Electrolytic Capacitor kit for repairing TV, Laptop etc?

[anoeev]

Hi DIY folks

I'm looking for a basic Kit of electrolytic Capacitors to learn to fix things such as TV, Radio, laptops.
I was wondering if a basic 1uF-1000uF kit from aliexpress is alright or will the quality be so bad that they won't work as replacement electrolytic capacitors?

A big pack could last many years and be used for DIY projects.
https://www.aliexpress.com/item/4000579643509.html?spm=a2g0o.productlist.0.0.14c321c3mU9mMB&algo_pvid=4b06afbd-d6b2-4dee-935d-2149cc374a89&algo_expid=4b06afbd-d6b2-4dee-935d-2149cc374a89-5&btsid=0bb0624716119534723722257e4f74&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

Let me know your thoughts and also which set you recommend
Electrolytic capacitors probably won't be the only type of capacitor i would need.

[Zero999]

No, I wouldn't recommend it. You don't know what crap it contains. It's worth spending extra on decent electrolytic capacitors. Cheap ones often fail and have high (ESR) effective series resistance.

A decent kit is good for prototyping, but no good for repairs. There are too many possibilities, regarding temperature & voltage ratings, value and very importantly case size.

[anoeev]

Which Capacitors would you recommend? I've found kits of several capacitors from 1uF to 1000uF that are more expensive, but i suspect they are still from China just sold under another name.

[james_s]

I keep a few common values around for projects but for repairs I usually buy what I need when I need it so I have fresh parts from reputable suppliers. Most of those cheap capacitors from China are garbage, I cringe whenever I encounter something someone has "recapped" by replacing a bunch of older but good quality parts with the cheapest new junk around. There are many good capacitors out there but my personal favorites are Nichicon, Rubycon and Panasonic. I've had good luck with all three.

[BrokenYugo]

Assuming we aren't talking about 40+ year old equipment, most failed electrolytic caps seem to be in positions where a low ESR cap is required, and/or high temp/high hour. And if we are talking old stuff you probably want something that will last (high temp/high hour) anyway. These are relatively expensive parts you aren't going to find in a big cheap assortment kit in acceptable quality like you can with resistors, small signal transistors, etc. As mentioned the cheap kits are going to be ok for playing around building your own stuff though, especially if you have some means of testing them (A cheap "transistor tester" component tester beats nothing). 

For repair work it's best to buy as needed from reputable vendors (in the US that'd be Mouser, Digigkey, etc.), maybe build up a small inventory for substitution. Barring that I'd rather use tested parts pulled from scrap than eBay grade stuff.

[TimNJ]

Not sure what electronics distributors are available to you in Norway. I think I remember that, for some European countries, might make more sense to buy from Chinese distributor as opposed to getting from the US, etc. You have TME though, right?

Murphy's law will usually get you. You won't have the value you want/need, or you'll have it in the wrong size, etc. I wouldn't bother maintaining a stock on the basis of "order once, set for life".

For replacing caps, it's important to match the ESR of the cap you are replacing. i.e. Check the datasheet of the cap you are replacing and aim to replace with similar or slightly lower ESR. Also, be mindful of the endurance rating. i.e. Stay away from 1,000 or 2,000 hour rated caps. Recommend to use 5,000 - 10,000 hour caps if you can get them.

As others have said, Nichicon, Rubycon, United/Nippon Chemi-con, Panasonic are all good. But, you'd honestly be just as fine with reputable Chinese/Taiwanese brands like Samxon, Taicon, Aishi, or Lelon, provided you pick a decent series from them...which again, goes back to selecting proper ESR and endurance rating.

But, just buy them each time you need them, and maybe buy a few extra each time. You might get lucky next time around, with something you bought before.

