But i still have the same result, Vout- = -3.3V and Vout+ = 5.25V...

I then replaced the chip and both output are equal to 0V.

Hi Alassane,

Can you please send an oscilloscope plot of the voltage on the Cin capacitor? The issue I am seeing is that you have a VIN = 5V however your VOUT+ is 5.25V which seems quite strange, since the connection between Vin and Vout is made through an LDO so the output voltage has to be smaller.

Also please provide the Vout+ and Vout- waveforms for more debugging.

Regards,

Davor

I made a mistake, My Vin is equal to 5.25V, so there is no regulation at all.

And as i said, since i changed the component, the outputs are both equal to 0V. I'll try to change the component once again as soon as i can, during the week, to see if i have something on the outputs, but with all the topics around the LM27762, i'm really worrying that it's not robust at all.

Hello,

I did some testing today and this is what i get :

- I replaced the chip on one PCB with Vin = 5.42 V. What i got at Vout- was -3.45V and Vout+ = 5.43V...(should be 3.3V)

- I took a new PCB and a new chip, i get : Vout- = -1.3 V and Vout+ = 3.3V.

I have used too many chips until today. Should i consider redesigning my card with a reliable chip? I see that many people seems to have issues with the LM27762.

I feel like i can't regulate Vout+ and Vout- at the same time.

Hi Alassane,

Your first picture shows the waveform of Vin correct? It does have an average value of 5.44V, but it seems that there is some ripple that goes quite a bit beyond 5.5V which is the maximum recommended voltage at the input – even though the absolute maximum is 5.8V this is a stress rating and not recommended during normal operation. Can you send a clearer waveform of Vin where the y-axis has a higher resolution, so that we can be sure Vin is within recommended conditions.

Regards,

Davor

Vin : 

- Vout negative : not regulated

- Vout positive: = regulated

For Vin, i'm using a DC power adapter. I can eventually use a lab power supply next time. For the osillscope screenshot, this is the best thing i can have.

HI Alassane,

Looking at the waveforms with a resolution as low as this will not provide much information. What I mean by higher resolution is having lower time and voltage values for each division, like the waveforms sent in this thread: https://e2e.ti.com/support/power-management/f/196/p/833835/3084725?tisearch=e2e-sitesearch&keymatch=lm27762%252525252520waveform#3084725. Also include waveforms for Cp, C+ and C- and if possible it would be best to have them on the same plot (if you have multiple probes at your disposal).

Regards,

Davor

THis is Vin (ch2) and C- (ch1)

This is 5V in (ch2) and C+

This is 5vIn (ch2) and Cp (ch1).

For the resolution, i can't do better than that.

Hi,

I am on vacation right now.

I will reply once I am back on the 17th of August.

Regards,

Davor

Hi Alassane,

These waveforms are not sufficient to help with the debugging. The time-division is at 20ms whereas the LM27762 operates typically at 2MHz, so with this time scale it is impossible to see what happens during the switching times. 
Since this thread is quite old, I will close it. But we can continue discussing the issues you are facing, by sending me an e-mail with the new waveforms at:
d-pavlovski@ti.com.

Regards,
Davor