How to choose the right capacitor type for a circuit?! || Film vs. Ceramic vs. Electrolytic

Recently, I've been playing around with some high power LEDs. To efficiently dim the brightness, I built up this simple test circuit, which features a function generator to create an adjustable PWM signal, and n-channel MOSFET in series to the LED, to actually turn it on and off rapidly, and a TC4420 MOSFET driver IC to charge / discharge the power MOSFETs gates as quickly as possible. Now in the low frequency range This circuits dims the LED perfectly fine by changing the duty cycle of the PWM signal. But while for example using a frequency of 100 kHz and a duty cycle of 1%, The circuit works for a couple of minutes, but then randomly stops working. Because the MOSFET driver IC apparently destroyed itself. After replacing it. This circuit worked fine once again but this time I examined the pins voltages of the IC with my oscilloscope, to determine the culprits. And while probing the supply voltage pin of the IC, I noticed that there occurred 100 kHz oscillations with peak voltages of 28V and 2V Since that is partly beyond the ICs maximum supply voltage. It is no wonder that it self-destructs after a while. To solve this problem, The Würth Elektronik eiSos Group recently sent me three of their capacitor design kits. The General-Purpose DC Film Capacitors Design Kit The Multi-Layer Ceramic Chip Capacitors Design Kit and The Aluminum Electrolytic Capacitors Design Kit. So in this video, let's solve this mysterious IC supply voltage problem, and learn the difference between those three capacitor types, to find out which one you should use for which circuit. LET’S GET STARTED! This video is sponsored by The Würth Elektronik eiSos Group. Let's start off with our MOSFET driver IC problem. The supply voltage breaks down and afterwards an oscillation occurs. This happens with a frequency of 100 kHz. Which not coincidentally is the exact moment the power MOSFET gates get charged up. So, if we break it down the input signal (Vᴅᴅ) gets put high, which ultimately connects the gate of our power MOSFETs to the supply voltage. This action requires current for our IC. In order to power its own components, and ultimately charge up its own MOSFET gates as quickly as possible. While observing this IC current through 1 ohm shunt. I noticed that it reached its first peak value of around 2A within only 15 nanoseconds. The only problem is that my power supply, due to its internal construction, is not the fastest acting energy source. That is why we can model its output impedance as a small resistor in series with an inductor. Now if the IC would require a constant 1A, we would only get a small voltage drop across the resistor, but no other problems! Since an inductor voltage drop only exists with a change in current flow. But since our IC wants to have 2A in a time of only 50ns, Our inductor now features a big voltage drop, which means we got a breakdown in the supply voltage of our IC. Combine that with a breadboard construction which comes with noticeable parasitic capacitances, and we got ourself a small oscillator on the supply voltage pin, that leads to problems. To solve that, we can add a capacitor in parallel to the supply voltage pin. Which is then often referred to as a bypass or decoupling capacitor. It's job is to basically provide the high current search for the IC, which the mains power supply can not offer because it is too slow. And thus it also suppresses noise for other ICs in the circuits. The only question is: “What capacitor type is best suited for this job?” The two main ratings, you usually see on them is their capacitance and their withstand voltage. Now since all of my capacitor voltage ratings are higher than the 12V I'm using, we should go for the highest capacitance rating. Right? I mean, since the capacitance rating is proportional to the stored energy of the capacitor, we should definitely be able to provide enough current with it. So I connected the 15,000μF electrolytic capacitor in parallel to the IC. And asserted that the oscillation peaks decreased to 16V and 8V. Seems decent. Out of curiosity though. I also tried out a small 150µF film capacitor as a decoupling capacitor, which worked even better! By decreasing the peaks to 13V and 10V But, why does such a puny small film capacitor whose capacity is 100,000 times smaller than the beefy electrolytic capacitor works better? Well, the reason is that while all capacitors share the same basic structure which means they got two metal electrodes, which are separated by a non conductive material called the Dielectric, in order to create an electric fields and the store energy when a voltage is applied, their materials all differ. My electrolytic capacitors for example, use aluminum foil in combination with an electrolytes. While my film capacitors use polypropylene and my ceramic capacitors use ... like the name implies Ceramic. This material choice influences electrical properties like the voltage or capacitance. But also other properties like for example, The expected lifetime or whether a capacitor is flammable But there are more