How do I determine LM5019 frequency tolerance?

LM5019 is called constant on time (COT) converter.  It is really more correctly called "adaptive on time" as the on time is adjusted with VIN to keep the switching frequency more or less constant.  There is no internal oscillator like you would find in traditional PWM control.  Figure 11 in the datasheet will give you some idea of teh nominal switching frequency variation with input voltage.  Perhaps someone from that product line has more details on frequency variation.  If you are looking for predictable, stable switching frequency, then a current or voltage mode converter with an internal oscillator may be a better choice.

Despite the fact that it doesn't have an oscillator, I was hoping for more hard data on the on and off times.  As you say, maybe someone from the product line will see this and have more details.  If I could find a step-down converter which was capable of handling 80+ volts in while being very small and having a more predictable frequency, I'd use it, but so far I have not found it.  Any ideas?

Thanks,
Scott

You did not mention your general requirements.  What are your input and output needs?

Input voltage range 20 to 60 V steady, but must withstand transients up to, say 95 V.  Output is 12 V at up to 75 mA.

Thanks,

Scott

The frequency varies in proportion to the ON time variation at a given condition. You can use the datasheet TON spec parameters in electrical table to get an estimate of the accuracy.

Looking at Figure 11 in the datasheet, the frequency appears to be stable with +/- 10% over a wide Vin range.  When I look at Figure 7, the curves have quite a different shape, with a huge variation in On Time, so much so that the vertical scale is a log scale!  So, it's clear to me that Ton does not by itself determine frequency, but rather Toff must be varying as well in order to have the frequency be much more stable than Ton as Vin varies.  Can I trust Figure 11 in terms of the amount of variation in frequency over Vin for a given Ron value?

Thanks,
Scott

Figure 7 shows Ton i a range I would expect.  Ideally Ton = ( Ron *10^-10) / Vin, so the on time will vary a lot with Vin.  i don't think figure 11 will tell you much about the tolerance.  If you look at the Ton specs on sheet three you can see that depending on Ron and Vin, the tolerance can vary by +/- 40 % or more.  it looks like teh switching frequency variation will greatly depend on Vin and Ron for your application.

But isn't Figure 11 indicative of the variation in frequency of a given unit, at least, even if unit to unit there is a larger variation?

Thanks,

Scott

That is the way i would read it.  I think figure 11 is "typical Fsw vs Vin"  with no indication of what the min and max frequency might be.  I think the electrical characteristics describes the min and max on time for 4 specific Ron and Vin conditions.  You would have to extrapolate the min and max on time for your specific Vin range and chosen Ron.  Then convert that minimum on time variation to frequency variation.