LM5164-Q1: WEBENCH® POWER DESIGNER Circuit Different than the application circuit in the datasheet problem

Roberto del Pozo Gallardo

Intellectual 440 points

Part Number: LM5164-Q1

Dear Sir/Madam,

I used WEBENCH® POWER DESIGNER for the design of a buck converter.

The parameters I filled in are:

Vin min: 15V
Vin max: 100V
Vout: 12V
Iout Max: 1A
IC: LM5164-Q1

When I tested the circuit it didn't give the right output.

The output was almost the same as the input and it is supposed to be 12V.

In the pictures are the measurements. (Blue is input voltage)

Output Voltage

Voltage before inductor

Feedback Voltage

I was looking at the datasheet and there was an application circuit with the same circuit only with different values.

In the application circuit was Rr the same as Rfbt so I changed Rr.

After that the circuit was working.

I don't understand why.

Could you explain it to me?

Or is the WEBENCH® POWER DESIGNER wrong and should I use the application circuit?

I would like to thank you in advance and I am looking forward to your reply.

Regards,

Roberto del Pozo Gallardo

over 6 years ago

0 Andy Chen over 6 years ago

TI__Mastermind 34785 points

Hi Roberto,

What do you mean Rr? do you use proper voltage rating on these ceramic capacitors?

0 Roberto del Pozo Gallardo over 6 years ago in reply to Andy Chen

Intellectual 440 points

Hi Andy Chen,

In my schematic is Rr = R36 and Rfbt = R39.

Yes de Capacitors have proper voltage ratings.

So can you explain this?

0 Andy Chen over 6 years ago in reply to Roberto del Pozo Gallardo

TI__Mastermind 34785 points

Hi Roberto,

Do you use 25V or higher voltage rating ceramic capacitor on C36 and C38?

The FB pin waveform looks quite strange as amplitude is much bigger than 1.2V, did you solder correct 332K resistor on R36?

Too smaller R36*C44 will inject higher ripple on FB which will cause higher output DC offset and maybe chip malfunction.

Webench use same equation on data sheet to calculate external device value, you can follow up either by same application conditions.

0 Roberto del Pozo Gallardo over 6 years ago in reply to Andy Chen

Intellectual 440 points

Dear Andy Chen,

I used 50V rated ceramic capacitors.

I know for sure R36 is 332K.

But when I changed R36 to 453K it worked.

So can you explain why that is?

Regards,

Roberto

0 Andy Chen over 6 years ago in reply to Roberto del Pozo Gallardo

TI__Mastermind 34785 points

Hi Roberto

From the scope shot the FB node waveform is irregular. It is overcharged up to 3V and then quickly falls to -0.7V.

I’ve seen FB node overshoot during startup and short recovery, but I’ve never seen fast discharge like that.

The FB node must see a low impedance to GND or SW during the off-time to discharge in 100ns.

Either 332kΩ is actually 332Ω or was somehow shorted by solder to cause a FB waveform like that.

The on-time is correct* but FB getting discharged leads to minimum off time and maximum duty cycle, hence VOUT being close to VIN.

I suggest you double check that the ripple resistor is 332 kilo-Ω and was soldered correctly.

0 Roberto del Pozo Gallardo over 6 years ago in reply to Andy Chen

Intellectual 440 points

Dear Andy Chen,

I measured the 332K resistor and it is actually 330R.

The distributor put a 330R resistor in the 332K resistor bag.

Thank you for your help!

Power management

Power management forum

LM5164-Q1: WEBENCH® POWER DESIGNER Circuit Different than the application circuit in the datasheet problem