#100

This tutorial video shows how to estimate / measure the self-resonant frequency (SRF) of a capacitor using an oscilloscope and a signal generator. It builds upon topics presented in two previous videos:
LC Resonant Circuits:
• #55: Back to Basics: T...
Capacitor and inductor self-resonance:
• #56: Basics of Capacit...
The video also mentions some caveats to be aware of when attempting this. In addition to those mentioned in the video, there are more... The parasitic inductance of the fixture and probe should be substantially lower than the parasitics of the capacitor under test. For small RF capacitors, well under 1nF for example, this is especially important. Special RF fixtures would be needed to accurately measure the SRF.
There are several other factors that can influence this measurement, but the basic concept works well for capacitors used for DC bypassing, broadband signal coupling, etc. For RF applications, it is much more difficult to apply this method and get accurate results.

[@Julius314159]

Why can't these videos go on for hours? So much great content! Much appreciated, sir!

[@txsharpshooter85]

Another excellent video. I went through a 2 year Electronics Technician tech program, and your videos drive home many of the points that were taught in that program. Your teaching style is clear and precise, and I appreciate the effort and time that you put into doing this.

[@CodfishCatfish]

I recently constructed an antenna analyser based on the VK5JST and keeping the cap leads really short improved the stability a great deal. I am glad this video shows this as my friend who is currently constructing one will find this video very helpful indeed. Thanks for the upload.

[@kibi15]

Nice one Alan. Yet another one of those simple experiments that you demonstrate that can provide hours of entertainment. Thank you.

[@markt4834]

Brilliant, practical measurement technique!

[@aduedc]

Thanks for another great clip. What is important in design is the knowledge of the maximum self resonance frequency of the capacitor( i.e. in perfect condition with as small lead as possible) and then you should use the capacitor for designs with maximum frequency 1/10th of self resonant frequency. Similar to rated voltage of capacitor which is used in designs that voltage does not exceed half of rated voltage.

[@JoshHopkinsYT]

TY I feel like I understand this a lot better now. Helps me to understand why its so important to have your capacitor tuned correctly in your primary coil of a Tesla coil.

[@Zeropointmicron]

Very good explanation! The best video I ever saw on youtube on this topic! Thank you very much!

[@alocin110]

Thank you for this upload. Very clearly demonstrated. Keep up the good work. I liked your video.

[@johnchung6900]

Explains inductance really well with the length of the leads.

[@DrHWO]

In the process of building the QCX+ kit and wondered why Hans stressed the importance of keeping the cap legs as short as possible (other than putting it in a box!). Now I know. Thank you.

[@w2aew]

Glad to help! I have a QCX+ that is still sitting in the box, waiting for me to build... Just need to find the time!

[@DrHWO]

@@w2aew My QCX+ may not be the best of ideas for the CW community. My fist has been likened to a spider with castanets !

[@w2aew]

@@DrHWO LOL, mine isn't much better, but I keep plugging along...

[@FRRRRAAANNK]

Thanks for getting back to me so lightning fast. I'll have to run that by a friend of mine as he has the equipment.

[@ronaldlijs]

I loved this video, keep up the great work Alan! Especially the very brief use of the "dip meter", which personally I didn't know about! So I learned something new today, thanks! A dip meter might be a GREAT tool to use for finding coupling issues in audio circuits, will build one asap and try it out! :-)

[@GaRbAllZ]

Great vid Alan. You do a great job explaining this and make it easy to understand!

[@shivaramarun8697]

Hello Sir, I am trying to measure the resonant frequency of a MEMS resonator in which the bottom electrode is fixed and the top electrode is a doubly clamped beam which is free to oscillate up and down (basically like a parallel plate capacitor). How do I measure the resonant frequency of the vibrating beam??

[@loglineproductions5944]

Thank you for a very cool and concise video. Would it be possible to do a video on "Inductor self resonance measured with oscilloscope and signal gen"? Perhaps you could add some detail regarding the difference between resonance, self resonance and spatial self resonance for caps and inductors and when they apply?

[@w2aew]

+Logline Productions An inductor's self-resonant circuit will look like a parallel LC tank circuit, thus will exhibit a high impedance. So, the process is basically the same, except that you look for a peak in the response instead of a dip in the response. Resonance is when the capacitive and inductive reactance are equal in magnitude. Self resonance refers to the resonant point that results from the component itself and it's parasitic characteristics. For example, a capacitor has some parasitic lead inductance - self resonance is when the capacitive reactance equals the parasitic inductive reactance of the leads.

