LT3086 as Adjustable Voltage Controlled Current Source: Low load behaviour

Hi,

I want to use the LT3086 as a voltage controlled current source. Based on https://www.analog.com/en/design-center/reference-designs/circuit-collections/lt3086-adjustable-voltage-controlled-current-source.html#cc-overview I realized the following circuit:

Imon/Ilim is connected to a 16k resistor (R67, R66 and R68 are not used), thus limiting the current to 50mA. TRACK-Pin is connected to the output of a MCP4725 DAC in order to be able to adjust the current via I2C. An on board INA226 measures the resulting current and voltage. Its output values have been validated via external measurements. The load is a light emitting diode.

To test the behaviour, I slightly sweep the TRACK-Voltage and measure the resulting constant load current. Additionally I simulated the circuit using LTSpice. The measured and the simulated curve read as follows:

As can be clearly seen, there is a slight shift between the simulated and the measured curve. If VTRACK goes below 0.12V, the LED starts flickering and finally just turns off. So, current below 15mA can not be reached, although simulation suggests so. Unfortunatly, the current region between 3mA and 15mA is my region of interest. 

What might cause this shift in constant current and how could current lower than 15mA be realized?

Thanks, Rob.

  • 0
    •  Analog Employees 
    on Aug 22, 2019 8:07 AM over 1 year ago

    Hi,
    You mentioned that in your design Imon/Ilim is connected to a 16kohm resistor but the circuit you posted shows that it is connected to a 1k-10nf RC (since you mentioned that R67, R66 and R68 are not used). What is the correct design you used in actual and in simulation? I'm a little confused.

    If I'll base it on what is stated on the datasheet, when you are trying to achieve an external current limit of less than 1A, you must connect a series 1k-10nF network in parallel with the RMON resistor for stability.


    Please let me know the design you used on your circuit and in simulation, maybe there is a discrepancy on your measured value and in your simulation because of that. 

    Regards,

  • Hi,

    sorry, I was not clear enough, I use a 16k resistor (R67) in both, the actual circuit and the simulation. R66 and R68 are not used, but optional in my circuit. 

    Thanks, Rob.

  • 0
    •  Analog Employees 
    on Aug 23, 2019 8:34 AM over 1 year ago in reply to robbeh

    Hi, 
    For your actual circuit, can you try changing the resistor on Imon/Ilim pin to 1k and connect a 10nf cap in series. 
    As reference, check typical applications "Adjustable Voltage Controlled Current Source" and "Ensuring External Current Limit Stability for Ilim≤1Aon page 27 of the datasheet. Then let me know if you still get the same result.
    Regards,

  • Hi,

    I replaced R67 with several other resistors, including 1k, this of course changes the slope of the response, but not the low current cut-off. I could not achieve a load current below 15mA. Following the datasheet, my circuit has the R-C stabilizing right from the beginning, see R65 and C69.

    Just to check, I also replaced the load LED by a 220 Ohm resistor in order to test with a resistive load, but get the same low current cut-off.

    Thanks

  • 0
    •  Analog Employees 
    on Sep 2, 2019 1:57 AM over 1 year ago in reply to robbeh

    Hi,

    Sorry for the late reply,
    After looking further into the datasheet and your design.
    The only thing I see that may affect your design is the PCB layout.

    Make sure to follow the reference (demo board) for your components value/rating and recommended PCB layout. 
    Layout can possibly affect the performance of your IC that may cause the problem.

    Regards,