Thanks for the video. I didn't realize there were LEDs that will light with only a microamp. But the longer glow from the electrolytic is simply due to the long time constant associated with its very large ESR. It doesn't really have anything to do with dielectric absorption which is associated with self recharging of a discharged capacitor as described in IEC 60384-1.
A very simple yet ingenius circuit. If there's leakage on the cap it acts as a resistance and LED glows depending on the leakage, why I never thought of that and it was under my nose. From this basis we can improve it so the indicator is more noticeable...I guess. Thanks!
I would love to build the tester Paul designed. I have hundreds of caps I collected over the years. Some of them definitely are bad, some are suspicious, but a lot should still be fine. Paul's device is able to predict the cap's future condition, which is amazing.
1959Berre He has a very simple inexpensive version that can be built on a terminal tie strip or protoboard. The 'deluxe' version is pretty inexpensive and easy to build as well. His videos have guided many to build their own, even after stating they didn't have the supplies or skills.
Does his amplify the leakage to make it more obvious? If you were to shunt the positive and negative feedback of an opamp with a current shunt, then use a 2.5 ohm power resistor to ground on the negative terminal, then connect to the positive terminal through the capacitor you are testing, you can charge caps fast and get very good sensitivity. A couple op amps in series to maximize gain if you want extreme sensitivity... for some reason... Something like this: www.falstad.com/circuit/circuitjs.html?cct=$+1+0.000005+49.687244006919805+50+5+43%0Av+96+80+160+80+0+0+40+12+0+0+0.5%0Ag+96+80+80+80+0%0Aa+208+96+368+96+9+2+0+1000000+0+0+100000%0Ag+208+144+208+176+0%0Ar+528+112+592+112+0+10%0A162+592+112+656+112+2+default-led+0+1+0+0.01%0Ag+656+112+672+112+0%0Ar+208+80+208+112+0+0.003%0Ar+208+144+208+112+0+2.5%0Aa+368+112+528+112+9+2+0+1000000+0+0+100000%0Ag+368+128+368+144+0%0Aw+208+0+208+80+0%0As+160+0+160+80+0+1+true%0A209+160+0+208+0+0+0.0047+0.001+1%0A
I wouldn’t think this would properly test for all leakage in a real-world sense as many caps have 25 to 400+ volts running through them in-circuit. At those voltages more leakage is likely to occur than at the low voltage this tester puts out. Nonetheless, what an absolutely wonderful “think outside the box” idea for a circuit! Fascinating use of the specialty LED and the device clearly works as designed at low voltages and I’m sure it’s useful for casual, quick tests. Thanks for an ingenious video and a great channel.
Great idea good job! You could pre charge the capacitors to the same power supply voltage as your circuit then plug them in and would get you a much quicker leakage current test on big caps!
My fast and dirty trick is to use my analogue needle meter to charge and then flip the probes on the capacitor to test it. Your idea is more appealing though.
Paul NJ I agree about using analogue meter. Its not just capacitors but also we can use it for finding leaking transistors diodes . I find it easier using analogue meters vs those digital meters when trouble shooting amps.😁
All electrolytic capacitors will leak a certain amount. It's true that as the capacitor reforms the amount of current will drop but it will NEVER go to zero current. There are some decent charts around that provide respectable limits to that current flow given the capacitance and voltage ratings of the cap. The higher the voltage, and the higher the capacitance rating the more current will flow on a perfectly good cap. The absolute best way to determine the maximum allowable leakage of a given electrolytic cap is the manufacturer's spec sheet. They all provide a formula for calculating that current value. The circuit is a cool idea, and looks like fun but with all due respect has very little value for testing allowable leakage on electrolytic caps.
