That was a very informative video. I've been hearing a lot about ESR lately and it was really clarifying to see it on an oscilloscope. That was a very clever and educational way to demonstrate ESR and show exactly the effect it can have! Keep up the great videos John!
Realy interesting video, very educationel, did not know that the difference between different cap types was that great. Looking forward to the next video, thanks. Greatings from Denmark :-)
I wish I had seen this several years ago when I was experimenting with the lm386 making an audio amp. I have since moved on to building and flying drones. I wonder how the x7r would help improve the video feed that gets distorted by emf from the electronic speed controllers and brushless motors on the drone? The FPV community are using the barrel-type electrolytic 440 - 1000uf. Can you give a reference to an X7R ceramic capacitor rated up to 24 volts?
So in your circuit, if the capacitor had no ESR of if you just connected your test probes together, would I see just a flat line? How much voltage and current does that circuit supply? Thanks for the educational videos!
Usually you can do this for power supply filter capacitors BUT the large capacitance is going to have a larger "inrush" current when you first hit it with 12 volts but in this case it seems harmless.
You frequency is too low to measure those small caps. The triangle is because the cap is being fully charged and discharged on each cycle. You're not seeing the resistance because the result is clipped off by the capacitor (ie. it's a high pass filter).
It would be relative to a cap of the same type. For example, that 330uf Xicon cap was not bad, it was just not a low ESR type. It measured about 17mv across its ESR in my jig, however, if it measured 50mv, I'd consider it likely to be failing.
Not very easy. Speakers suffer power compression when the voice coil starts to heat up. It might be possible to measure that with some risk of damage. If the driver has a part number, you can search online.
So i guess the power of a speaker depends on the frequence. first find the ideal frequence of the speaker like you explaned in one of youre video's. And then use that frequence to test the maximum power. (with ear caps :-) )
JohnAudioTech I wonder if you could by monitoring the temperature of the magnet and increase the power slowly? Or maybe use a mic with real time analyzer app to look for change in noise content?
Many thanks for another great video. One thing I've always wondered about polarized caps like electrolytics; They must of course be DC biased the correct way. But what if an electrolytic is biased with 0V DC, and then as the frequenzy of the signal passing through it goes down - what happens then? Say a signal of 0.1Hz - wouldn't it look pretty much like DC to the cap? So it would be biased the wrong was about 50% of the time. How low a signal can such a cap handle? Many thanks.
Polarized electrolytic capacitors require DC voltage and in the application you describe its positive side, for instance, would idle at 1/2 the power supply voltage and as the signal came through it would vary down to nearly zero and up to nearly the supply voltage. On the other side it becomes AC (alternating current) but that's okay because the capacitor *itself* never becomes reversed polarized. Of course, non-polarized electrolytic capacitors exist for situations where a DC bias is not present.
I work on power supplies that I take in for repair. On one type I get in on a regular basis I see a lot of spikes like you have on your scope. Problem with this power supplies are that some resisters are overheating. these are resistors that are meant to heat up and their off set high off of the board but they heat up to 300- 400° and overtime day darken their surrounding board area. What would you suggest can be done about this. By the time I get these power supplies they're not working due to component failure. I can locate and replace the failed components. but I don't seem to be able to figure out how to lower the heat of these resistors. Any suggestion would be greatly appreciated.
Persistent overheating of a power supply regulator suggests excessive voltage in the transformer and the transistors and resistors must regulate this down to whatever is the specified voltage. Spikes suggest an inductive load somewhere in the mix, particularly with switching power supplies. A small capacitor in the right place (more than one probably) and probably with a modest resistor in series becomes a "snubber" to control these spikes.
Fantastic video, thanks! I have heard some audio-experts talk about the capacitors reaction time. That a capacitor with large capacity are longer time to "react" then a small one, therefor it better to have different capacities so the small capacitors can start delivering while the large are getting ready. Is that right?
Fried Mule I think John demonstrated that this is not true. The same value of capacitance in 3 different types of capacitors provided much different response. If the point is to keep the signal clean, it would appear that the ceramic capacitor was even better than the poly and both were far better than the aluminum electrolytic. But if you need bulk capacitance, it is hard to find huge capacitance in ceramic, so you can parallel an aluminum with a ceramic to get the best of both worlds. Make sense?
"Is that right?" That is correct. It is common to parallel a small capacitor across a big electrolytic capacitor. The "getting ready" you speak of is actually inductance in a rolled-up electrolytic capacitor. You can see some odd behavior in the oscilloscope that appears not to be explained. That's most likely the parasitic inductance of the capacitor. Abrupt changes in current will produce a voltage spike on an inductor. So, there's inductance present, not just series resistance.
@@thomasmaughan4798 Thank you very much for your great reply! But would that not mean it is better with lets say 10 small caps than 1 large and 1 small cap? Assuming both solutions have the same total value.