[@Vintage_USA_Tech]

Very nice indeed..... so that make my mind wander to all the cool squiggles and shapes we see in the printed circuit boards of these new high end scopes which i can only guess as to the reason is to manipulate parasitic capacitance and resonance frequency?

[@mxcollin95]

Excellent video as always! 👍

[@researchandbuild1751]

Just for kicks i tested this with a ceramic cap but i went all the way to 50mhz and still didnt hit an srf. So maybe its also depending on the cap obviously too..i tested a film cap and hit srf around 7.2mhz

[@tonybell1597]

Excellent vid, all the theory, than a simple demonstration. Thanks, keep p the good work.

[@wa4aos]

Always enjoy the review Professor!! Care to tackle the infamous RIAA filter used in phono stages? Vinyl media is still out there and many of us prefer it to other types of media. The RIAA filters are tricky due to in and out Z with regard to freq and trying to get the line to match the chart. Vry 73, Glenn

[@FRRRRAAANNK]

Hi there. I love your video. Very educational. I've used different brands of metal film resistors in an audio amplifier. One is a Takman and the other is a PRP. Both are 1W.The Takman sounds bassy and the PRP sounds lacking in bass. Both are super quiet. Is there a test we can do to see what's going on here?

[@drtidrow]

The capacitance in the scope probe may also make a difference, depending on the size of the capacitor. If the cap you're testing is in the tens to low hundreds of picofarads, the probe's capacitance could be as much as 10% or more of the cap being tested. That'll definitely skew the results a bit.

[@christophschuermann6512]

Was this cap a CK05 npo-type cap from AVX ? As always great video! Thanks for helping to understand every detail!

[@chandrashekarg2073]

Hi can you show signal without 10db attenuator

[@JacGoudsmit]

So is the attenuator necessary to make up for the parasitic capacitance and inductance of the wires from the frequency generator and the scope probe? I would expect that with measurements like this one, you can't simply neglect the parasitic capacitance/inductance of those wires but apparently you can, otherwise clipping the leads off that capacitor wouldn't make such a difference. I'm curious as to why.

[@laharl2k]

insnt there any vco circuit that would adjust itself to the self resonance frequency, like plls do with the phase?

[@wa4aos]

Hi Al, I need to fabricate 75 to 50 ohm adapters to go from a distribution panel to various HP devices in my lab that require a 5 or 10 MHz ref. I have this working but as I add more devices the distribution signal is too low. The Z match should help. I know how to make a resistive and inductive pad. Have you done anything with SMD devices similar to filters you made recently. Space is tight on the distrib. panel and trying to get it all to fit. Any suggestions appreciated 73, Glenn WA4AOS

[@Btw_visit_____todacarne-com]

I learned something today. Thanks.

[@laharl2k]

i was thinking about using the frequency as a cheap touch sensor or general sensor, as the frequency would change both by capacitance and inductance, and as R is constant if you put a metal near it or touch it hence changing L and C respectively, then just pass it through an RC filter and feed it to an adc it would be less cpu intensive than meassuring the rise time of an rc constant and would just use a single input, no need of especial ic or anything and you could get the value on demand.

[@Afrotechmods]

Great topic and technique!

[@BrendaEM]

Again, thank you for making these excellent videos. As good as your demonstration is, you have incited me, once again go experiment, firsthand. Changing the lead shape also reminded me of value preserving the return path, by not interrupting the ground plane. I have a Siglent SDG2042X function generator, and a Rigol 1054z, which even though they are entry-level "instruments," they can be hooked together to sweep something like this to make a Bode plot, which might make it easier to find the self-resonance. Ironically, the Siglent function generator's ability to trigger external on a sweep, was a later-added feature. Though, it does work, as tested. Still, a VNA would be nice to have. You have seen the LibreVNA 2nd version, correct?

[@w2aew]

I know the LIbreVNA 2nd version exists, but I have never seen/used one.

[@pe0sat]

Great Video's thanks for sharing all this information.

[@Pwaak]

Incredibly interesting video! Thank You!

[@felixruiztorrez5192]

Awesome videos. Thank you.