[Per Hansson]

I would avoid capacitor kits from China like the plague!
There is a reason you are replacing the caps to begin with, it is because they came from the same shop in a back alley in the Shenzhen market.
The only capacitors to this day that can be generally trusted come from Japanese brands: and the Chinese love to copy them!
So if you buy Rubycon, Nichicon, Sanyo/Suncon or United Chemi-Con and they come from China via Aliexpress I can guarantee you they have a high percentage risk of being counterfeits!
There are some small stores available to you, if you don't wanna buy from Elfa (that sadly moved their warehouse away from Sweden) or TME as was suggested or Mouser then you can try Electrokit.
It is a small Swedish reseller that stocks some common parts otherwise sold by Elfa and the likes, though for capacitors their stock is quite limited.
I would just do as was suggested and buy from the large trusted sources and each time you need buy a few extra, most give you free shipping if you spend like $50.
So it is a good excuse to buy some extra caps

[Siwastaja]

Capacitors fail because the original designers failed to design correct capacitors in. They were professionals, maybe with hard-to-meet cost constraints, but professionals nevertheless. You need to be better.

Replacing with random parts will result in a "repair" which will likely last for much shorter time than the original. OK in a pinch if you just need some weeks of operation out of a device you are going to replace anyway.

You need to understand the circuit and why the part failed, what were the ratings and the quality of the failed part, then substitute with something better suited.

Generally, if the capacitor is in a high ripple current switch mode converter and the original part is a no-name low-cost brand, replace with Panasonic FR, FM series or something similar, with similar or slightly higher ripple current ratings.

If the original part was already a good known brand, then the original problem likely wasn't low quality part, but choosing a part with ripple current or temperature rating too small, then you need to substitute with one having better ratings, i.e. higher ripple current rating, higher lifetime and/or temperature rating.

[MathWizard]

Hi DIY folks

I'm looking for a basic Kit of electrolytic Capacitors to learn to fix things such as TV, Radio, laptops.
I was wondering if a basic 1uF-1000uF kit from aliexpress is alright or will the quality be so bad that they won't work as replacement electrolytic capacitors?

A big pack could last many years and be used for DIY projects.
https://www.aliexpress.com/item/4000579643509.html?spm=a2g0o.productlist.0.0.14c321c3mU9mMB&algo_pvid=4b06afbd-d6b2-4dee-935d-2149cc374a89&algo_expid=4b06afbd-d6b2-4dee-935d-2149cc374a89-5&btsid=0bb0624716119534723722257e4f74&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

Let me know your thoughts and also which set you recommend
Electrolytic capacitors probably won't be the only type of capacitor i would need.

To get you out a of a jam, I'd say most cheap capacitors, as long as they are rated for what their label says......they should be ok for short term, especially ceramic caps, and low power caps, and DC blocking caps, etc.

But for long term replacements in things like a Switch Mode Power Supply, like in computers, TV, pretty much everything these days that's low power consumer electronics, etc...or high voltage tube radios/etc.....I'd say then it's much safer to order more expensive brand name caps, with the true proper spec's for the job.

I have some cheap caps of most types, a few nice caps from salvage that are still good, and hardly any caps that I could vouch for tho, if they wanted top join the Mafia I guess.

Do I regret buying cheap caps from ebay over the last few years ? I only remember blowing up 2, and that was by mistake. But that was just VERY VERY VERY low power stuff, like a few Watts. So no, I guess not for breadborading and the odd repair.

I don't have any electrolytic cap's I trust for any length of time in a computer SMPS, or TV/monitor SMPS. And hardly have any caps I'd call accurate for timing.

If the cap is in a high power, and/or /high temp/frequency setting, and it's an electrolytic cap, then get something brand name for long term confidence.

[Renate]

For replacement parts, especially on stuff that you're going to repair for other people I don't like to screw around with iffy components.
I bought this electrolytic assortment from Digi-Key with caps from Panasonic.
https://www.digikey.com/en/products/detail/panasonic-electronic-components/M2-KIT/4383781
(Funny, they don't seem to have a price on it now.)
When I use some of the caps I just put some more on my next order.

[james_s]

To get you out a of a jam, I'd say most cheap capacitors, as long as they are rated for what their label says......they should be ok for short term, especially ceramic caps, and low power caps, and DC blocking caps, etc.

But for long term replacements in things like a Switch Mode Power Supply, like in computers, TV, pretty much everything these days that's low power consumer electronics, etc........I'd say then it's much safer to order more expensive brand name caps, with the true proper spec's for the job.