hidden properties which we can discover by examining the capacitors with an LCR meter. (L: Inductance C: Capacitance R: Resistance) with an LCR meter. Sadly though the 15,000µF one overloaded the meter. But as a replacement, I used a 10µF one which works similarly as a decoupling capacitor. The first thing we notice is that the capacitor not only features a capacitance but also a resistance and inductance Those are called Equivalent Series Resistance (ESR) Those are called Equivalent Series Resistance and Equivalent Series Inductance. (ESL) and Equivalent Series Inductance. And they do exist in a practical capacitor due to its internal structure. The big problem with that though is that the parasitic resistance creates a power loss. As an example, we can use the 100 Hz measurement of the LCR meter to determine a dissipation factor of 0.097. The dissipation factor describes the relation between the ESR and the capacitive and inductive reactance. But let's neglect the inductive one for now. That means the overall impedance of our capacitor acts around 92% like a capacitor and 8% like a resistor. Which on the other hand means we waste energy that goes in and out of the capacitor as heat If we increase the frequency to one kilohertz We can see how the dissipation factor increases to 0.220 which means the capacitor now features an even bigger resistive components. With rising frequency this DF value increases because the dielectric ohmic value increases while the capacitive reactance decreases with rising frequency. It gets especially interesting when the capacitive reactance = the inductive reactance of the ESL which happens at the self resonant frequency of the capacitor. Above this frequency, the capacitor acts more like an inductor than a capacitor. And thus, It’s not interesting for us when it comes to decoupling. Even the data sheets of the electrolytic capacitor gives us a dissipation factor of 16% at 120 Hz which means such electrolytic capacitors are better suited for ULF applications (ULF: Ultra Low Frequency) are better suited for ULF applications. But if we insert the 150µF film capacitor into the LCR meter. We can see that its dissipation factor is pretty much 0 at 100 Hz and 1 kHz and Only goes up to around 0.001 So 0.1% at 10 kHz The datasheet of the capacitor pretty much confirms those values. By giving a DF of only 0.26% at 100 kHz Meaning such film capacitors have a very low ESL and ESR rating and thus a high self resonant frequency. Which makes them suitable for LF & MF applications (LF: Low Frequency) Which makes them suitable for LF & MF applications (LF: Low Frequency MF: Medium Frequency) Which makes them suitable for LF & MF applications like our decoupling task. But we should not forget about our super tiny ceramic SMD capacitors. For which there apparently exists different classes like NP0 and X7R In a nutshell those two kinds feature a different base material. Which has the effect that class want ceramic capacitors like the NP0 are very stable over a wide temperature range while class two ceramic capacitors like the X7R are not as stable over a wide temperature range but feature way higher voltage dependent capacitances. That makes class 1 ceramic capacitors perfect for something like oscillators while class two ones could be used for decoupling. Right? To find that outs, I grabbed the 10µF one and checked it with my LCR meter. At 1 kHz, We got a dissipation factor of around 3% and at 10 kHz around 15% . So not as low as the film capacitor. But after soldering it to a THT breakout boards and connecting it to my MOSFET driver IC. It reduced the oscillation to better values than what the electrolytic capacitor offered. Now, of course a capacitor datasheet depending on its type can give us even more information like the insulation resistance which basically sits in parallel to the actual capacitance or the leakage current. Whose name pretty much speaks for itself. But you should now understand that while electrolytic capacitors can be used for buffering energy which is why you see them often in power supplies they are generally not well suited for higher frequency filters or decoupling. And if you want more information about other applications of capacitors and the usage of different capacitor types in general, then I highly recommend having a look at the webinar of The Würth Electronik eiSos Group which you can find in the video description. As always, thanks for watching Don't forget to like share and subscribe. STAY CREATIVE AND I’LL SEE YOU NEXT TIME! (As alway, Subtitle by PolaX3)

Channel: GreatScott!

Views: 464,372

Rating: 4.932363 out of 5

Keywords: capacitor, type, compare, comparison, decoupling, decouple, bypass, film, ceramic, electrolytic, tutorial, guide, beginner, beginners, tc4420, mosfet, driver, ic, oscillation, fix, problem, inductor, resistor, reactance, structure, esr, esl, equivalent, series, resistance, inductance, self, resonant, frequency, df, dissipation, factor, lcr, meter, datasheet, buffer, smoth, energy, filter, emi, greatscott, greatscott!, electronics

Id: 2v8zBj7_sxg

Length: 12min 16sec (736 seconds)

Published: Sun Apr 21 2019