Introduction

Hello. It's phi.
This is the last of the series "I made a DCDC converter". In the final episode, I would like to introduce a stumbling block during the evaluation.

But before that, let me review the DCDC converter I made this time.
Inputs 5 to 7V and outputs 3.3V.

Evaluation start

I measured the ripple voltage with an oscilloscope. Ripple voltage is minute vibration and noise in a DCDC converter, and the smaller this value, the better the performance of the DCDC converter.

Here are the actual measurement results (Table 1).

I also tried a simulation with the circuit in Figure 2.

Here are the results (Table 2).

I just noticed it here.
"Huh...? Is the value completely different between the actual machine and the simulation?" (Table 3)

Why is there a difference between the simulation and the real machine?

Looking at Table 3, the measurement results of the actual machine are larger than the simulation results.
Why…. I was very worried.
If it is different from the simulation, then the simulation has no meaning. . .

After examining the difference between the real machine and the simulation, it was found that there is a tolerance for all coefficients.

There was a 5% tolerance on the resistor and a whopping 20% tolerance on the inductance value of the inductor which affects the ripple voltage.
Therefore, when we calculated the actual inductance value using Equation 1, we found a value of 3.8uH.
Applying this value to Equation 2 and recalculating the ripple voltage yields the results shown in Table 4.

From this result, it is thought that the reason why the actual device was larger than the simulation was largely due to the inductance value.
(There was also an error in Vd, but it had little effect on the value of L.)

Another reason for the large error is that the capacitance of the capacitor is small due to the DC bias characteristics of the capacitor.
If Co decreases, the ripple voltage will inevitably increase, so the actual device will be larger than the hand calculation.

I thought that if I had a device to check resistance and inductance values, I would be able to obtain more accurate simulation results by simulating using those values.

what i learned

There may be errors between the simulation and the actual machine.
I felt that it was important to know the characteristics of each part in order to understand the cause of the error.
(DC bias characteristics of capacitors, temperature characteristics of inductors, etc.)

We also learned that we can grasp the range of error in the actual machine by performing a simulation that takes tolerance into consideration. Based on this, we would like to use the simulation more effectively in the future.

I felt that it was difficult because the designer had to consider tolerances and other factors while designing the device so that it would operate without problems even under the worst conditions.

last greeting

Thank you for reading to the end.

I have been blogging for about a year.
By putting the events of the production training into words, I was keenly aware of my lack of understanding of my technical knowledge.

I realized that there is one wall in the stage of conveying from the state of knowing.
I would like to continue learning while being conscious of the understanding of knowledge.

We would like to thank you again for your cooperation over the past year.