John, as you have just demonstrated that this type of tool isn't just for Repairs on old equipment. You could be building a new circuit not knowing the condition of those new capacitors that you might have purchased from eBay. I have purchased a capacitor kit from eBay that was supposed to have very low ESR capacitors, out of the 10 470uf capacitors only 4 were actually low ESR! The other 6 were really bad, I was wondering why the next stage was saturated until I decided to check out the new capacitors, the problems went away as soon as I replaced the bad capacitors with the good ones that I finally checked. So basically! If you are working with electronics, old or new, you need the ability to check capacitors. Mr Carlson has built a really good capacitor checking device which uses low voltage to test and while he showed how good it works, I still am kinda leaning towards the old way of testing by using high voltage to it's rated voltage... Maybe I would change my mind if I knew how Paul's tester worked but for now I'm sticking to the old, albeit dangerous way because as you switch up through the voltages, you get instant feedback on how the capacitor is charging, whereas with Paul's tester, you have to wait not knowing if it is charging at all... Incidentally, with your circuit, you could add a bypass switch that allows you to charge the capacitor to it's rated voltage faster and possibly add another led or two to speed up the process with the larger capacitors! That way you can speed up the process and see that it's being charged at the same time, so when the Led's go out, you switch off the high speed bypass and the other led takes over to indicate whether any leakage is taking place... It's just a thought that I had while watching your video... Capacitor testing always makes for great watching...
it's normal for an electrolytic capacitor to have a leak, but it should be minimal, some vom with capacitance will test caps good or bad detecting high leakage as open or short or wrong caps measurement.
I noticed while working on an LED project that white LEDs would glow at very low currents, so I had to use some parallel resistance to allow the voltage to drop below the LED threshold.
put the LED inside a plastic tube 20-30mm (1 inch) long, like an old pen (not transparent obviously) and looking inside the tube you can notice even faint glows.. do it by myself and it works.
Well, Mr Carlson's video was a over one hour long advertisement ;-) Without any explanation, tutorial, schematics etc. Pretty boring for me. So, I prefer your video much more ;-)
@@trieck Mr. Carlson is the best. Not everything can be free. He deserves some support for all he demonstrates and explains. That capacitor tester he designed, tested and put up for sale is awesome.
Excellent - yes the time it takes a good electrolytic to cause the LED to go out is governed by CR, the time constant of the circuit. Dielectric absorption is a different affect. It can be thought of as a charged capacitor when discharged not discharging completely to zero volts - this is not due to the time constant of the circuit but the fact that some molecules in the dielectric remain polarised for longer than they should meaning that a directly shorted capacitor can still hold charge for quite some time. Mainly apparent in large value industrial capacitors in high voltage applications with the associated risk therein. Hope that helps explain things a bit and you have given me an idea for a video on exactly this phenomenon! Thanks for making the video - I enjoyed it! [BTW have you thought of pre-charging the capacitor under test to the supply voltage and then putting it in the tester to avoid the CR wait as it were?]
their are meters that only test capacitor. they cost like $35 however mr carlsons lab has meter that cost $600 that dont work so are these meters for testing capacitor actually work? consider reviewing them one is "Proster LCR Meter LCR Multimeter Tester for Capacitance Resistance Inductance Measuring Meter with LCD Over-range Display"
Indium gallium nitride LEDs in green will have the rich green color. If you are not sure, quality LED manufacturers will provide the information in the datasheet.
I cannot find this type of small LED on AliExpress. A new thing? Have recently been impressed that my 3mm blue LEDs will glow well with a 4.7k resistor, at TTL voltages .
Dear John, would you mind to share the LED part number? I would like to try this out but couldn't find the right LED with such low current capability. Thanks!
@@JohnAudioTech I checked the data sheet for both of these. They seem to be rather typical LEDs; Vf=3.2V-3.4V If=20-25mA. From the technical name that you used to refer to them, I thought that they were some type of higher technology. Now, I think the term has to do with their color and, yes, the green has the highest output intensity. Meaning, that they are a good quality LED, having testing and data, but they are not a special technology. Do I have that correctly? I am not intending to disparage the video. I just want to learn from it correctly. Thank you for all you do for us. Much appreciated.
@@t1d100 Yes, they are not special, just good consistent quality. I stressed that point because people often go to ebay or the other cheap sources and end up with parts that don't perform as intended.
charles klein It is a way to “reconstitute” the cap and restore it close to it’s original value/function by applying a series of increasing dc voltages to it over time. In my experience this is not a very permanent fix. Better to replace it.