[@mjlorton]

Great video, thanks for posting it Alan. Thumbs up and shared.

[@RanHam]

Thanks. Great video as always. I'm very jealous of your shop. Is it just for hobby use?

[@mstjerning8919]

@6:18 The loop area? I would think it is the parasitic capacitance that changes as a function of distance between the leads. The parasitic capacitance is in parallel with the ideal capacitance so this would yield a lower measurement. 1/2*pi*sqrt(L*C)

[@w2aew]

The difference in the parasitic capacitance between the leads is a very tiny fraction of the capacitance of this device - probably just a few pF at most, while the capacitor used here was a 1000pF or 10000pF cap (can't remember which), so the change in parasitic capacitance by dressing the leads wouldn't account for the 10% change in SRF. It was primarily due to the change in parasitic inductance.

[@ovalwingnut]

A2AEW "Making America great again one video at a time". Thank you

[@mob7782]

very interested in purchasing ...

[@alessandrosaporetti9778]

Hi , thanks a lot for all your effort to educate well about EE topics. Have a question for you , can you measure the resonance freq of a loudspeaker with a pulse gen and an O'scope ? thanks a lot Alex

[@absurdengineering]

With a pulse gen it’d be harder, but with an oscillator it’s easy. The mechanical resonance is coupled to the electric domain, and presents as an electrical resonance!

[@tunicana]

very instructive vid.thx

[@uniment]

Many MCU's have a built-in comparator. You can measure RC rise-time by running a counter, and when the comparator trips, run an interrupt that stops the counter and reads the value. Not CPU intensive at all.

[@amikhimji]

Great video thanks

[@TheChipburner]

If I connect the capacitor to inverting amplifier with AGC, will it oscillate as normal series LC circuit?

[@w2aew]

Probably not. In order to oscillate, you need an addition 180 degree phase shift at the oscillation frequency and enough gain to overcome the losses on the feedback network.

[@snnwstt]

Cutting the leads also reduces the enclosed area. Dramatically. So, it is the "same" effect in disguise (cutting the leads == reduction in the enclosed area) or is there more to it ? Would we get the same self resonnant frequency without cutting the leads, but making the leads really close (isolated by a sheet of paper) as example? or it is that the (straight) leads, as a cable in itself, are inductive too?

[@w2aew]

I believe the inductance will be lowered if you brought the leads together, also adds the capacitive coupling between the leads.

[@snnwstt]

Good observation, indeed, that would be an indirect and unwanted (for purpose of the intended measurement) modification to the original cap. Thanks.

[@wa4aos]

BTW, I found these distribution panels on ebay CHEAP, under $20. One has 10 and the other has 15 outputs at 75ohm for CATV work. Initially, I used it to reference 3 sig gens and 2 counters but as I add more devices, the sig is getting hammered even though these panels have amps. I built a freq divider to get 5 MHz's on one panel feeding both panels with a GPS standard backed with a very high res TXCO. All of my test gear is waiting for 50ohm inputs. 73, WA4AOS

[@FrozenArchLich]

Hello sir, Does 10dB attenuator attenuate gains (in decibel) at every frequency down 10dB? Thanks

[@w2aew]

+Nattapong Chotikorn In general - yes, attenuators are almost always broadband devices and provide the rated amount of attenuation over a wide frequency range.

[@user-fs8xx1cy5h]

Hi, is it possible to replace an attenuator resistor 1M?

[@brijeshyadav-mq2xy]

not recommended Стас Негруша, because output impedance of signal generator & input impedance of Attenuator is 50 ohm. so there will not be any reflection coming from the input of attenuator to signal generator because both having similar value.but if you use a 1M resistor instead of Attenuator as because of difference in output impedance of generator and input impedance of resistor large signal gets reflected back to generator @ RF or even higher frequency.

[@coilsmoke2286]

"signal gets reflected back to generator @ RF or even higher frequency." ...What *Would* that be , frequency greater than RF ?

[@abod4927]

Thanks for the effort that you put in your videos. I love your videos. don't you think that the input capacitance of the oscilloscope would give you a false resonance frequency?

[@w2aew]

The probe's input capacitance is only about 9-10pF, so as long as the capacitor value is substantially higher than that, the probe won't affect the self resonant frequency by a noticable amount.

[@abod4927]

Thanks for the explanation. :)

[@tcarney57]

Thanks once again!

