Reforming Capacitors - Everything you wanted to know

hello there this video covers the technical information process necessity and results of reforming capacitors put simply this is a procedure that takes capacitors of an unknown health and reconditions them to potentially be usable again this says by farb in the most controversial topic I've mentioned in any video and I've been waiting until I had all the resources and examples ready to give a comprehensive look into it several projects have come together recently to make this possible so I'm ready to discuss it if your opinion is to initially be against this process I encourage you to refrain from commenting until you finish the video and have seen the full technical explanation of it there aren't many complete explanations of capacitor reforming out there and a lot of misconceptions about it as a result so my goal is to lay out all the details explain how it works why you would want to do it and when it isn't worth doing I'm hoping this will finally definitively answer once and for all that reforming is worth doing and how you can go about doing it yourself when it makes sense [Music] the first thing we need to cover is some of the more technical information of how capacitors work because that is a critical component of understanding reforming electrical capacitance is created by having parallel electrical conductors the amount of capacitance is influenced by multiple factors it increases with more surface area of the conductors and as the distance between them is reduced the distance between them also affects the maximum rated voltage the substance between the conductors is considered an insulator and incapacitors is known as a dielectric as an insulator the dielectric will have a rated breakdown voltage this is a known voltage limit that the insulator can prevent the flow of electricity up to mostly if a voltage higher than the breakdown voltage for the thickness of the insulator is applied an arc will form between the two conductors this can be seen in examples like a Jacob's Ladder where air is the insulator and with a high enough voltage an arc will form between the two conductors we'll come back to this more later on but due to the breakdown voltage requiring an insulator thick great enough to withstand the voltages the capacitor needs to operate at you have a limit on how close you can make the conductors in the capacitor this means the surface area becomes a primary means to set the capacitance value of a capacitor the material used as a dielectric in capacitors varies by the chemical type of the capacitor for the purposes of this video our Focus will be on aluminum electrolytic capacitors due to the method of construction these capacitors have a fragile and variable dielectric which is the reason reforming may be needed these capacitors are built by taking two aluminum sheets with metal tabs for the legs and wrapping them around each other separated by sheets of paper the roll is put into a metal can with an electrically conductive electrolytic fluid in it the fluid is soaked into the paper getting in between the two sheets of aluminum a low voltage is applied and slowly raised to the operating voltage to the two legs over a period of time this is the Catalyst to form a chemical reaction that causes some of the electrolytic fluid to decompose and build an oxide layer on the positive input creating the anode of the capacitor the oxide layer is itself the dielectric and will completely cover the anode insulating it from the electrolytic fluid the negative side becomes the cathode and mostly does not have an insulating layer which means it conducts directly through the electrolytic fluid and the paper this makes the oxide layer the only insulator in the capacitor this entire process of building up the oxide layer is called forming and sets the maximum voltage for the capacitor but this isn't the end of the process if you've ever shopped for electrolytic capacitors you'll probably be familiar with their rated life and operating temperature if you haven't done that it's pretty simple because more is better it's well understood that electrolytic capacitors don't last forever there's a reason I built a whole website specifically for the purpose of cataloging capacitors and devices so you can replace them more easily but not everyone understands what goes into the rated life of a capacitor capacitors commonly fail by drying out or bulging or leaking but these are anomalous problems though as we can see from this failure mode chart from United chemicom when everything works correctly capacitors should be able to operate until they run out of electrolyte but normally this isn't caused by a leaking fluid which we can see chemicon considers a separate failure during the normal operation of an electrolytic capacitor the electrolyte itself is consumed by the capacitor and it's because of the oxide over time the oxide slowly deteriorates which would reduce the breakdown voltage except that the oxide is continuously rebuilt by a chemical process that depletes the electrolyte this is known as self-healing and is a normal process as capacitors are in use this can also be called reforming though as it's the same process when capacitors were initially constructed now let's finish this up by bringing it back to the rated life and temperature the higher the temperature the capacitor operates at the more quickly the oxide will Decay because the chemical processes are excess accelerated the expected lifetime of a capacitor is therefore the time it will take for the electrolyte to be depleted at the operating temperature as the electrolyte is depleted it will lose conductivity and as it is the cathode side of the capacitor the overall capacitance will go down if the cap is left running like this it will eventually destabilize the circuit is in or Mayfield short as the oxide is no longer being rebuilt that was a lot to go over but now we can talk about what happens with old capacitors the decomposition of the oxide layer on the anode of capacitors continues as they sit unused but without the self-healing action caused by having a voltage applied this means the insulator thickness slowly decreases over time reducing the maximum voltage the capacitor is able to handle if the capacitors are installed into a device when it is first powered on after long-term storage the capacitors will begin the self-healing process like normal but the byproducts of this will become a concern the capacitors can draw significantly more sustained current