• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 63 subscribers
  • Views 327 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Lm3478 : sub-harmonic issue

Yoshwa
Expert 4050 points

Hi,

Customer developed about 200 power sources of Vin=8V to 18V, nominally 18V, Vout=12V, Iout=1A.

However, he has got a sub-harmonic issue. It occurred in the rate of one for 30.

From his inquire, parameters are changed to L1=68 uH, L2=100 uH and clock=about 400kHz, other parameters are same as the Figure 36 of the data sheet, and RSL is not yet used.

He is considering to add an RSL to overcome the sub-harmonic issue.

He is worrying about inserting RSL effect to the circuit operation.

Are there any effect for operation by adding an RSL, and what kind of problem may be expected if double values of the optimum RSL is used?

  • 0
    TI__Genius 15810 points

    Hello Yoshwa,

    In the figure on the datasheet, there was no mention of the Ccomp value. It usually helps to debug problems when the BOM is known. You mentioned that the customer did not change any other component apart from the ones listed. I noticed that the output voltage set in the datasheet schematic is 5V, but you are attempting to regulate at 12V. This would require the feedback resistors to change as well. The new values could be, Rfbbottom = 10k and Rfbtop = 84.5k.

    The SEPIC design for your requirements can also be realized using a coupled inductor. The part number you could use is MSD1278T-123MLB. Quite often if the BOM is not optimized for the operation, then there will be oscillations in the operation. The RSL resistor adds more slope to the already existing internal slope. Larger value of slope result in two conditions. One, the part starts behaving more like voltage mode than current mode because the comparing ramp signal is now quite large in comparison to the sensed current signal. This makes the design more challenging to compensate and the resulting design would be slow to respond to transients. Two, additional ramp also causes the effective current limit to happen at lower currents. The internal threshold is fixed at about 130mV. With larger ramp values, this threshold will be met at lower values of load currents. I hope this was clear. 

    Now for your design, based on my calculations, your Rsense resistor value could be 25mohms. This is calculated based on the desired load current while operating at minimum input voltage of 8V. Next, if you use the 100uF output cap, your resulting compensation components could be as follows Rcomp = 8.3k(next standard value), Ccomp = 3.3nF. I hope this is helpful.

    Regards,
    Akshay