[@timbo01889]

Hi there great videos have you ever seen any electronics books by Forrest Mims by any chance or are you related

[@w2aew]

Definitely not related to him, but I read many of his books (Notebook series) when I was just beginning in electronics in the late 1970s.

[@friedmule5403]

Great video as always!!:-) May I please ask you what if we do talk about GHz? It looked like your capacitor would be "100%" inductive?

[@w2aew]

True - you would use much lower value capacitors at GHz frequencies. They have a higher SRF.

[@absurdengineering]

When you go high enough, there will be several resonances, each reflecting a physical dimension or some physical aspect of the device, like electro acoustic coupling in capacitors, etc.

[@impsick]

good video.

[@s_amoku]

Thank you!

[@GOWTHAMKRISHGowthamShanmugaraj]

CAN YOU DO A VIDEO ON "IMPEDANCE MATCHING" ??

[@w2aew]

+GOWTHAM KRISH (Gowtham Shanmugaraj) It is on my list of topics to cover in the future.

[@szekerespista3758]

Sorry, i’’m a beginner in electronics. My question is: In a resonant circuit, using a capacitor at his resonance frequency is the best solution? (lowest ESR and lowest ESR). Thanks!

[@w2aew]

It is usually NOT a good idea to use a capacitor at its self-resonant frequency - mainly because this is not well controlled or predicted, and will vary by manufacturer and installation configuration.

[@szekerespista3758]

@@w2aew thank you for your response!

[@Muck-qy2oo]

And the grounding lead should also be shorter for these measurements!

[@placebomessiah]

Why would the signal amplitude drop if the impedance dips?

[@w2aew]

Because the signal generator has a 50 ohm output impedance. This forms a voltage divider with the load. For example, if the load is 50 ohms, then the voltage across the load will be 1/2 of the open-circuit voltage from the generator.

[@placebomessiah]

@@w2aew ohhhh....so the fact that the cap is resonating, effectively creating a short, it is ROBBING the scope of its signal??

[@Diegoescalante0]

If i don't have an attenuador. What can I do? I don't speak english so i don't understand a lot of words... And another question, if i connect the generator like you directly to the capacitor, the generator will not be in parallel with the capacitor, and the voltage in parallel it's the same right? Sorry and thanks a lot.

[@w2aew]

You can make a simple attenuator, I have a video showing that. Here is a link to the attenuator video: Basic RF Attenuators - Design, Construction, Testing - PI and T style - A Tutorial If you can't make one, just connect the generator directly to the capacitor.

[@Diegoescalante0]

***** Thanks, and If i have a generator capable of attenuating -20dB, works? or it isn't the same thing?.

[@w2aew]

Not the same. The purpose of the attenuator is mainly to provide a controlled termination for the generator and transmission line.

[@tomasmatejka3448]

​@@w2aew When I don't have any attenuator, can i connect directly to capacitor? I assume that there will be a reflection, but I think it will able to measure self-resonance. When I use 50 ohm termination in my oscilloscope, transmission line is terminated and i can measure well, isn't it?

[@w2aew]

@@tomasmatejka3448 It will work without the attenuator, but sometimes reflections can cause standing waves that will vary with frequency which might "fool" you.

[@nlimchua]

Bravo!

[@ShopperPlug]

That DIP meter is interesting, you do not plug anything to it and can meter the capacitor using RF.

[@laharl2k]

but that's provided u use a mcu. if you have just a cheap microprocessor with just io pins and 1 or 2 adcs. If you need to make a product and get the costs as low as posible then a mcu is an overkill. You'd end up having a lot of hardware you would never use but had to pay for. This way you can use transistors to connect and scan through the dc values, which is way cheaper (cash and cpu) than using a mcu with 1 adc for each button and interupts on top of that.

[@rapsod1911]

How to measure nH?

[@LHJARS]

Connect a capacitor of known value parallel to the inductance. Use a grid dip meter to determine the resonant frequency. From this value you can calculate the inductance.

[@acmefixer1]

Good video. But I was very disappointed that I didn't hear what the actual value of the capacitor was. What are you trying to hide?

[@w2aew]

I believe it was a 10nF cap

[@adminalane4303]

I had old one. 1959 old I want to sell

[@InsurgentX]

You continue to annoy me with the great videos. Quit forcing me to learn stuff! It hurts!