than usual and internally will heat up additionally the self-healing process releases hydrogen gas inside the capacitor and can build up pressure if the capacitor has dropped significantly below its rated voltage it may generate considerable amounts of heat and pressure enough to rupture the case this process after long-term storage is considered reforming even if you don't go out of your way to do any anything with the capacitors in a device that has not been powered for a long time this will still happen to them in it the high current draw and potential ruptures are risks that you take if you do not use precautions to ease the load onto capacitors before powering them the severity of the risks though are dependent on the values age and internal surface area of the capacitors but you could potentially over draw a power supply causing more issues or have the capacitors blow in the device causing potential physical and electrical damage these risks can be mitigated by changing the power going into the capacitors to reform them at a slower and controlled rate when it is worth doing for a device and its capacitors is not a firm rule but I would say it's worth considering when the potential energy of the capacitor approaches one joule here's a prime candidate for this a 100 000 microfarad 10 volt monster its lower voltage means the oxide layer was thinner to begin with and the massive capacitance means it has a large surface area which increases the likelihood for failure it also clocks in at potentially five joules deciding on the risk does require the Restraint of opening and inspecting the insides of the device before powering it to make this determination though if you've decided you do want to attempt safely reforming capacitors and have attempted to look it up you've probably seen two types of results either industrial applications that don't really apply to you or vintage audio gear enthusiasts recommending you slowly bring up the voltage what they're talking about is using a variac to reduce the AC Mains voltage to the device for True analog electrical equipment this is a valid option as the majority of the voltages will be divided down from that this option is not valid though for anything with a regulated power supply like computers and pretty much anything made after the 80s they require a specific input voltage to start generating their output voltages reducing the AC Mains voltage into these is more likely to cause more damage than help in these capacitors must be individually removed from the circuit and worked on independently the safest way to reform capacitors in an unknown condition is to slowly apply a rising voltage to them in much the same way the manufacturer did when they were first made this is the literal process of reforming there are actually standards for how to do this if you look at a data sheet for any reputable Japanese capacitor you'll see them say you should apply a voltage treatment which is reforming based on the JIS c5101-4 Clause 4.1 I would love to show you this standard but it is proprietary and you have to pay a lot for it so screw that the process for this is simple enough that even the United States Department of Defense has a guide on capacitor reforming in short you increase the voltage to maintain an optimal current until you reach the rated voltage of the capacitor as we would expect enough Theory though we've covered in enough groundwork to get into practice let's start with what you need to actually reform a capacitor there are three things required at minimum an adjustable power supply a multimeter and a resistor you don't need very high end or specific equipment for this these are some cheaper options that I have that I want to show you with first now for the resistor you really don't need to sweat it too much when picking one the purpose of the resistor is that it allows you to scale the current with the voltage acting as a current limiter the resistance used is not a hard value and has some fairly wide variants one manufacturer recommends approximately 100 ohms for capacitors rated for less than 100 volts and 1000 ohms when rated for more I've also heard of using resistors that are 100 to 1000 times the rated voltage of the capacitor lower will be better though if you're trying to work with a capacitor that will Top out the voltage range of your power supply I have been using a 200 Ohm resistor for most of the Caps I've been working with the basic setup I'm using is this can connect the output of the power supply to one side of the resistor and the positive terminal of the capacitor to the other then connect the positive side of the multimeter's current input to the negative terminal of the capacitor and the meter's common connection to the power Supply's negative side the multimeter can really go before or after the capacitor but this is how I've been doing it the multimeter is set up to measure current at a low scale at this point you need to decide on what your limit for acceptable current is that DOD guide I mentioned earlier recommends 5 milliamps while the TDK doesn't directly mention one but would produce much higher ones with the resistors they recommend as we'll see later I don't think the maximum value is something you need to worry about too much but I've been limiting mine to two milliamps but my final setup spends very little time near that now to demo this I'm going to need some old caps so I'm going to grab some from this dual 8 inch floppy drive to start there are some of the power supply and I especially like the pair of identical ones on the left measured before and after so I'll get those out of there which will is a fair bit more work than I expected it to be but I got them so now we can do some work with everything connected you can start if you have an analog power supply like this make sure you set it to zero volts before turning it on this multimeter can measure current without a battery so it's ready as well turn on the power supply and slowly raise the voltage until you see it Peak at the limit you decided on you'll see that it will quickly reduce down but the speed of that also ramps down even when it seems like it's stopped it is actually still reducing just very slowly this is the current breaking down the electrolyte to rebuild the oxide layer but there will always be some current draw here because capacitors have a leakage current where power will flow through them so you will need to pick a current minimum above that where you feel safe and has been reformed up to that point and this will