LM3409HV: No output signal when changing output coil part (with same inductance value)

Hello Ramiro,

Replacing L12 should not be difficult.  Do you have the original L12 datasheet as well as the new parts you are trying?

What switching frequency is this running at?  Roff is very small for 8V.

Best Regards,

Hi Irwin

Thanks for your reply.

Former L12 is ELC18B470L, which is an obsolet part (http://industrial.panasonic.com/cdbs/www-data/pdf/AGB0000/AGB0000C1.pdf ) . Is 47uH with 10% tolerance, Idc 4.3A, Rdc=37mohm.

I have change to murata 19R473C, that has different pin distance, but has basically the same specs.( https://docs.rs-online.com/038b/0900766b80ebc5af.pdf). Once I have change the coil, then the problem appears. 

I have even change to other inductance, slighty different but at closer values, ( included in the 10% of toerance of the original coil)

The frecuency is theoretcally 525 k I think

regards

Hello Ramiro,

Seems that part should work fine in your application.  These are non-shielded inductors so layout and orientation of the inductor can affect the operation.  What this means is all the field going through the core has to go outside of the inductor as a return path.

Best Regards,

Not sure if I understand what you mean. What should I check that makes a difference?

Hello Ramiro,

If it is not working correct you have to look at the signals to see what is happening.  Is it true that the original inductor works on a board but when switched to another inductor is does not?  Everything else is the same besides changing the inductor?

Best Regards,

Correct. Everything is exactly the same. The signals at the inductor are as I attached in my first post. But not sure what other signal could I check

Hello Ramiro,

Your Roff is much too low at 1 Kohm, at 12V input and 8V output.  Your on-time is approximately 160 ns and off time is approximately 80 ns.  This is over 4 MHz switching.  Try changing Roff2 to 5 Kohm, why is it 1 Kohm?

What does the output voltage look like when this is happening?  It looks like the LM3409 is hitting maximum off-time.  The schematic doesn't show it but is there any capacitance across the LEDs?

Have you tried other inductors?  Perhaps try a shielded inductor.  Another thing to try is to move the inductor away from the circuitry by means of short wires.  Unless this is changing leading edge blanking.

Best Regards,

Hello Irwin:

There is no Capacitance across the leds 

You are right, I do not remember why is 1 k in Roff. It shouldnt be.  I will check with 5 k or higher. Anyway, I have check next thing with Roff=1k

1st picture: circuit with former (and obsolete) coil. Yellow line ==>output, blue Line V(Coff). As you can see, Voff goes over the blue cursor in 1.24, triggering the Threshold 

2nd picture: circuit with new coil coil. Yellow line ==>output, blue Line V(Coff). As you can see, V(Coff) never goes over the threshold.... Why? 

In addition to the previous message, which is not clear for me, and I guess have a big clue to understand the issue, I have followed your sugestion changing Roff. as I havemt any 5 k R, I have change to a 6k98. And I have found similar behaviour, but now in both circuit (never mind f it has the obsolete coil or the new one)

Former Coil:

New coil

Regards

Hello Ramiro,

There are a few things I don't understand.  I will calculate this design and ask more questions.

Best Regards,

Hello Ramiro,

Check to see that your Coff is 470 pF and not larger.

This should be switching at a very high switching frequency.

Have you looked at the output current?  Is the LM3409 reaching peak current threshold of 2.4A?

Can you look at the switch node when it is not operating correct?

According to the scope picture this is switching around 10 KHz.

Best Regards,

Hi Irwin

You are right. Doing a quick calculation, the Coff seems about 0.1 uF

I am going to proceed to change Coff and see what happens. Give me 1-2 days and I will get back to you

Regards