A burning led is an indicator for nothing. How to test a cap out of an old tupe radio? Needs about 300Volt... hmmm! Or filtercap-cans.... 350Volt and 50microFarad..... This Video was good for intruducing the led.
Mr. Carlsons Lab has a great polarity tester for all types of caps. Thanks John for all the great stuff you post. I'm STILL enjoying my bench preamp
Thanks for the video. I didn't realize there were LEDs that will light with only a microamp. But the longer glow from the electrolytic is simply due to the long time constant associated with its very large ESR. It doesn't really have anything to do with dielectric absorption which is associated with self recharging of a discharged capacitor as described in IEC 60384-1.
A very simple yet ingenius circuit. If there's leakage on the cap it acts as a resistance and LED glows depending on the leakage, why I never thought of that and it was under my nose. From this basis we can improve it so the indicator is more noticeable...I guess. Thanks!
Thank you. This has given me food for thought and filled in a few blanks concerning capacitive materials.
Very cool, I built the Mr. Carlson's leakage tester and love it. Yours works well too with a kiss design.
Nice! Carlson is awesome. I was thinking of that tester while watching this.
I would love to build the tester Paul designed. I have hundreds of caps I collected over the years. Some of them definitely are bad, some are suspicious, but a lot should still be fine. Paul's device is able to predict the cap's future condition, which is amazing.
1959Berre He has a very simple inexpensive version that can be built on a terminal tie strip or protoboard. The 'deluxe' version is pretty inexpensive and easy to build as well. His videos have guided many to build their own, even after stating they didn't have the supplies or skills.
Does his amplify the leakage to make it more obvious?
If you were to shunt the positive and negative feedback of an opamp with a current shunt, then use a 2.5 ohm power resistor to ground on the negative terminal, then connect to the positive terminal through the capacitor you are testing, you can charge caps fast and get very good sensitivity. A couple op amps in series to maximize gain if you want extreme sensitivity... for some reason...
Something like this: www.falstad.com/circuit/circuitjs.html?cct=$+1+0.000005+49.687244006919805+50+5+43%0Av+96+80+160+80+0+0+40+12+0+0+0.5%0Ag+96+80+80+80+0%0Aa+208+96+368+96+9+2+0+1000000+0+0+100000%0Ag+208+144+208+176+0%0Ar+528+112+592+112+0+10%0A162+592+112+656+112+2+default-led+0+1+0+0.01%0Ag+656+112+672+112+0%0Ar+208+80+208+112+0+0.003%0Ar+208+144+208+112+0+2.5%0Aa+368+112+528+112+9+2+0+1000000+0+0+100000%0Ag+368+128+368+144+0%0Aw+208+0+208+80+0%0As+160+0+160+80+0+1+true%0A209+160+0+208+0+0+0.0047+0.001+1%0A
I thought your videos are great and you're a really good teacher.
I wouldn’t think this would properly test for all leakage in a real-world sense as many caps have 25 to 400+ volts running through them in-circuit. At those voltages more leakage is likely to occur than at the low voltage this tester puts out. Nonetheless, what an absolutely wonderful “think outside the box” idea for a circuit! Fascinating use of the specialty LED and the device clearly works as designed at low voltages and I’m sure it’s useful for casual, quick tests. Thanks for an ingenious video and a great channel.
Great idea good job! You could pre charge the capacitors to the same power supply voltage as your circuit then plug them in and would get you a much quicker leakage current test on big caps!
My fast and dirty trick is to use my analogue needle meter to charge and then flip the probes on the capacitor to test it.
Your idea is more appealing though.
Paul NJ I agree about using analogue meter. Its not just capacitors but also we can use it for finding leaking transistors diodes . I find it easier using analogue meters vs those digital meters when trouble shooting amps.😁
All electrolytic capacitors will leak a certain amount. It's true that as the capacitor reforms the amount of current will drop but it will NEVER go to zero current. There are some decent charts around that provide respectable limits to that current flow given the capacitance and voltage ratings of the cap. The higher the voltage, and the higher the capacitance rating the more current will flow on a perfectly good cap.