@@friedmule5403 "But would that not mean it is better with lets say 10 small caps than 1 large and 1 small cap?" It *might* depending on various circumstances. Large capacitors tend to have significant inductance that can be a problem at higher frequencies but is not a problem at low frequencies more commonly encountered with large capacitors used as power supply filter caps. The small mylar or cereamic capacitor is a high frequency bypass to keep stray radio signals out of the voltage regulator since that can produce oscillation that destroys the regulator and whatever is hooked up to that power supply. But there's a gotcha: the inductance of the large capacitor will form a "tank circuit" and resonate with the small capacitor at a specific frequency and at that exact frequency it will not only not bypass, it will enhance that signal. So the "Q" is reduced by a series resistor, a low value so we don't lose bypass effectiveness. A VNA (vector network analyzer) or BODE plot can determine the frequency at which this takes place; not that its a big problem but if you have CB or ham radio (for instance) then it could become important.
ESR is also problematic in tweeter X-overs 4 ohm ESR dulls the treble, I put in WEAH 2.7uF poly caps and the difference was clear ( I used the "singing Budgie's" Locomotion) as the test record. Kylie Minogue was nick named that in 1987 and hated that nickname
Interesting, I think the capacitors used in a power supply like and audio amp you need to bring the frequency to double the line frequency. Note that is for the Power supply. As for a SMPS PWS you need caps with low ESR for high frequencies depending on the frequency produced by the SMPS PWM circuit. I am sure you get some different results with different frequencies.
Electrolytics are used in audio power supplies a lot because of the much higher capacitance. They are trying to minimize ripple as much as possible. Ceramics are also avoided to avoid microphonics, but they are sometimes used right next to the op amp if a switching power supply is used. In audio amps with linear power supplies, you won't see ceramic capacitors because they aren't necessary, and it's better to not introduce the potential for microphonics.
You might of been able to reduce some of the inductive effects by using an old fashion Carbon composition (core) type resistor instead of that Carbon film type with it's spiral cut.
I can solder the circuit on proto board with better components for improved performance but this is thrown together for the video and to test the caps that came with the LM1875 kit. You could (should) use higher current drive for the low ESR caps to see the differences, but the idea here to keep it all the same to show the relative performance differences.
When scope does not trigger because of "small signal" make the signal "big" bij lowering Volt /div one step it wil trigger better. Or move with trigger level knob tp the middle of the wave - it will trigger beter
Man I wish I were more like my dad. However, being myself I got some basic info out of this, which kind of confuses me more. I'm trying to buy PSU caps for my DEQ2496 (smps). I could really use a consult. Most EEs are insulted if money gets offered, but I want to do my due diligence in the matter and not just do what everyone else is doing to fix their DEQs. So if you wouldn't mind looking at my schematics and recommending caps I'd be willing to pay you for your time. If this is insulting to you, my apologies upfront.
By this test, ceramic caps are best. However, I just watched one of Mr. Carlson's shows where he says they are not a good choice for signal path due to high microphonics. I guess even the top contender components each have their own unique shortcomings, so I guess you have to pick your poison and go with it.
That was a very informative video. I've been hearing a lot about ESR lately and it was really clarifying to see it on an oscilloscope. That was a very clever and educational way to demonstrate ESR and show exactly the effect it can have!
Keep up the great videos John!
Realy interesting video, very educationel, did not know that the difference between different cap types was that great.
Looking forward to the next video, thanks.
Greatings from Denmark :-)
Very educational video.
Really enjoyed it,simple to understand but not too simple to skip basics.
Keep up great work...
Cheers
Love the outtakes at the end. Good kitty! Oh, and thanks for the informative video.
Thank you, great video. When you measured the ESR voltage drop for the bad capacitor, how can you from that value calculate the ESR in ohms?
Thaks a lot. Very technical yet short and simple explained.
This was a very enjoyable video. Looking forward to the next video. And hello to Snickers from me!
Excellent demo, Thank you very much John
Fine video. Also, enjoyed outtakes and guest star SNICKERS.
I wish I had seen this several years ago when I was experimenting with the lm386 making an audio amp. I have since moved on to building and flying drones. I wonder how the x7r would help improve the video feed that gets distorted by emf from the electronic speed controllers and brushless motors on the drone? The FPV community are using the barrel-type electrolytic 440 - 1000uf. Can you give a reference to an X7R ceramic capacitor rated up to 24 volts?
wish i could thumbs up this videos hundred times.
Very good informative video, just noticed I didn't knew that much about capacitors, thanks.
super video, it's not too fast, perfect! Thanks for sharing!
Even the outtakes! Neat!
Great video! Feel bad for Mr. Inductance, he got blamed for everything!
