0
\$\begingroup\$

I am trying to test an SMPS controller IC (like KA3882) by isolating the controller from the PWM FET circuit and de-coupling the SMPS output (to optocoupler and the feedback pin of the controller). By applying the required feedback voltage to the Vfb pin (2.5v in my case), I expect to get the square wave pulses on the output pin. Is this procedure good enough? I also felt that I-sense pin must be for protection against shorting of SMPS output power supply. Hence , I skip applying a current to this pin, but I am not sure and need advice.

What is the purpose of I-Sense current in a PWM controller IC?

\$\endgroup\$
0

1 Answer 1

2
\$\begingroup\$

The documentation surrounding the KA3882 is pretty poor compared to devices such as the UC3843 made again by Fairchild, ON semi and TI. So make a comparison: -

enter image description here

Make that comparison against the TI UC3843: -

enter image description here

You can also note that the pin numbers are the same and you can pretty much assume that functionally they are the same: -

enter image description here

What is the purpose of I-Sense current in a PWM controller IC?

It's protection against too much current being fed through the inductor and it saturating excessively: -

enter image description here

So, to get more information functionally about the Fairchild KA3882 I'd look into the TI UC3842/3843 range of devices.

I skip applying a current to this pin, but I am not sure and need advice.

Don't apply a current input - it requires a voltage that represents current BUT, you can connect it to 0 volts for just a bench test.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks for the detail answer. But what 0V at ISense pin signifies ? Will I see a pulse train at the ouput? \$\endgroup\$
    – seccpur
    Jul 5, 2018 at 14:06
  • 1
    \$\begingroup\$ Yes, you will still see a pulse train - the Isense input is just to protect the inductor from overload currents and isn't necessary for the circuit to oscillate - Vfb pin will ensure that there will be oscillation close to it being at 2.5 volts - but take care - the open loop gain is high and it may produce 0% to 100% duty cycles on the output over a few millivolts change in Vfb AND it may be at 2.45 volts or as high as 2.55 volts (device dependent). \$\endgroup\$
    – Andy aka
    Jul 5, 2018 at 14:13
  • \$\begingroup\$ Do the totem output need a pull up resistor to see the pulse train 'cause I'm seeing none? \$\endgroup\$
    – seccpur
    Jul 6, 2018 at 1:11
  • \$\begingroup\$ No, it's a push-pull output but, it might help if you posted a schematic of your circuit. I might spot an error. \$\endgroup\$
    – Andy aka
    Jul 6, 2018 at 9:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.