TPS549D22: why the feedback resistor divider should be less than 100kohm?

Frank Xiao

Genius 12055 points

Part Number: TPS549D22

Hi team,

if the input impedance is larger, why it still needs to be much less than 100kohm?

What's the meaning of much less than 100kohm? 10kohm or 1kohm, or 50kohm which ones are less than 100kohm?

If we use a larger resistor here, for example 400kohm, what's the risks here?

Thanks.

over 5 years ago

0 JohnTucker over 5 years ago

TI__Guru**** 191445 points

Frank,

Values around 10 kohm will work. I would select bottom resistor at 10 k ohm and solve for the top resistor value. If it is above 50 kohm, I would make the top resistor 10 kohm then solve for the bottom resistor value.

0 Frank Xiao over 5 years ago in reply to JohnTucker

TI__Genius 12055 points

Hi John,

what's the risk of higher value resistor, for example 400kohm ? Thanks.

0 JohnTucker over 5 years ago in reply to Frank Xiao

TI__Guru**** 191445 points

Frank,

Higher values are very much susceptible to noise. Further and more important there are internal resistance that will be in parallel with the external FB resistors. With FB resistors approaching 100 kohm, the effective FB resistance path will be much lower than expected (i.e. 100 k || 100k = 50 k while 1k || 100 k + .99 k). So you need to ensure the FB resistors are much less than 100 kohm. I would even consider going lower than my previous post.

0 Frank Xiao over 5 years ago in reply to JohnTucker

TI__Genius 12055 points

Hi John,

understand. Larger resistor will give more noise. Whether it will only influence the voltage voltage accuracy? Thanks.

0 JohnTucker over 5 years ago in reply to Frank Xiao

TI__Guru**** 191445 points

Frank,

Yes I would expect it to affect the accuracy.

0 Frank Xiao over 5 years ago in reply to JohnTucker

TI__Genius 12055 points

Hi John,

Currently my customer uses up resistor 1kohm, low resistor 39kohm, What's the risks here?

Thanks.

0 JohnTucker over 5 years ago in reply to JohnTucker

TI__Guru**** 191445 points

Frank,

Why would you reject this answer?

0 Frank Xiao over 5 years ago in reply to JohnTucker

TI__Genius 12055 points

Hi John,

Need your support to double check my customer's feedback resistor about 1k+39k. Whether it has riks, and what't the risk?

thanks.

0 JohnTucker over 5 years ago in reply to Frank Xiao

TI__Guru**** 191445 points

Frank,

That is a rather unusual combination. What is the desired output voltage? What is the VSEL pin strap selection?

0 Frank Xiao over 5 years ago in reply to JohnTucker

TI__Genius 12055 points

Hi John,

The desired output voltage is 1v, the Vsel pin strap selection is 10k+68.1k, thanks.

0 JohnTucker over 5 years ago in reply to Frank Xiao

TI__Guru**** 191445 points

Frank,

I did some research on this today. I attempted to simulate the design in pspice and discovered that the RSP input current is modeled as a constant 10 nA. Further resarch reveals that the remote sense amplifier is NOT as depicted in the functional block diagram in the datasheet. The datasheet shows a traditional differential op amp with balanced resistor dividers internally. In fact, TPS549D22 uses an extremely high impedance dual transconductance amplifier. The warning verbage about using low value resistance values for the external divider is most likely a carry over copied from a datasheet using the older remote sense amplifier. Using a 1kohm and 39 kohm resistor divider will not pose any issues for you.

Power management

Power management forum

TPS549D22: why the feedback resistor divider should be less than 100kohm?