LM5642 buck converter not work

[SILVESTROS9]

Hello to all !
getting some info from AN1292 (Texas application for LM5642 ),I made dc-dc converter to use at a solar controller. At test with no load the converter not work . no output, no pulses 200khz to drive mosfets. it seems to be disable ( on/ss pins below 2volts ). I was checked hardware for errors many times , but I don't find something. The measurements shows that ic is ok , but something disable output. Any idea for possible reasons ?View attachment AN1292-LM5642 EVAL. BOARD.pdfView attachment LM5642.pdf

 

[KlausST]

Hi,

Dataheet says:

PCB Layout Considerations
To produce an optimal power solution with the LM5642 series, good layout and design of the PCB are as
important as the component selection. The following are several guidelines to aid in creating a good layout.

...and a lot more on page #28.

Thus, please show us your PCB layout and give the exact partnames (type, vendor) of the critical parts.
Additional photos could be helpful, too.

Klaus

 

[BradtheRad]

At test with no load the converter not work.

Sometimes a load is needed in order to get any sort of normal operation. Try a resistive load whose value draws 5 or 10 percent of your intended current.

 

[SILVESTROS9]

no output with a light load. pcb was replicated from AN1292 Texas application.
below are some components that I use in demo board.

LM5642 , TEXAS INSTR. vendor FARNELL code : 8181772
PSMN4R5-40BS , NEXPERIA vendor FARNELL, code : 2114728
RLF12560T-4R2N100 , TDK vendor FARNELL , code : 1669981

 

[KlausST]

Hi,

so you used the eval-board?

and you modified it?

Why not step by step modify it back until it works again?

Klaus

 

[FvM]

Connection of KSx and Ilimx is wrong in your schematic. Probably only a drawing error.

As your measurement shows, the chip is hold in shutdown. May be due to previous OVP or UVP fault.

 

[SILVESTROS9]

its a drawing error. my doubt remains ..according to block diagram there is a delay to OVP, UVP ..a possible reason is mosfets bad drive . I check mosfets for damage but nothing found. I replace them , but chip remains in shutdown. maybe inductance of power inductors are wrong. unfortunately Texas Webench Power Designer not support LM5642 for simulation. I think that the only way to find something is the simulation...is there some simulator that can do that ?

 

[BradtheRad]

Your schematic appears to be twin interleaved, buck type, synchronous operated.

Try starting out with a single basic buck converter. Use just one mosfet and one diode (asynchronous operation). See if your control IC operates properly. If it does, then add another section of your design.

At 200 kHz your smoothing capacitor can be 1 uF. (It reaches output voltage fast as compared to your values totaling 2,000 uF).

If you get no output, then try driving the mosfet manually from a pulse generator. Gate voltage needs to reach more than 12V, so that it can turn on an N-device at the high side (N-device placed close to positive rail).