vary depending on each capacitor once you are there just slowly raise the voltage up again until you reach your maximum current then stop and let it saddle again you repeat this process again and again until the voltage is increased to the rated voltage of the capacitor this cycle can take a long time both of the documents I mentioned before recommend an hour total of this but that is for newer capacitors the manufacturer dock is just for ones that are only two years old these are two decades old so I like to take it even slower since it takes a while I'm going to stop this and switch over to my better gear now I've been building up a lot of stuff with this use case in mind specifically primarily I'm using an HP 34401a DMM and an HP 6633a power supply for this like before I'm connecting the DMM for current measurements after the capacitor and resistor now I can just use the button panel on the power supply to directly set the voltage I want and read out the current clearly but there is another reason I use this Hardware it has gpib which makes it computer controllable the process we're doing here is not all that complicated if the current is below a raise the voltage to B using Ohm's law C resistor value and maximum current D so I wrote some software to automate this using this I can not only have the voltages automatically increase when needed but I can better understand what is happening by seeing the history of it these values are even logged which lets me open them in a spreadsheet program and chart them to see how it worked in action which actually gave me another idea and I coded a whole web front end for this with real-time plotting so uh yeah both the web version and the CLI versions of the software are available on GitHub as open source and I even remembered MIT licensing before release this time if you want to use these they are currently programmed for the interface language for the test equipment I have but it wouldn't take much work to adapt them now this software is performing the exact same processes I've been going over before when you start the web interface you have to configure the voltage resistor value and Min and Max currents then it will handle the rest of it looking at the charts you can see when the current reading dips below the minimum point the target voltage is adjusted up and the current shoots up with it but never beyond the maximum allowed values now that this is automated I'm going to let it finish off the first capacitor we've been looking at all the way up to 50 volts once it's done just for comparison I'll put it back on the analog gear and now you can see how much faster and further the needle Falls when measuring the current because it's not spending a bunch of time reforming now this still seems a little slow compared to a newer capacitor which doesn't let the needle jump up as much at all but that is a way newer cap rated to 200 volts so it has everything going for it to perform better now it could do the other caps on the logic board we saw to reform them all but there is another problem in these drives themselves which are Mains AC powered directly there are some very different capacitors rated over 300 volts it's not possible for me to reform these right now with some hacky stuff the highest voltage I've been able to reach was around 120 but if you have Parts like this you can either get a much higher voltage power supply or put several in series the series option is all I could do but 300 volts would be a lot of power supplies even if they were all 50 volts like the two I've shown so far now as I'm in the US these likely will only ever see 120 volts which I could potentially do but I'm also looking to buy a high voltage power supply so I may as well put this off for now all right now that you've seen this work and have tangible effects on the capacitor after reforming let me talk about when you shouldn't reform the devices I've been looking at here are presumed to have been put in storage just from being obsolete they were still likely fully functional the only thing they have wrong with them is age which is exactly what reforming helps with if I had inspected these capacitors and seen bleaks swelling ejected plugs or corrosion there would be absolutely no reason to attempt reforming because they have physical issues that cannot be solved they must be replaced if the capacitors don't show any visible signs of failure they are likely worth trying to reform based on the risk assessments that I mentioned earlier but this still doesn't mean they will be good the capacitor I fully reformed at 50 volts I personally think may have a higher ESR than normal since it seems like it's a little slow to respond it's also a bit difficult to measure ESR with something like this on higher capacitance Parts like this so I would recommend for any caps that you would attempt reforming that you connect an oscilloscope to the voltage rail after reassembly to look at the Ripple current on it and check if it's getting hot you'll have to make a judgment call on replacing it or not here though you can at least use the original or a modern replacement data sheet to see if it may be out of spec lastly if you attempt reforming and the capacitor draws a significant amount of continuous current or never has the current reduce down it's just past its usable life and has depleted the electrolytic fluids self-healing capacity it then must be replaced all right after all that I hope those who doubted the reforming process can see that it is worth making an attempt at and it has real outcomes that can help make the capacitors work after long-term storage it is of course an option to just wholesale replace all capacitors in an older device but then you have to research them to find suitable Replacements and for these monster capacitors you may be looking at a substantial cost to buy them as well so I will be reforming capacitors using the methods here when it makes sense to help get the older devices I have up and running and I had my eyes set on some already but that will have to wait for another video well I hope you guys enjoyed this in-depth look into reforming capacitors a kind of touchy and not very easy to research subject if you did you may want to subscribe to the channel to be notified when I release another video and if you want to help the channel I am on patreon but for now that's it and I will see you next time [Music]

Channel: Tech Tangents

Views: 183,265

Keywords:

Id: Z-jKsVbKFqM

Length: 19min 8sec (1148 seconds)

Published: Sat May 06 2023