• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 615 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

TL497 vs. TL497A

TL497A: TL497A Sensitivity to Feedback Pin Noise, & Was The TL497 More Sensitive to Feedback Pin Noise Than TL497A?

Andrew Gibson
Intellectual 640 points
Part Number: TL497A

Good Day.  Apologies in advance for asking about such a venerable part.  It is still in current production in lots of packages, though, so there is demand and field use of the part.

Some documentation refers to protection circuits which may be added to the feedback pin (COMP INPUT) to stop noise on that pin from interrupting the inductor charge cycle by driving the pin above the 1.2x V reference level momentarily.  This contradicts the description and normal operating plots in which a full charge cycle is completed even after the feedback pin voltage is higher than the reference in the same documents.  Was this a legacy problem from the TL497 which made its way into the documents for the TL497A? 

More important to current actions, referring to document SLVA143, "Designing Switching Voltage Regulators With TL497A":

Does the TL497A ignore changes in feedback pin status during a charge cycle as shown in Fig. 4?

or

Does a momentary high voltage on the FB pin momentarily interrupt the timing capacitor charge process as shown in Fig. 14(b)?

Thank you in advance.

Andrew

  • 0
    TI__Guru 71675 points

    A momentary high voltage on the FB may interrupt the Ct and hence the on time. Fig 4 is just used to show how a cycle can be skipped and the magnitude was exaggerated (not drawn to scale versus Vprogrammed) for easy illustration. To affect Ct the FB should be much higher than caused by the regular output ripple voltage.

  • 0 in reply to Youhao Xi
    Intellectual 640 points
    Thank you for your answer. It helps a lot, to see how the different sections of the same document still agree. Can you share an idea of how large a bump is needed on FB to cause an effect? I am curious about the general scale of noise needed i.e. 10 mV, 100 mV, 1 V... .
  • 0 in reply to Andrew Gibson
    TI__Guru 71675 points
    We need to find out the info for you. This part was released in 1976 and I started to support it last year. Please allow us some time in searching the info.
  • 0 in reply to Youhao Xi
    TI__Guru 71675 points
    Our design team studied the design and found no hysteresis voltage built in. It seems when FB exceeds the ref voltage the Ct charging current is interrupted, which is why the app note recommended the diode clamp of FB to avoid affecting the Ct charging cycle.
  • 0 in reply to Youhao Xi
    Intellectual 640 points

    Hi Youhao,

    I would like to make sure I understand.  Does this new information mean that any time the FB exceeds the reference voltage the Ct charging is interrupted?  Even during normal feedback operation?  I have seen application circuits without the clamp diode in full production - would those circuits only end an output current pulse and Ct charge cycle when the inductor saturates and the current limiter circuit turns on to charge Ct faster?

    Thank you for your help and thoughts on this.

  • 0 in reply to Andrew Gibson
    TI__Guru 71675 points
    It looks like that. I am going to order an EVM myself to have a better study of this lagacy device.