This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM3409HV: Saturation af the LM3409HV driver at low Iadj values 1-3472849393

Part Number: LM3409HV

I have been working with the LM3409HV for some years now. I like it very much.

My parts have identical package code SYHB

Bath3 code: 73EJ

Bath1 and 2 code. 5C2V and 655F

When I test the behavior of my design I have been used to that the old batches (1 and 2) had a linear section (at the high currents values) that DID not pass through 0,0 of my plot, when the current was controlled at pin 2 Iadj.

My new batch (3) has now a linear section (at the high currents values) that DO pass through 0,0 of my plot, when the current is controlled at pin 2 Iadj.

All bathes still saturates at low Iadj values. Bath3 at a level > 0 Amps, Bath1 and 2 at a level = 0 Amps.

Can this be explained in any way

Best Regards

If schematics is required ask for them at fr@chemometec.com

2273.LM3409HV.pdf

  • Hello Frans,

    Has anything else changed for the testing? Or is it the exact same PCB and BOM for each device? Is the LED load the same with the same Vf and dynamic resistance? There are a lot of things that if they are just a bit different can affect the dimming profile and range.

    If they are identical did you try and differentiate by testing a batch 2 device and then replacing it with a batch 3 device on the same PCB and re-testing?

    Let me know and if we need to explore further I will e-mail you directly for a schematic. Thanks.

    Regards,

    Clint

  • Hi Clint and thanks.

    It is two different designs and PCBs.

    I have been working with the LM3409HV for quite a while now, with peak currents in the range from 1A to 9 Amps and a lot of different LED types, in parallel and in series.
    However this was the first time I noticed the saturation current higher that 0A. 

    These two design are also my first designs with a capacitor at the output.

    I have used the webbench tool for finding the components, but the Iadj is not adjustable in the webbench tool.

    Have you got a reference design with LM3409HV that will work in the Tina 11 - Classic simulator?

    Best

    Frans

  •  Hello Frans,

    Ok, that makes perfect sense now. With IADJ at 0V the output current will never actually go fully to zero, EN or UVLO need to be pulled down to achieve zero current.

    But, there will be a big difference between using an output capacitor and not using an output capacitor. Basically with no output capacitor the voltage ripple at the output only depends on the dynamic resistance of the LED, so it tends to be pretty large. As you analog dim you will get to a point where the output voltage falls below the 1.24V COFF threshold voltage. This forces the device into a maximum off time condition and the current will fall very low. But this also causes a bit of a discontinuity so that the current as you dim isn't totally linear at the crossover.

    With an output capacitor the output voltage is held up by the capacitor and the output voltage ripple is much less. This gives a smoother dimming profile but it also keeps the output above the COFF threshold so maximum off time doesn't happen, the off time it is still dictated by the output voltage. So the effective minimum duty cycle is higher and the minimum current is higher. This explains the discrepancy between your two circuits.

    As for TINA, I am not sure about those models. I am not familiar with any of them believe it or not (don't tell anybody). I always just do paper designs or use Mathcad or excel. I hope the above explanation helps you.

    Regards,

    Clint

  • Thanks Clinton.
    Do you know of any one to contact regarding a reference design for Tina Spice?
    Best
    Frans
  • Hello Frans,

    Your best bet is probably to post in the "Development Tools" forum. If that doesn't pan out let me know and I'll see if I can track somebody down. But the right people should be monitoring that forum.

    Regards,

    Clint

  • Hi Clint.

    Thanks.

    Best fRans