TPS54340 variable output resistor

It is pretty common to use a variable resistor, switched resistors or PWM signal to adjust the output of a dc/dc converter over a small range  such as for "margining".  Adjusting over a large range may be problematic as many of the external component values are calculated with Vout as a variable.  It is usually possible to find a compromise design that will work over a fairly wide range but not be optimum for any one particular output.  CAn you tell us about your specific application?

Thank you John for a quick reply.

Yes that was my consideration in part. I`m looking to select parts for let`s say a output range of about 26-8V.

Now this will have some effect on efficiency, heat production, etc.

I`m not sure if this type of range is acceptable and what it will do to performance.

It`s for a power supply with multiple "end" applications.

Will try setting different variables to see what works (or not).

Do you have any recommendations on the type of variable resistor to use?

I've used these types before:

It all depends on your actual application and how you want to physically interact with the circuit.

Thanks again.

Do you have any experience using digital pots for this type of application?

Like this one: AD5274BRMZ-100 (256 steps).

Or the more precise variant (1024 steps).

Thanks.

Can I possibly power such a device from the EN pin?

As the current requirements are low and this pin has an internal pullup function.

Any other suggestions on powering this device directly from the TPS54340?

Powering it from another source seems less attractive.

You do think this type of resistor can be used for this TI product?

Thanks.

Yes however that I2C master would probably be temporary.

It`s possible to use this kind of product?

Any reasons you can see why it will not work?

Thanks.

Thanks John you answered my questions. 

Are all these AD digital pots useable? Both volatile and non volatile?

I have never personally used one.  Let me ask around and see if any of my colleagues have any recommendation.

Another question. I can`t find it in the datatsheet directly.

What amount of ripple can this device handle on Vin?

At maximum loads there`s a bit of a ripple that I possibly need to reduce.

One thing to watch is the current going through the resistor A, B and wiper, selecting the 10kohm version could be a problem.

Regarding the ripple current. I calculated the input ceramic capacitors using your web software and manually. Considering the switching frequency.

Using the SLTA055–FEBRUARY 2006 document.

(Current)ly I have a X amount of ripple at absolute maximum loads coming from bulk capacitance.

What amount of ripple is acceptable to put into the ceramics? 100mV?

100mV would be great however that would require a (very) large amount of capacitance.

I`m looking to find a workable input ripple for this device at 50Hz.

Do you have some reference material on this variable?

Let`s say I switch at 500kHz... It possbily can handle a bit more than 100mV?

Yes that`s correct.

I`m working on the filter. I`m trying to find acceptable ripple levels on the rectified line voltage.

I calculated the ceramic input capacitor requirements based on current and switching speed.

By the way: your WEBENCH application keeps changing components without request or approval.

It can be frustrating when trying to find optimal values.

When setting COMP values for instance the application asks: are you ok to set these values -> and changes other values in addition like Vout resistors etc.

And it  just changes values after selecting a new component. (automatically and randomly -> I don`t understand that combination)

I can change the value 10x times, the 11 time for instance it mixes up the whole setup.

And finally: I can`t find reference on the values of Ro and Co.

I used the re-comp section however if I ask the program not to auto re-comp (option 3)... it changes the design by itself.

Automatic functions can help however changing randomly (or without notice) can be a bit of problem.

Now with some work I got a good looking design however I need to check the current going through Rlow (R3) and the thermal values.

If I want to manually calculate frequency compensation values I would need Ro and Co?

Currently I`m not able to retrieve the previous settings. Could you provide a (rough and or maximum) number on the Rlow (Rfbb) current?

In relation to this current: what would be a minimum recommend value for Rlow? (considering Rhigh, Rfbt is >100k)

There`s no current outflow from FB? 

I would like to get your view on the 50-60Hz input ripple for this type of regulator.

If I have a 1V ripple at 50Hz that`s a small relative input ripple for 500kHz.

Would you have any reference voltage when working at full load?

"If that ripple is 50 - 60 Hz, I suppose it is due to less than ideal filtering of rectified line voltage?"

So just to clarify, your input voltage source is derived from 50 Hz ac that is rectified and filtered, correct?  Typically the input voltage is ripple is generated by the switching action of the converter.

Yes correct.

If I have a 10V ripple at 50Hz the device will see 1mV per switching operation @ 500kHz from the filtered input.

The more ripple the more difficult it becomes for the device to reference the voltage correctly?

Any recommendations?