Great video, very informing. Thanks!
So in your circuit, if the capacitor had no ESR of if you just connected your test probes together, would I see just a flat line? How much voltage and current does that circuit supply? Thanks for the educational videos!
Yes (ideal).
That "yellow" ceramic capacitor was an "MLCC(Multi-Layered-Ceramic-Capacitor)" type, NOT the standard cheap single ceramic disc type.
Hello sir
Can I replace 16v 1000uf to 16v 3300uf ?
Atx power supply for car amplifier
Usually you can do this for power supply filter capacitors BUT the large capacitance is going to have a larger "inrush" current when you first hit it with 12 volts but in this case it seems harmless.
Thank you very much, Wish you successful and glorious!.
The switching noise is caused by the capacitor self resonating from the high frequency content of the switching transition.
You frequency is too low to measure those small caps. The triangle is because the cap is being fully charged and discharged on each cycle. You're not seeing the resistance because the result is clipped off by the capacitor (ie. it's a high pass filter).
indeed
Loved the video. So, higher ESR does/doesn't mean the capacitor is going bad?
It would be relative to a cap of the same type. For example, that 330uf Xicon cap was not bad, it was just not a low ESR type. It measured about 17mv across its ESR in my jig, however, if it measured 50mv, I'd consider it likely to be failing.
JohnAudioTech I guess the datasheet needs to be referenced to determine the design tolerance then. Thx!
Hi john, how about two eletrolithic caps in serie, like (+-solded-+) or (-+solded+-)?
Hi John. :-)
I have a question.
Is there a way to measure the max rms power that a speaker can handle witout the risk of overloading it ?
Not very easy. Speakers suffer power compression when the voice coil starts to heat up. It might be possible to measure that with some risk of damage. If the driver has a part number, you can search online.
So i guess the power of a speaker depends on the frequence. first find the ideal frequence of the speaker like you explaned in one of youre video's. And then use that frequence to test the maximum power. (with ear caps :-) )
JohnAudioTech I wonder if you could by monitoring the temperature of the magnet and increase the power slowly? Or maybe use a mic with real time analyzer app to look for change in noise content?
Ceramics still have that feature of piezoelectric sensitivity, so they're okay for the power rails but not the audio stream.
All videos verry verry useful
Thanks, that helped me a lot
Many thanks for another great video. One thing I've always wondered about polarized caps like electrolytics; They must of course be DC biased the correct way. But what if an electrolytic is biased with 0V DC, and then as the frequenzy of the signal passing through it goes down - what happens then? Say a signal of 0.1Hz - wouldn't it look pretty much like DC to the cap? So it would be biased the wrong was about 50% of the time. How low a signal can such a cap handle? Many thanks.
Polarized electrolytic capacitors require DC voltage and in the application you describe its positive side, for instance, would idle at 1/2 the power supply voltage and as the signal came through it would vary down to nearly zero and up to nearly the supply voltage. On the other side it becomes AC (alternating current) but that's okay because the capacitor *itself* never becomes reversed polarized.
Of course, non-polarized electrolytic capacitors exist for situations where a DC bias is not present.
I work on power supplies that I take in for repair. On one type I get in on a regular basis I see a lot of spikes like you have on your scope. Problem with this power supplies are that some resisters are overheating. these are resistors that are meant to heat up and their off set high off of the board but they heat up to 300- 400° and overtime day darken their surrounding board area. What would you suggest can be done about this.
By the time I get these power supplies they're not working due to component failure. I can locate and replace the failed components. but I don't seem to be able to figure out how to lower the heat of these resistors.
Any suggestion would be greatly appreciated.
Persistent overheating of a power supply regulator suggests excessive voltage in the transformer and the transistors and resistors must regulate this down to whatever is the specified voltage. Spikes suggest an inductive load somewhere in the mix, particularly with switching power supplies. A small capacitor in the right place (more than one probably) and probably with a modest resistor in series becomes a "snubber" to control these spikes.
nice video! can you post the specs of all the caps you tested?
Strange. The PNP BC327 should be put on the high side bridge for CAP charging. And NPN should be put on the low side for discharge.
Thank you, nice explanations.All the best.
Cheers John.
Fantastic video, thanks!
I have heard some audio-experts talk about the capacitors reaction time.
That a capacitor with large capacity are longer time to "react" then a small one, therefor it better to have different capacities so the small capacitors can start delivering while the large are getting ready.
Is that right?
Fried Mule I think John demonstrated that this is not true. The same value of capacitance in 3 different types of capacitors provided much different response. If the point is to keep the signal clean, it would appear that the ceramic capacitor was even better than the poly and both were far better than the aluminum electrolytic. But if you need bulk capacitance, it is hard to find huge capacitance in ceramic, so you can parallel an aluminum with a ceramic to get the best of both worlds. Make sense?