The absolute best way to determine the maximum allowable leakage of a given electrolytic cap is the manufacturer's spec sheet. They all provide a formula for calculating that current value.
The circuit is a cool idea, and looks like fun but with all due respect has very little value for testing allowable leakage on electrolytic caps.
John, as you have just demonstrated that this type of tool isn't just for Repairs on old equipment. You could be building a new circuit not knowing the condition of those new capacitors that you might have purchased from eBay. I have purchased a capacitor kit from eBay that was supposed to have very low ESR capacitors, out of the 10 470uf capacitors only 4 were actually low ESR! The other 6 were really bad, I was wondering why the next stage was saturated until I decided to check out the new capacitors, the problems went away as soon as I replaced the bad capacitors with the good ones that I finally checked. So basically! If you are working with electronics, old or new, you need the ability to check capacitors. Mr Carlson has built a really good capacitor checking device which uses low voltage to test and while he showed how good it works, I still am kinda leaning towards the old way of testing by using high voltage to it's rated voltage... Maybe I would change my mind if I knew how Paul's tester worked but for now I'm sticking to the old, albeit dangerous way because as you switch up through the voltages, you get instant feedback on how the capacitor is charging, whereas with Paul's tester, you have to wait not knowing if it is charging at all...
Incidentally, with your circuit, you could add a bypass switch that allows you to charge the capacitor to it's rated voltage faster and possibly add another led or two to speed up the process with the larger capacitors! That way you can speed up the process and see that it's being charged at the same time, so when the Led's go out, you switch off the high speed bypass and the other led takes over to indicate whether any leakage is taking place... It's just a thought that I had while watching your video...
Capacitor testing always makes for great watching...
Is that a cheaper device that the one, extremely sensible and powerful design by Mr. Carlson Lab?
😄😆😄
it's normal for an electrolytic capacitor to have a leak, but it should be minimal, some vom with capacitance will test caps good or bad detecting high leakage as open or short or wrong caps measurement.
I get so excited when you repeat the gallium nitroanalytical Led thingy.
It's usually a better idea to chuck a leaky capacitor instead of temporarily reforming it yes ?
Yes, I'd just get a newer one. I pulled that leaky one out of a bag of old caps I keep around for some reason.
I noticed while working on an LED project that white LEDs would glow at very low currents, so I had to use some parallel resistance to allow the voltage to drop below the LED threshold.
put the LED inside a plastic tube 20-30mm (1 inch) long, like an old pen (not transparent obviously) and looking inside the tube you can notice even faint glows.. do it by myself and it works.
Well, Mr Carlson's video was a over one hour long advertisement ;-)
Without any explanation, tutorial, schematics etc.
Pretty boring for me.
So, I prefer your video much more ;-)
couldn't agree more
Explanation? His Patreon site shows every imaginable detail you could ever want to know about it.
@@trieck Trouble is if I pay Patreon $2 a month my bank takes another $16.50 on top because of currency exchange fee.
@@harrystevens3885 can't you just use paypal?
@@trieck Mr. Carlson is the best. Not everything can be free. He deserves some support for all he demonstrates and explains. That capacitor tester he designed, tested and put up for sale is awesome.
In the 3rd approximation there are also inductors in the drawing of the cap besides added resistamce and capacitance but it is negligible.
Thanks! Interesting LED.
Is this inspired by Mr Carlson's lab leakage tester?
Yes... he mentioned it at the end of the video.
He does indeed. I should have watched the whole video first
This vid is Old put I just watched it. can you design a tester for 1000v + Caps using this Led?
I tried this but when i cut power to see if led would drain the capacitor and light up
it didnt work. It took know time to charge 300 and 47 uf caps.
