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.

TPS62560: Feedback accuracy

Part Number: TPS62560
Other Parts Discussed in Thread: TPS62821, TLV62568

Hi, 

I'm looking at the TPS62560 for a new design and am trying to establish the voltage output accuracy. 

I noted that the Vfb is specified for PWM mode as +/- 2.5% over 2.5V to 5.5V and full current + temperature range, however, there is a footnote that suggests that this is "For Vin = Vout + 0.6V". The switcher claims to be capable of 100% duty cycle so I was planning on using this device to switch in some situations as low as Vin=3.6V to Vout=3.3V. Will the Vfb accuracy be affected if I operate with VIN less than "Vout + 0.6V"? 

Thanks

Phil

  • Hi Phil,

    That note just keeps the IC out of 100% mode, for purposes of the accuracy specification. In 100% mode, Vout just follows Vin, since Vin is too low. Thus, the output voltage will drop.

    But this is an older IC. Why not use the newer TLV62568, smaller with a PG pin, or TPS62821, better accuracy?
  • Hi Chris,

    Many thanks for this - that makes sense.

    The TPS62821 is twice the price and this is for a very cost constrained design.

    The TLV62568 looked like a great candidate but the feedback voltage is only specified at Tj=25 and not specified over line and load. In fact Figure 10 in the datasheet suggests that in PSM mode, the load regulation accounts for at least 1% and Figure 11 for 0.25% line regulation for Vin=5, Vout=3.3 which our application would also need to support. It's unclear what the total error would equate to - webench suggested it was 5.04% which is out of specification for our design. Do you know what the actual error is for this part as I'd like to use it?

    Are there any longevity issues with using the TPS62560 given it's an old device?

    Thanks
    Phil
  • Hi Phil,

    Yes, TPS62821 has higher efficiency and tighter accuracy and is more expensive.

    Figure 3 shows the typical Vout variation over temperature. We only spec 2% at room temp to save cost. TLV62568 should be cheaper than TPS62560.

    If you aren't concerned with efficiency at light loads, the forced PWM feature of TPS62560 would give higher accuracy. In PFM, the TPS62560 has dynamic voltage positioning which sets Vout 1% higher already.

    There are no issues with using the TPS62560, if you want.
  • Hi Chris,

    Thanks - the explanation is greatly appreciated. This design does not have low power requirements so I think the MODE pin on the TPS62560 is useful to increase accuracy.

    One other thing to check is with the TPS62560, to find the output voltage accuracy, do I need to also consider the PWM load regulation (listed as 1%) on top of the +/- 2.5%? It looks like the spec. for the Vref is specified over the full load range 0 to 600mA so wasn't sure if this 1% was already accounted for.

    Will the output voltage accuracy therefore be +/-3.5% or +/-2.5% over load, line and temperature? I'm aware I also need to account for accuracy in my resistor feedback network.

    Thanks
    Phil
  • Hi Phil,

    The load regulation is in addition to the 2.5%. And it's 1%/A, which is less than 1% for this 600 mA device.

    You could also check the TPS62237, which has internal FB resistors, a MODE pin, and tighter accuracy.