I have some cheap caps of most types, a few nice caps from salvage that are still good, and hardly any caps that I could vouch for tho, if they wanted top join the Mafia I guess.

Do I regret buying cheap caps from ebay over the last few years ? I only remember blowing up 2, and that was by mistake. But that was just VERY VERY VERY low power stuff, like a few Watts. So no, I guess not for breadborading and the odd repair.

I don't have any electrolytic cap's I trust for any length of time in a computer SMPS, or TV/monitor SMPS. And hardly have any caps I'd call accurate for timing.

If the cap is in a high power, and/or /high temp/frequency setting, and it's an electrolytic cap, then get something brand name for long term confidence.

I've repaired a lot of CRT monitors for vintage arcade games and for many years there have been cap kits on the market for specific monitor models so it's very common to find monitors that have had the whole set replaced with what tend to be the cheapest parts available at the time. I've seen lots of cases where the original capacitors were replaced when the monitor was 25-30 years old and then after just a few years of home use there were bad capacitors again. That's what made me realize that a lot of the cheap electrolytic capacitors are just junk.

I agree about the ceramic though, I have not experienced the high failure rate with cheap ceramic capacitors, and the cheap film capacitors seem ok too.

[wraper]

For replacement parts, especially on stuff that you're going to repair for other people I don't like to screw around with iffy components.
I bought this electrolytic assortment from Digi-Key with caps from Panasonic.
https://www.digikey.com/en/products/detail/panasonic-electronic-components/M2-KIT/4383781
(Funny, they don't seem to have a price on it now.)
When I use some of the caps I just put some more on my next order.

Such kit is pretty much pointless for repairs. You'll be quite lucky if there will be something, and especially everything you need for actual repair. Not to say it does not even contain low ESR capacitors which are a must for repair of TV, monitors and similar.

[Zero999]

To put this into perspective, try searching TME, RS Components, or Farnell for a 100µF 400V capacitor. Look at how many options there are, in terms of physical dimensions, for a start. There have been plenty of occasions when I've had the correct value and voltage in stock, but it's the wrong shape to fit. Sometimes I can just bodge it on with wires, but quite often it just won't fit in the case. My capacitor kit is primarily used for prototyping, hardly ever for repairs.

[Siwastaja]

Kit wouldn't make any sense; you would need say some 20-30 different parts about 5-10 pcs each to be able to actually repair something. If decent quality, such kit would cost way more (at least some $50 for just the parts) than paying for Digikey/Mouser postage once or twice, getting just what you need, and 99% of the parts in the kit would slowly age without ever being use.

Now, a professional repair shop theoretically could end up using the parts of such kit, but OTOH, they have no problem just ordering the correct parts when they need them, it's their core business.

[Renate]

You'll be quite lucky if there will be something, and especially everything you need for actual repair.

Gosh! I guess that I'm lucky then.

Not to say it does not even contain low ESR capacitors which are a must for repair of TV, monitors and similar.

Yes, low ESR is desirable, but since when have we raised this to a fetish?
Usually I just replace a bad cap with one of these:

[Siwastaja]

Some circuits, for example those filtering rectified 50/60Hz AC, are just fine with high ESR.

But ripple current rating is highly important for switch mode converters. With wrong type of capacitor, the capacitor can self-destruct in matter of minutes due to too much internal heating. Low-ESR types are used in such circuits. Even then, it's not enough to see a "low ESR" claim, the actual ratings need to be understood (ripple current rating in mA @ 100kHz).

This is engineering, nothing to do with "fetish". You can't eyeball it or base it on feelings, it's simple math and real physics.

Obviously if it wasn't important at all, how did the original capacitor die?

[james_s]

Some circuits, for example those filtering rectified 50/60Hz AC, are just fine with high ESR.

But ripple current rating is highly important for switch mode converters. With wrong type of capacitor, the capacitor can self-destruct in matter of minutes due to too much internal heating. Low-ESR types are used in such circuits. Even then, it's not enough to see a "low ESR" claim, the actual ratings need to be understood (ripple current rating in mA @ 100kHz).

