LM5116: Weird HO signal with ringing - without MOSFET connected.

Pawel Staszewski

Prodigy 160 points

Part Number: LM5116

Just checking from where are my FET ringings at turn-off.

And checking HO when transistor is disconnected i found that there is already ringing...

Below osc connected to HO pin.

Attached schematic and PCB layout - probing directly at Q1_1 VIA:

Any ideas from where this ringing can come ?

Checked power supply and there is stable 46.9V DC (without any riple - from bench PWS).

over 4 years ago

0 Pawel Staszewski over 4 years ago

Prodigy 160 points

After connecting FET.

HO signal is mostly just bad ringing...

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

By the way - currently snubber RC is not installed.

0 Andy Chen over 4 years ago in reply to Pawel Staszewski

TI__Mastermind 27115 points

Hi Pawel,

The PCB layout looks not good:

1. C5 was not placed close to Pin16.

2. No ceramic capacitor placed on VIN rail

3. the SW trace not fully displayed, but seems it's too thin and long.

Pls change R8 to 0.015ohm and load current below 2A for circuit debugging, add few 1uF-2.2uF 100V ceramic cap on C15, add 0.1uF 100V on Pin1 and Q1 drain to power GND.

How do you measure HO signal? use differential probe? pls check HO signal again with above changing.

B R

Andy

0 Pawel Staszewski over 4 years ago in reply to Andy Chen

Prodigy 160 points

Thank You for reply.

1. Ok will modify PCB layout (as there is not much space to fit it will be a little hard but will try)

2. Corrected in new pcb.

3. will 1mm width 25mm len. be good ?

Currently im making tests without much load (200ohm resistor as a load).

HO was measured by single-ended probe.

Below attaching a little changed layout for my test PCB . (LM5116 pins are 0.4mm - so the starting point for track can't be more than 0.4mm if any of track already is for example 0.5mm - current clearance is 0.152mm)

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

Hi again.

Waiting for new pcb - but have also other pcb with 5116 - a little different layout (just for testing)

Schematic of this test board:

And HO signal when mos is connected is same as in previous pcb. (below osciloscope signals)

CH1 - HO

CH2 - LO

CH3 - Inductor between fets

Can this be FET problem ?

Used fets:

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

Also with increased values (1/2.2uF) C5/C6 and soldered directly to IC pins - not much change.

Below osc signals:

Currently DC input 15V

DC output set by divider to 12.4V

Attached load - 200Ohm resistor.

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

Also - what can cause no RAMP signal on RAMP cap ?

CH4 of the osc screen above shows vsignal on RAMP CAP.

After adding reistor between VCC and RAMP CAP i can observe ramp signal... but without resistor there is just some close to ground noise.

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

With attached 2ohm resistor as a load + input current limited to 330mA (bench pws)

Below osciloscope signals:

CH1 - HO

CH2 - LO

CH3 - Inductor/SW

CH4 - Ramp CAP

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

And 27 Ohm load without current limit on DC input.

0 Pawel Staszewski over 4 years ago in reply to Pawel Staszewski

Prodigy 160 points

Anyone have any ideas what can be wrong ?

Currently changed caps/fets plus added more ceramic caps at vin but not much change.

0 Andy Chen over 4 years ago in reply to Pawel Staszewski

TI__Mastermind 27115 points

Hi Pawel,

What's your application requirement?

Vin range=?

Vout=12.4V

Load=??

Fsw=250KHz

Can you down load circuit calculator for LM5116 application?

http://www.ti.com/tool/lm5116-5116wg-25116design-calc

The PCB layout you posted on above thread was not good, I strongly suggest you to apply LM5116EVM and modify the circuit to your application, then design your PCB again.

B R

Andy