LM3450AEV120V30: Output voltage oscillation issue

Hello Ali,

Is it regulating and putting the the correct output power?  Since this is a current regulator it won't regulate to 36V it will provide current and the load will determine what the voltage will be.  Have you run your load on a lab supply to see what the voltage of the LED stack is at the current you want to operate at?

The bias voltages will be set according to the output voltage times the turns ratio between the bias winding and output winding.

Make sure the 'dots' of the windings are in the correct place as well.

Best Regards,

Hello Irwin,

That is true and this is a two stage (flyback+buck) current regulator. By saying VOUT, I mean the output of the first (flyback) stage which provides a constant voltage to the second stage which is the current regulator you mentioned.

You know what, I think I just got what the problem might be. :)

Using the original xformer in the LM3450AEV120V30 User's Guide, the output voltage of the flyback stage is VOUT=50V. This voltage is achievable by putting both secondary winding in series. Looking at the original xformer datasheet, the voltage across both secondary winding in series is 12.5VDC+50VDC=62.5VDC. Using this voltage, the VOUT=50V is achievable.

Now, by looking at the datasheet of the xformer I used (760871333), there are three 12V secondary winding and by putting them in series I get 36V. Using this 36V I don't think I would be able to get 36VDC regulated voltage on the output of the flyback stage.

Solution or next step: I will calculate what the maximum VOUT could be using these three winding (36V) and will set the VOUT on that by adjusting R81.

Thank you,

Ali

Hello Ali,

The schematic shows turns ratio on the transformer:  80:40 for primary to secondary, bias windings 10 each.  Secondary shows 30+10 to get 40 turns.  At 50V the two bias windings should be around 12.5V.  I cannot find the original transformer drawing.  So it isn't 50V + 12.5V, it is 37.5V+12.5V.  The design is set up to be Class 2 so it has to be under 60V.  The output regulates at 50V with both windings in series.  Another reason to do this is to avoid voltage peaking on the bias winding.  Leakage current can cause the bias voltages to rise if they aren't loaded much.

If you set the divider up as you have it should try and regulate at the voltage you set it to.  It is a flyback it will generate the voltage until something clamps it.

I just checked again, the primary to secondary ratio is similar to the design calculations.  In reality you should go through all the design equations to make sure it is operating how you want it to.

Best Regards,

Hello Irwin,

I have to correct something here. I should have say VDC as the output of the flyback converter. VOUT is the output of the buck converter or the LED voltage which I measured as 18V for my LED. I did not measure the LED current or power but I expect it to change from 0 to 700 mA by the triac dimmer.

I attached the original transformer drawing here for your reference. I think you are right. The secondary bias voltage is 12.5VDC and the two secondary winding in series will provide the 50 VDC.

I will go through the design equations to see what other parameters need to change.

Thank you,

Ali

750813651.pdf

Hello Ali,

For Wurth 750813651:

2:1 for secondary with the two secondary windings in series

7.43:1 for both bias windings

The bias winding voltage would be 13.4V with a 50V output

The datasheet calls out Wurth 750813550, this part you can find online and it shows the two secondary windings tied in series also.  It gives 13.4V to the bias windings.  It is for 15W output.

Best Regards,

Hello Irwin,

For 750813550, I am not able to find it online. I checked Digikey and Mouser Canada and they don't have it in stock. Can you please provide the link to the online store?

Thank you,

Ali

Hello Ali,

Sorry, I meant you can find information on line about it.  It has a lead time on the website I looked at (digikey).

Best Regards,