"Is that right?"
That is correct. It is common to parallel a small capacitor across a big electrolytic capacitor. The "getting ready" you speak of is actually inductance in a rolled-up electrolytic capacitor. You can see some odd behavior in the oscilloscope that appears not to be explained. That's most likely the parasitic inductance of the capacitor. Abrupt changes in current will produce a voltage spike on an inductor. So, there's inductance present, not just series resistance.
@@thomasmaughan4798 Thank you very much for your great reply! But would that not mean it is better with lets say 10 small caps than 1 large and 1 small cap? Assuming both solutions have the same total value.
@@friedmule5403 "But would that not mean it is better with lets say 10 small caps than 1 large and 1 small cap?"
It *might* depending on various circumstances. Large capacitors tend to have significant inductance that can be a problem at higher frequencies but is not a problem at low frequencies more commonly encountered with large capacitors used as power supply filter caps. The small mylar or cereamic capacitor is a high frequency bypass to keep stray radio signals out of the voltage regulator since that can produce oscillation that destroys the regulator and whatever is hooked up to that power supply.
But there's a gotcha: the inductance of the large capacitor will form a "tank circuit" and resonate with the small capacitor at a specific frequency and at that exact frequency it will not only not bypass, it will enhance that signal. So the "Q" is reduced by a series resistor, a low value so we don't lose bypass effectiveness.
A VNA (vector network analyzer) or BODE plot can determine the frequency at which this takes place; not that its a big problem but if you have CB or ham radio (for instance) then it could become important.
@@thomasmaughan4798 Thank you so very much, this makes so much sense! Are you a teacher or something? :-)
How about tantallum capacitors?
Nice vídeo!
Cool. I got the same cat, I mean scope; both actually.
Can replace with Ic 555 ???
Cats ALWAYS "pick a fine time."
6:30 "About 10 volts of ESR in there". So resistance is now expressed in volts.
ESR is also problematic in tweeter X-overs 4 ohm ESR dulls the treble, I put in WEAH 2.7uF poly caps and the difference was clear ( I used the "singing Budgie's" Locomotion) as the test record. Kylie Minogue was nick named that in 1987 and hated that nickname
Interesting, I think the capacitors used in a power supply like and audio amp you need to bring the frequency to double the line frequency. Note that is for the Power supply. As for a SMPS PWS you need caps with low ESR for high frequencies depending on the frequency produced by the SMPS PWM circuit. I am sure you get some different results with different frequencies.
Electrolytics are used in audio power supplies a lot because of the much higher capacitance. They are trying to minimize ripple as much as possible. Ceramics are also avoided to avoid microphonics, but they are sometimes used right next to the op amp if a switching power supply is used. In audio amps with linear power supplies, you won't see ceramic capacitors because they aren't necessary, and it's better to not introduce the potential for microphonics.
he gives out good info.
Thank you.
You might of been able to reduce some of the inductive effects by using an old fashion Carbon composition (core) type resistor instead of that Carbon film type with it's spiral cut.
I also wonder about the 22 Ohm value, how you come up with that and if it's value should be changed for testing electrolytic VS small caps?
I can solder the circuit on proto board with better components for improved performance but this is thrown together for the video and to test the caps that came with the LM1875 kit.
You could (should) use higher current drive for the low ESR caps to see the differences, but the idea here to keep it all the same to show the relative performance differences.
JohnAudioTech well done!
Would a 555 Timer work as a Square waver generator for this CCT?
It could work if the rising and falling edges are fast enough.
When scope does not trigger because of "small signal" make the signal "big" bij lowering Volt /div one step it wil trigger better. Or move with trigger level knob tp the middle of the wave - it will trigger beter
Man I wish I were more like my dad. However, being myself I got some basic info out of this, which kind of confuses me more. I'm trying to buy PSU caps for my DEQ2496 (smps). I could really use a consult. Most EEs are insulted if money gets offered, but I want to do my due diligence in the matter and not just do what everyone else is doing to fix their DEQs. So if you wouldn't mind looking at my schematics and recommending caps I'd be willing to pay you for your time. If this is insulting to you, my apologies upfront.
I inductance only am AC product?
By this test, ceramic caps are best. However, I just watched one of Mr. Carlson's shows where he says they are not a good choice for signal path due to high microphonics. I guess even the top contender components each have their own unique shortcomings, so I guess you have to pick your poison and go with it.
Nice!
You should of showed what it looks like to put a ceramic cap in parallel with an electrolytic to gain the best of both worlds.
Yes, I am doing this in my audio amp project.
很棒, 多謝
We call that a "Cat-Scan"
ЕSR нужно мерить синусойдой. 250 кГц это много. Большие емкости это 100кГц, а маленькие от 100Гц.