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.

TPS61500: TPS61500 Inductor selection

Part Number: TPS61500

Hi, 

I'm looking at using the TPS61500 for a 10V, 0.3A TFT backlight design and I had some questions regarding the selection of the inductor. 

1) There's no guideline for frequency selection vs. the inductance value. I understand it's a trade off between efficiency and component size but how do we calculate the correct value to use in the 4.7uH to 22uH suggested range for a specific frequency?

2) I assume that frequency selection is a continuous range and that values between those listed in table 2 of the datasheet can be used? 

3) It's been recommended previously for buck regulator design, that ideally the inductor saturation current should be larger than the worst case current limit of the internal switching FET. Is this still the case for the boost topology used in the TS61500 and if not, what's a sufficient de-rating for the inductor saturation current? Is it possible to have large current transients or does the feedback resistor ensure that the current is kept within the designed limit? 

4) There's no information in the datasheet for ensuring control loop stability - does this mean that operating in the recommended conditions with the recommended values is guaranteed to stable?

5) The datasheet recommends the compensation value being set to 470nF but the EVM appears to use a different value. Which is correct and is there a particular reason this is different?

Thanks

  • Hi Pjs,

        Here is an application note that gives you some guidelines for choosing the inductor for the backlight driver:

        

        A quick method is to test the loop stability, since larger cap will lead to small bandwidth, using the 470nF as the start point and test with your target application.

    Regards

    Sean

  • Hi Sean, 

    Many thanks for the SNVA763 app. note - this was very helpful. I had a few more follow-up questions that I was hoping you might be able to help with:

    1) Is there any guidance on selecting the inductance value vs. switching frequency? The TPS61500 datasheet suggests that increasing the switching frequency will reduce the value of the external capacitors and inductors as you would expect but provides no formula for this calculation.

    2) With regards to the saturation current, during normal PWM operation does the control loop respond sufficiently quickly to prevent current transients that far exceed the nominal current limit set by the FB pin? I.e. when the output is turned on am I likely to temporarily see an output current spike similar to the internal switch current limit or does the control loop respond quickly enough to ensure the output never exceeds the nominal current limit set by the resistor on the FB pin?

    3) Could you elaborate on selecting the DIMC capacitor value used for analogue dimming? It sounds like the value is dependent on the PWM speed but it's not clear how to calculate a suitable value. 

    4) Finally, the datasheet suggests, that for analogue dimming, the PWM duty cycle should ideally remain above 1%. Can you confirm why this is and what's likely to happen if a finer duty cycle is used - say 0.1% for example. 

    Many thanks

  • Hi Sir,

         1. For the inductor, you can select between 4.7uH to 47uH with the different switching frequency;

         2. When it triggered the current limit and it will stop switching until it decrease to the normal value;

         3. For DIMC pin cap, The pulsed reference voltage is then filtered by a low pass filter that is composed of an internal 25-kΩ resistor and the external capacitor C5. The output of the filter is connected to the error amplifier as the reference voltage for the FB pin. Therefore, although a PWM signal is used for brightness dimming, only the LED DC current is modulated. This eliminates the audible noise which often occurs when the LED current is pulsed during PWM

    dimming. Unlike other methods for filtering the PWM signal, the device analog dimming method is independent of the PWM logic voltage level which often has large variations. For optimum performance, TI recommends that the value of C5 be as large as possible to provide adequate filtering for the PWM frequency. For example, when the PWM frequency is 5-kHz, C5 equal to 1 μF is sufficient.

        4. The recommended minimum PWM on time at start-up is 200 μs. you can use this to calculate the minimum duty cycle.

  • Hi Sean, 

    Thanks for the quick reply. 

    1) Just to confirm I understand, for the inductor are you saying any value between 4.7uH to 47uH can be used with any switching frequency on the TPS61500 for a stable output? 

    2) How quickly will the current limit trigger? Am I likely to see transients that exceed the current limit set by the current sense? 

    3) OK so I need to ensure that the RC network filters the PWM waveform sufficiently so I get a DC average value that relates to the square wave duty cycle?

    4) Thanks but can you confirm what's likely to happen if a finer duty cycle is used - say 0.1% for example. I'm interested in understanding why 1% is the limit and what the effects would be of exceeding this.

    Many thanks

  • Hi Sir,

        1. High inductor for low switching frequency, low inductor for high switching frequency.

        2. Please test by yourself.

        3.yeah.

        4. This is for internal start detection and 200us is the requirment.