Capacitor Selection & “DC Link”: De-Mystified!

[@baberraza4231]

simplest, most useful explanation I've seen in a long time, of capacitor construction, specs, types, and uses.

[@kerokero3970]

just found your channel. really nice. came for some explained calculations for selection, but liked the content anyways.

[@EEforEveryone]

Hi Kero Kero! I'm glad you enjoyed! Yes, choosing a DC link voltage is a very complicated matter. It will depend on many factors... But for a ups, it must likely be greater than the peak output voltage (plus a bit of margin for losses in the inverter). Thanks for watching!

[@muhammedkara9185]

You are making great content which is very rare. But please change your voice level sometimes. It is incredibly hard to focus all along these long videos.

[@EEforEveryone]

Hello Muhammed! I'm glad that you're excited about our content, and I hear you. :) Maybe I'll print off this comment and put it by the camera for extra motivation. As a small thank you for helping me make this community better... If you'd like a EEforEveryone sticker, send an email to the business inquiry email posted on this channel and we'll work out the details. support.google.com/youtube/answer/57955?hl=en Cheers!

[@dennismalandro2110]

0:14 Stabilize a voltage _what_ ?

[@Luca-fq9cd]

Cool video like always

[@EEforEveryone]

Thanks Luca! Thank you for watching!

[@EEforEveryone]

By the way, Luca, I really appreciate all the feedback you've been giving us - It has been very positive and constructive. It pushes me to do more fun and great things! I want to say thank you, so I'll make you the first lucky viewer to win our latest giveaway. It's nothing crazy - but It's a token of my appreciation, and I hope you like it. Unfortunately, I can't find a way to PM you on youtube... so maybe you can help me out there or try reaching out to our business inquiry email? support.google.com/youtube/answer/57955?hl=en

[@Luca-fq9cd]

@@EEforEveryone thank you very much i am very excited but for the moment my business mail is offline so on that way it wont work Furthermore i am vacation. When i am back i will write you. See you, thanks for your great videos.

[@akshaykumarvyas]

i have a few questions: 1. i have seen some welding inverters with full bridge topology which use electrolytic capacitors after mains rectification, but some welding inverters with half bridge topology using film capacitors after mains rectification with very small capacitance around 50~100uf. how does such small capacitance work for such powerful welding inverters which are rated at 7000 watts. 2. i have seen some chinese welding inverters which have only a few capacitors after mains rectification for very high power output upto 4KW. is this ok? what can be the consequences. 3. what will happen if we avoid capacitors after mains rectification in very high power switching power supplies.? thank you

[@EEforEveryone]

Hello Akshay, Thanks for watching, and thanks for the great comment! A few answers: 1.) You make a good point that I didn't mention in the video: We can make decisions in a design that change what is required for a design. For example: If I design an inverter to operate at a higher switching frequency, less input capacitance will be required. If less capacitance is required, then perhaps a film capacitor becomes cheap enough, and this could lead to an overall more reliable design. (An inverter that lasts longer) Another approach could be to use electrolytic caps, but WAY more of them than is truly requires so as the capacitors dry out (and the capacitance value drops) the inverter continues to function. It's also possible that the design using electrolytic capacitors won't last as long, and will require a person to either throw it away, or replace the capacitors to restore functionality! It's a very complex (and incomplete) answer... but you asked a very complex question. Without the context for all the other decisions made around that design - all I can say is that it's possible to design an inverter that will work, for an amount of time, with any type of capacitor. Some designs may just last longer than others. 2.) If the design seems too good to be true... it probably is. It's possible they're using aftermarket components, re-using stolen designs, or something along those lines. Duplicating a circuit without understanding can lead to premature failures. That said, it may also work fine! If you're willing to take the risk of buying some E-waste(because you REALLY need that welding Inverter)... I'd recommend opening it up first to make sure everything is fused / isolated properly before plugging it in. 3.) It's possible! For example, current fed inverters (instead of voltage fed inverters) require very little input capacitance. Of course, the voltage will fall to 0 on the input (when the mains voltage crosses 0), so I'd expect to see a bit more voltage ripple on the output. Again, see answer 1 - depending on the requirements, it's possible to design circuits in many, many ways! It has less to do with power level, but for a high power designs, the high-frequency energy needs to come from somewhere.... and without capacitance it's coming across a cable, which could lead to electromagnetic radiation problems down the road. This aren't complete or universally applicable answers, but I hope you find it useful or interesting! Cheers!

[@akshaykumarvyas]

@@EEforEveryone hi there, thank you for answering. The welding inverter which i looked into, having a single plastic film capacitor (50uf) running on half bridge topology, was switching at only 30KHz. That's not too High I guess. It ran on 3 phase 415VAC input. Is there any calculation which shows the minimum capacitance required if plastic film capacitor are being used . Also I didn't know that the switching frequency mattered for input capacitance after mains rectification. Is there any calculation for this too. Please let me know

[@EEforEveryone]

Hello! The calculations for how much capacitance is required is the same for every type of capacitor. Consider how much current is being pushed into or out of the capacitor, and for how long. That will tell us how much the voltage will change during one cycle. If the voltage will change too much, there isn't enough capacitance! This is calculated using the fundamental equations for a capacitor. Rectified 3 phase is very different than rectified 1 phase power, due to the 120 degree phase shift. This means the input voltage will never truly hit zero. For something like these capacitors, there is the mains input, which is at a fixed frequency, but the other half of the circuit matters too! The switching frequency of the downstream circuit will put some requirements on the DC link capacitance because it's effectively serving as input capacitance. Again, these calculations are not revolutionary or different from other DC-DC converters, we're just talking about higher voltages. The calculation for quantity of capacitance required is the same for this (Current and Time), but ripple current, ESR, and self-resonant frequency must be considered as well.

[@zarif_23]

Great content. If possible can you help me design a low voltage high current electrolytic capacitor for my project.

[@EEforEveryone]

Hi Zarif! Sounds interesting. I try to help where I can, but can't always help as much as I'd like to. :) Reach out to the business inquiry email (on our about page)!

[@Fridelain]

There is something unnerving about the way you present.