I suggest u use your multimeter same test but much better indicator specialy if you use a analoge multimeter in the range of 10K or 100K setting
Excellent - yes the time it takes a good electrolytic to cause the LED to go out is governed by CR, the time constant of the circuit. Dielectric absorption is a different affect. It can be thought of as a charged capacitor when discharged not discharging completely to zero volts - this is not due to the time constant of the circuit but the fact that some molecules in the dielectric remain polarised for longer than they should meaning that a directly shorted capacitor can still hold charge for quite some time. Mainly apparent in large value industrial capacitors in high voltage applications with the associated risk therein. Hope that helps explain things a bit and you have given me an idea for a video on exactly this phenomenon! Thanks for making the video - I enjoyed it! [BTW have you thought of pre-charging the capacitor under test to the supply voltage and then putting it in the tester to avoid the CR wait as it were?]
their are meters that only test capacitor. they cost like $35 however mr carlsons lab has meter that cost $600 that dont work so are these meters for testing capacitor actually work? consider reviewing them one is "Proster LCR Meter LCR Multimeter Tester for Capacitance Resistance Inductance Measuring Meter with LCD Over-range Display"
what is reformation?
i noticed on your diagram you had the cap to be tested on positive side, but on breadboard you were testing on negative side. great video though
Good catch. It will work either way.
How do you know wether the LED is the indium gallium nitride type?
Indium gallium nitride LEDs in green will have the rich green color. If you are not sure, quality LED manufacturers will provide the information in the datasheet.
I cannot find this type of small LED on AliExpress. A new thing?
Have recently been impressed that my 3mm blue LEDs will glow well with a 4.7k resistor, at TTL voltages .
Cause it looks like one....lol
Very nice. How would you test an electrolytic cap. rated at 450 V? thanks
At 450 volts I would buy a DMM that can measure capacity and a ESR meter. Safely discharge them first. 450 volts can be lethal.
Pretty cool. Makes sense
Dear John, would you mind to share the LED part number? I would like to try this out but couldn't find the right LED with such low current capability. Thanks!
OVLBG4C7 (3mm) OVLFG3C7 (5mm) Green high brightness LEDs. I got them at DigiKey.
@@JohnAudioTech thanks!
@@JohnAudioTech I checked the data sheet for both of these. They seem to be rather typical LEDs; Vf=3.2V-3.4V If=20-25mA. From the technical name that you used to refer to them, I thought that they were some type of higher technology. Now, I think the term has to do with their color and, yes, the green has the highest output intensity. Meaning, that they are a good quality LED, having testing and data, but they are not a special technology. Do I have that correctly? I am not intending to disparage the video. I just want to learn from it correctly. Thank you for all you do for us. Much appreciated.
@@t1d100 Yes, they are not special, just good consistent quality. I stressed that point because people often go to ebay or the other cheap sources and end up with parts that don't perform as intended.
nice, thanks ! is this how capacitor testers work ?
depends on the tester. but the most simple ones, yes
can it be done with opamp with its inputs across resistor?
Is the 4148 actually fast enough to provide protection ?
Yes. It is a fast switching diode.
Any diode should be used,
Fast diode is fast on reverse bias only which is not in our case.
Nice info, thanks :)
Why not just measure the current with a multimeter on peak hold? Mine goes down to 0.01 uA resolution.
I popped a cap in my face today, watch out for it. (make sure u check what the voltage is! BOOOM! cop it in the eye would be real bad.)
Useful video
REFORM? WHAT IS IT
charles klein It is a way to “reconstitute” the cap and restore it close to it’s original value/function by applying a series of increasing dc voltages to it over time. In my experience this is not a very permanent fix. Better to replace it.
Ya blast all of the smoke causing stuff out of it by using it like a spark plug....lol
nice work
"Acrost" :D
Hiiiii how are you
The word is 'across'
Not "acrosst "
Rather use a cap tester.
Stop saying "acrossed". ;)
Why not just put a micro-current meter in series Mr Carson with all his junk rules
Yes you could as long as you protect it from the charging surge current with a resistor.
most vom today has a capacitance test features, no need to do this circuit to find out capacitor leakage problem.
Most multi-meters will test capacitance. They do not test for leakage.
A burning led is an indicator for nothing. How to test a cap out of an old tupe radio? Needs about 300Volt... hmmm! Or filtercap-cans.... 350Volt and 50microFarad.....
This Video was good for intruducing the led.