This is engineering, nothing to do with "fetish". You can't eyeball it or base it on feelings, it's simple math and real physics.

Obviously if it wasn't important at all, how did the original capacitor die?

Any number of things could have caused the original capacitor to die, but yes you are correct. I think his point is some people don't really understand any of this, they just blindly go with low ESR capacitors because they're "better". In a lot of cases it doesn't really matter, modern general purpose capacitors can be quite good compared to what was considered low ESR or high ripple current 25 years ago but once in a while you come across something where it the choice of part is more critical.

[wraper]

Any number of things could have caused the original capacitor to die, but yes you are correct. I think his point is some people don't really understand any of this, they just blindly go with low ESR capacitors because they're "better". In a lot of cases it doesn't really matter, modern general purpose capacitors can be quite good compared to what was considered low ESR or high ripple current 25 years ago but once in a while you come across something where it the choice of part is more critical.

Nonsense. Modern general purpose caps generally are nowhere close to entry level LOW ESR and about the same as 25 years ago. Not to say that digikey kit has 85^oC capacitors. Modern SMPS almost exclusively contain LOW ESR capacitors, except high voltage cap on the primary side. If LOW ESR capacitor failed, replacing it with general purpose stuff is insane. In most cases it will last a few months or less.

[CatalinaWOW]

Lots of good advice here.  But some needs to be taken with a bit of skepticism.

Often these failed parts are in a bit of gear that was used full time (sometimes even 24 hours a day) for decades.  And often stacked with a bunch of other heat generating gear.  Gear that old is unlikely to ever see that type of service again.  If it will, fine, replace the electrolytics with the best you can buy.  If it is a SMPS good (but perhaps not top of line caps) are warranted.  But in other cases the remaining actual service life of the device may well be served with much more pedestrian parts.

Case in point.  The monitor I am using now failed from bad electrolytic caps.  I re-capped with a modestly priced set from Ebay (not the lowest price, but far from Rubycon prices).  It has been working fine for ten years so far, which matches the performance of the original installation. 

[james_s]

Nonsense. Modern general purpose caps generally are nowhere close to entry level LOW ESR and about the same as 25 years ago. Not to say that digikey kit was for 85^oC capacitors. Modern SMPS almost exclusively contain LOW ESR capacitors, except high voltage cap on the primary side. If LOW ESR capacitor failed, replacing it with general purpose stuff is insane. In most cases it will last a few months or less.

I never said anything about using a cheap general purpose capacitor in a modern SMPS, and I never said it was ok to replace a low ESR capacitor with something that has a much higher ESR. There are lots of SMPS from the 80s and even 90s that do not have special low ESR capacitors in them, they just use larger capacitors and usually several of them in parallel on the output rails. Ultimately it is usually the ripple current that matters, not the ESR. Most "low ESR" capacitors have a high ripple current rating but not all.

[wraper]

I never said anything about using a cheap general purpose capacitor in a modern SMPS, and I never said it was ok to replace a low ESR capacitor with something that has a much higher ESR. There are lots of SMPS from the 80s and even 90s that do not have special low ESR capacitors in them, they just use larger capacitors and usually several of them in parallel on the output rails. Ultimately it is usually the ripple current that matters, not the ESR. Most "low ESR" capacitors have a high ripple current rating but not all.

I doubt very much that such equipment is repaired that often, unless someone is collector of physically and morally obsolete stuff. Most of it has gone to landfills a long time ago. Most of the stuff with SMPS made in last 15-20 years has LOW ESR capacitors.

[TimNJ]

Please someone explain to me how ripple current rating is not related to ESR? Endurance rating is given as: Number of hours before capacitor degrades (to some point), when operated at rated ripple current and max rated temperature.

ESR drives self heating of the capacitor can by (Iripple)^2 * ESR. Can size, specifically surface area exposed determines how well the can dissipates heat generated by the capacitor “core”. I guess in engineering terms, can size influences the thermal resistance from core to ambient.

From my understanding, the ripple current rating is derived from the ESR and size of the capacitor.

