Author Topic: Boost feedback resistor  (Read 925 times)

0 Members and 1 Guest are viewing this topic.

Offline LaserTazerPhaserTopic starter

  • Regular Contributor
  • *
  • Posts: 203
  • Country: us
Boost feedback resistor
« on: May 21, 2019, 07:36:53 am »
How much current do the feedback voltage divider resistors need to conduct to allow this boost IC to work properly?
Seems only ~50nA max is required.


https://datasheet.lcsc.com/szlcsc/XI-AN-Aerosemi-Tech-MT3608_C84817.pdf
 

Offline pwlps

  • Frequent Contributor
  • **
  • Posts: 372
  • Country: fr
Re: Boost feedback resistor
« Reply #1 on: May 21, 2019, 10:34:46 am »
Typically it is recommended to have a divider current at least 100 times larger than the feedback input leakage current.
You might look at  ยง8.2.2.2 in https://www.ti.com/lit/ds/symlink/tps61372.pdf
« Last Edit: May 21, 2019, 10:48:06 am by pwlps »
 

Offline LaserTazerPhaserTopic starter

  • Regular Contributor
  • *
  • Posts: 203
  • Country: us
Re: Boost feedback resistor
« Reply #2 on: May 21, 2019, 12:00:42 pm »
Datasheet only lists SW Leakage nothing about input leakage current.
 

Offline pwlps

  • Frequent Contributor
  • **
  • Posts: 372
  • Country: fr
Re: Boost feedback resistor
« Reply #3 on: May 21, 2019, 12:34:38 pm »
Datasheet only lists SW Leakage nothing about input leakage current.

I wanted to say input bias current which is probably the right terminology here. But in some docs it was called input leakage current. Anyway there is only one "input current" going into the FB pin, whatever they call it.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Boost feedback resistor
« Reply #4 on: May 21, 2019, 09:57:54 pm »
The feedback voltage could be as far off as +/-0.012 volts so the feedback parallel resistance should not be any higher than the worst case voltage error of 0.012 volts divided by the worst case input bias current of 50 nanoamps or 240 kilohms to double the error.

In practice, it will be much lower than this and typically the AC impedance at the feedback pin needs to be low anyway to prevent capacitive coupling into it from preventing proper operation.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf