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.

LM5117: Designing an adjustable constant current (CC mode) supply

Part Number: LM5117

Hello TI experts,

I am trying to design an adjustable constant current supply using the LM5117 chip that can source ~0-2A. Currently, I plan to modify the CC example in the datasheet:

Originally, I tried to replace the voltage divider in the current control box (bottom left) with a potentiometer. The potentiometer seems to have little to no effect, however.

After looking at the datasheet again, however, I realized that because V_CM ~= 2*I_out*R_s*A_s, and the reference voltage for the error amplifier is 0.8V, the small 40milli-ohm current sense resistor is probably too small when trying to regulate the output to 100mA (V_CM = 2*0.1A*40E-3 ohms*10 = 0.08V << 0.8V), for example. My next idea would be to add a variable gain op amp to amplify the output of the CM pin, something along the following lines:

 

Firstly, does this approach make sense, or do I have a conceptual misunderstanding of how the chip works? Secondly, the capacitor connecting the COMP node to the FB node, although it is in the datasheet example, I'm not quite sure how to pick its value, as the datasheet equations consider a filter network with the resistor in series with the cap there.

  • Hi Zachary,

    2A is not much for a controller implementation, so yes a larger shunt resistor value is required. Note that the CM output is only valid in CCM operation. Take a look at app note snva829 for more detail related to CC/CV circuits.

    Regards,

    Tim

  • Does this mean a really accurate digipot on the shunt resistor is the best way to make this thing adjustable?

  • No, the shunt resistor (40mOhm in the original circuit) is a power component. 50mOhm is possible as this increases the signal-to-noise ratio but still keeps it below the internal current limit threshold (you can check with the LM5117 Excel quickstart calculator). Changing the gain from CM to FB, maybe by injecting a  current into FB, seems feasible.

  • Hey Timothy,

    The help has been greatly appreciated, but after frying my brain with simulators all day (I'm very new to designing switching regulators), I'm still a little confused. Firstly, I want to confirm that I follow your last advice. Assuming I use a 50mOhm or greater sense resistor, is something like the following what you imagined in order to adjust current? To recap, I'm hoping to be able to adjust the output from ~100mA to ~2A.

      

    With this idea, a DAC could be used to inject or even remove current from the FB node. Although this intuitively makes sense, my hands on testing and simulations make this seem like it's going to be a difficult task. It seems like everything I have tried so far to be able to modify the CM gain generally doesn't change the output once the circuit reaches CCM. Also, most the supporting documentation for the chip seems to assume a fixed voltage or current, and even knowing the load resistance (which will also be variable). Any help finding the right direction (I have a lot to learn, but I am determined) would be greatly appreciated. 

  • Hi Zachary,

    See Figure 2 in app note SNVA829. The resistor divider tap goes to FB and you can inject current there using the injection resistor as shown in your schematic. See Figure 35 in the LM5117 datasheet as well.

    Regards,

    Tim