Now, there exists low ESR caps with varying endurance ratings, which is a little confusing at first glance. But, basically, as I understand it, different electrolyte formulations have different tolerance to heat. I think some of the low ESR caps with only 1,000-2,000 hour lifes are “water-based” or something like that, which allows low ESR, but dry up more quickly compared to a cap with same ESR but 10,000 life.

[CatalinaWOW]

Please someone explain to me how ripple current rating is not related to ESR? Endurance rating is given as: Number of hours before capacitor degrades (to some point), when operated at rated ripple current and max rated temperature.

ESR drives self heating of the capacitor can by (Iripple)^2 * ESR. Can size, specifically surface area exposed determines how well the can dissipates heat generated by the capacitor “core”. I guess in engineering terms, can size influences the thermal resistance from core to ambient.

From my understanding, the ripple current rating is derived from the ESR and size of the capacitor.

Now, there exists low ESR caps with varying endurance ratings, which is a little confusing at first glance. But, basically, as I understand it, different electrolyte formulations have different tolerance to heat. I think some of the low ESR caps with only 1,000-2,000 hour lifes are “water-based” or something like that, which allows low ESR, but dry up more quickly compared to a cap with same ESR but 10,000 life.

They are related.  But it isn't a one-to-one relationship.  In addition to the chemistry differences you mention there are physical choices on the dimensions of the conductors inside the case, their arrangement, seals, amount of excess electrolyte at manufacturer and more.  The manufacturer should and usually does know all this and the choices they made in design and manufacture.  This, and how conservatively they wish to make their claims based on market strategy determines what goes on the data sheet.

[rsjsouza]

I never said anything about using a cheap general purpose capacitor in a modern SMPS, and I never said it was ok to replace a low ESR capacitor with something that has a much higher ESR. There are lots of SMPS from the 80s and even 90s that do not have special low ESR capacitors in them, they just use larger capacitors and usually several of them in parallel on the output rails. Ultimately it is usually the ripple current that matters, not the ESR. Most "low ESR" capacitors have a high ripple current rating but not all.

I doubt very much that such equipment is repaired that often, unless someone is collector of physically and morally obsolete stuff. Most of it has gone to landfills a long time ago. Most of the stuff with SMPS made in last 15-20 years has LOW ESR capacitors.

You would be very surprised to see the vast amount of people that performs repairs in Amigas, Apples, early PCs, old monitors... All of them with ancient SMPS circuits.

[wraper]

You would be very surprised to see the vast amount of people that performs repairs in Amigas, Apples, early PCs, old monitors...

IMHO you overestimate that grossly, confirmation bias. A few enthusiasts who do that relatively often and some very rare random jobs done by others. In comparison to repairs of more modern equipment, it's just a miniscule amount. Unless person states intention to repair exactly those, it's not even worth consideration.

All of them with ancient SMPS circuits.

I did not say they don't have it.

[helius]

It may depend on where you are, but in the relatively affluent US, repair shops for "more modern equipment" have largely gone out of business. The TV/radio repair shop used to be a familiar sight in the mall or the phone directory, but they are very rare now, so most "modern equipment" is being scrapped rather than repaired.

[Siwastaja]

General purpose elcap series do not typically report ripple current rating at 100kHz, only at 120Hz, but if it is rated/simulated/measured @ 100kHz, it tends to be roughly half of the ripple current rating of a similar sized "low ESR" type.

Given that even the low ESR types tend to be quite stressed in SMPS circuits, replacing with one with only half the ripple rating, the result seldom lasts very long.

But of course, no one says an SMPS design must use a type labelled as "low ESR". One with high enough ripple current rating is obviously enough. Similarly, you don't actually shop for a "low noise opamp" or a "high performance DSP", instead you buy one with the numbers you need.

Ripple current rating and ESR of course go hand-in-hand. Given same I_ripple, lower ESR means less internal heating. Other parameters affecting this is the surface area of the capacitor, internal heat conduction from the core of the capacitor to the outer sleeve, and the details of the electrolyte, how quickly a certain temperature degrades it.

Ripple current rating exists solely so that you don't need to understand the internals of the capacitor, model the heating and then try to find out how much ripple current you can put through it. Just look at the table and you are given the number you need for your design, directly. This is handy.

