LM5069: LM5069 issue

Pierre Carrié

Part Number: LM5069

Hi,

I need your help about the LM5069MM-2/NOPB. The goal of the application is the following :

The productX have to supply the productY with 48V and 2 A. If the power is upper than 90 W, the LM5069 will shutdown the output power so the productY will shutdown.

I have downloaded the calculator from your website. You will find it filled with the application.

This is the schematic of productX :

The VIN pin(second image) is connected to C25 capacitor (first image).

Note :

-diode D9 is not soldered so the LM5069 circuit is isolated from the rest of the product. The output voltage is between the Gate and AGND_BNC.

-Following components soldered are : R76 = 24 mOhm, C62(Ctimer) = 47 nF, R75(RPWR) = 267 kOhm

-Rest of components are the same as schematic so C73, R78, R21, R55, R74, R39

If the I connect nothing to productX, the output voltage between the gate and GND is 48 V. If I connect a productY the productX, the output voltage is 0 V.

I tryed to connect a pure resistive load from 15 W to 90W to productX but there was 0V too.

I hope you can help me, thank you for any help.

Regards,

Pierre

over 2 years ago

0 Rakesh over 2 years ago

TI__Guru* 95551 points

Hi Pierre,

Welcome to E2E!

Thanks for sharing your details. As I understand, that you have issue while starting with load. Can you share startup waveform of Vin, Vout, input current, timer voltage using 4-channel oscilloscope.

What are the load resistance value you tried ?

Also, share me the excel calculator.

Best Regards, Rakesh

0 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

Hi,

Thank you for your reply. First there is the calculator file.

I will share you the waveforms as soon as possible.

5001.LM5069 Design Calculator with FET Recommendation _V3.xlsm

Regards,

Pierre

0 Rakesh over 2 years ago in reply to Pierre Carrié

TI__Guru* 95551 points

Thanks Pierre.

With the current design you are very marginal.. the timer will timeout during startup.

While testing increase the timer cap to 100nF and check what amount of load you are able to startup.

Best Regards, Rakesh

0 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

I forgot to tell you that I have tested with C62(Ctimer) = 390 nF and it fails

I'im testing with 100nF and give you a feedback

0 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

Here is a document with the different tests I made. I can't measure input current with osci but it's around 10 mA.

debug_LM5069.pdf

0 Rakesh over 2 years ago in reply to Pierre Carrié

TI__Guru* 95551 points

Hi Pierre.

Measuring current will help to debug whether the device gone into fault mode or not. Please use 4-channel oscilloscope and capture startup waveform of Vin, Vout, input current, timer voltage.

You mentioned that you used pure resistive load from 3 W to 30 W. What is the resistance value ?

Best Regards, Rakesh

0 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

Right now, I can't do measurement of 4 channels but 2 channels at the same time. Moreover, I don't have current probe. I can measure the voltage difference at Rsense to see if the condition of 55mV is OK and calculate the input current.

The resistance is 39 Ohm.

Best regards,

Pierre

0 Rakesh over 2 years ago in reply to Pierre Carrié

TI__Guru* 95551 points

Thanks Pierre.

I can understand your constraint. With those design parameters, I don't see any reason for startup issue with 39 Ohm.

Best Regards, Rakesh

0 Rakesh over 2 years ago in reply to Rakesh

TI__Guru* 95551 points

May be you can ship us the board for debug as I see that as the only option.

Regards, Rakesh

0 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

Yesterday, I was working on the issue and some things come to mind :

-With load for 42V, if you look at Vin_VCtimer waveforms the LM5069 seems to power up in restart sequence. The waveforms is similar as the restart sequence waveform given in the datasheet. The voltage at pin Timer goes over 4 V.

-At power up with productY for Vin(productX) = 42 V, the waveforms of the timer (blue) and the gate (yellow) are below :

Power up sequence waveform in datasheet :

We can see three things:

-the gate pin voltage is rising up at the same time with the timer pin voltage

-the timer voltage pin goes over 4V and reaches 11.5 V

-the load capacitor influences the timer pin voltage

I will discuss it with my boss about the shipping of the product.

Best regards,

Pierre

0 Rakesh over 2 years ago in reply to Pierre Carrié

TI__Guru* 95551 points

Hi Pierre,

That is the expected timer behavior under fault mode of operation.

Best Regards, Rakesh

+1 Pierre Carrié over 2 years ago in reply to Rakesh

Prodigy 40 points

Hi Rakesh,

I would like to thank you for your help. I fixed the issue 10 minutes ago by finding a layout issue. The top first of ctimer is connected to Ctimer pin but the second one was floating. I pulled it to the ground.

Have a good day,

Pierre

Power management

Power management forum

LM5069: LM5069 issue