Low ESR types are easily available in 5000-8000 hour ratings @ 105degC, this specification is actually the ambient temperature (around the capacitor, not around the complete device) and at rated ripple current. It follows, it's fairly easy to derate the operating temperature and/or ripple current so that 5000 hours becomes 50000 hours. The practice of derating is widely understood and known and supported by documents from the capacitor manufacturers, so a 5000 hour rating does not mean the capacitor is actually supposed to only last for 5000 hours, not at all. Like always in engineering, you need to know what the numbers mean.

"Repair people" tend to have strong biases. In reality, repair is a tiny tiny business compared to manufacturing new units to replace broken stuff. If you and your friends like to repair your broken devices, you are missing the 100000 average Joes who just ditch them.

[rsjsouza]

You would be very surprised to see the vast amount of people that performs repairs in Amigas, Apples, early PCs, old monitors...

IMHO you overestimate that grossly, confirmation bias. A few enthusiasts who do that relatively often and some very rare random jobs done by others. In comparison to repairs of more modern equipment, it's just a miniscule amount. Unless person states intention to repair exactly those, it's not even worth consideration.

I never compared with modern equipment, so this is a moot point.
In my experience, the amount of people that are interested in maintaining these equipments is quite large, even if they outsource the repair work (which is still imposes a demand on regular capacitors, which is the subject of this thread). You don't know the entire world.

All of them with ancient SMPS circuits.

I did not say they don't have it.

Neither I suggested the opposite.

[wizard69]

What I don't see stressed enough here is the fact that you will likely never have what you need in the kit.   On top of that a good portion of the kit will just age and eventually become a problem because of that.   Al the technical discussions in this thread are completely true but the number one point for me is the waste of money that kits of Electrolytic Capacitors are.   Even worse is the temptation to put that kit to work, because you spent all that money, on bodged installations that likely will not be reliable.   Bad voodoo very bad VOODOO!

Now that is in the context of a repair lab.   A designer might have use for a kit but then he will have a good idea what he expects to be designing and how a kit might fit.   Even here it often makes more sense to simply buy the right parts to fit the design.

The story might be a bit different for other capacitor technologies.   Ceramic capacitors will likely last for ever on the shelf, but even here I can't see buying a kit in the hopes that it would have the right chip when you need it for repair work.   Again a different story on a hobby or design bench where the projects can be varied.

Beyond all of that if you really want to make up an on going inventory; simply order more parts than the last repair required.   At least then you know you have a use for the parts.   As a business you would want to keep your inventory concentrated on stuff that will move out the door fairly quickly.   In other words keep stock focused on stuff that drives the business and avoid tying up cash in stock that will not generate revenue quickly.   Sorta JIT thinking, though I don't see JIT as ideal for a repair shop, it isn't like you will know what you are working on from one hour to the next.

So baring all of the well reasoned points earlier in this thread I just see kits like this as a bad business move.   There are far better things to tie your cash flow up in.

[Jwillis]

Pretty much every thing others have said . But the biggest annoyance with electrolytic cap kits for me is getting a bunch in the kit with voltages you never use .This of course depends on what type of electronics that interests you. Better to get good E caps as you need them . You really don't save much getting cheap ones from unknown sources . I've not had any problems with E caps from Ebay and they have tested fine but I can't really say how they will perform as they age .
Just my opinion.

[wizard69]

"Repair people" tend to have strong biases. In reality, repair is a tiny tiny business compared to manufacturing new units to replace broken stuff. If you and your friends like to repair your broken devices, you are missing the 100000 average Joes who just ditch them.

Sadly much of what you said is correct, however it would do us a lot of good to try to change the mind set with respect to repair.   Part of the throw it away mentality has come from the rapid advancements in compute technology which made old computers obsolete pretty quickly.    That rapid obsolescence is no longer the case, a smart computer purchase can last for years now for many users.   Repairability just means a device will last even longer.

On the flip side I worked with a guy for years that had a computer repair business on the side.   Most of his repairs involved reinstalling Windows and fixing other OS related issues.   Actual repairs would be at the sub assembly level.   However when products have known failure points that are easy to address one can save a lot of money over the board swaps that large computer companies do.

In the end capacitor replacements can be a win win situation when it comes to repair.   It is something that usually is easy to diagnose, fairly cheap to repair, and makes the customer happy when charged far less than a board swap when going through the manufactures support.   In a nut shell I really see repair as something that needs to be focused on more as it is good for the economy, good for the environment and keeps some cash out of the hands of the largest corporations in the country.   So yeah lets do cap replacements right so that repair can maintain a good public image.

[Zero999]

What I don't see stressed enough here is the fact that you will likely never have what you need in the kit.

I agree. That has been my main arguement: it's highly unlikely any of the capacitors will fit, so you'll be tempted to bodge. Often I will use a capacitor from my kit to piggy back on to a faulty capacitor, to prove the circuit works, when it's replaced, then I'll by a new one, in the correct case size for the propper repair.

On top of that a good portion of the kit will just age and eventually become a problem because of that.

I have not found this to be the case. I have some very old capacitors in my capacitor assortment which I salvaged from some old toys/radios/TVs, when I was a child (yes I know it's not advisable to use second hand capacitors, but I didn't know any better back then and it was one of the few sources I had) and guess what? Many of them are perfectly fine, even though they're over 30 years old. For prototyping it's a good idea to keep a stock of the most commonly used electrolytic capacitor values, at least 10µF, 100µF and 1000µF, which are handy for decoupling and output filtering on power supplies, but I try to stock electrolytic capacitors in the E3 values, i.e. 10, 22 and 47, from 2.2µF to 10 000µF, for general prototyping purposes.

[Renate]

I've been surprised at the level of disapproval here.
Yeah, it's all fine to say that a part should only be replaced by its exact equivalent (after checking with the manufacturer to make sure there hasn't been an engineering change order).
Maybe some of you should go look in at the scrap recycling center and see how many TVs and monitors there are there.
Would a quick replacement cap that gets another 5 years of life out of some of them be such a bad thing?

Ok, if I were fixing a spectrum analyzer I would be more picky about a replacement part than a $99 Walmart TV.

And yeah, for some of us, the hassle of ordering a single part would be a hassle.

[wraper]

Would a quick replacement cap that gets another 5 years of life out of some of them be such a bad thing?

When replacing LOW ESR cap with some random general purpose cap, likely will get you less than a few months of life. And as I already said, basically any LCD TV or monitor will have low ESR capacitors. As of laptops, you're not likely to find any electrolytic caps in anything made in last 15 years. And even in a rare case when there is one or two, unlikely to be a reason of failure.

[Dacke]

Ok, if I were fixing a spectrum analyzer I would be more picky about a replacement part than a $99 Walmart TV.

It's not just TV's and monitors,  switch mode supplies are in literally everything now.  Low ESR caps are par for the course.  I do a lot of SMPS repair,  and it's rare that I would use gen purpose caps for anything but audio circuits,  testing,  and prototyping.  As wraper has stated,  you will not get "a few years" of life from a gen purpose cap being used in these power supplies,  you would get weeks or a few months at best.  I speak from experience and it's annoying as hell to go back into something you've recapped only to have to do it again in 2-3 months,  even if it's just something I'm repairing for myself.

[Siwastaja]

If the original, designed-in-by-a-professional part failed in a few years of use, don't expect 5 years lifetime out of random parts. You need to do better, not worse.

I don't repair that much, mostly I design, but personally, some 15 years ago I "repaired" a PC PSU with a general purpose "I have this on my self" elcap which had the correct voltage, temperature and capacitance ratings, and it lasted in the circuit for about 1 month until I had to order proper low-ESR caps so this isn't just theoretical talk.

[TimNJ]

Not to  but another two reasons:

1. ESR of output caps often plays a large role in ripple suppression. GP caps might have 5...10x ESR of a low ESR cap. As power supplies are commodity items, they are built to a price, and usually don’t have a ton of margin with respect to their output ripple specifications. So, put in a GP cap and you might be very far out of the specification. Will it cause problems? Maybe. Maybe not. Might look fine at first but start doing weird things once you really start to use it.

2. Depending on the topology the ESR zero of the output capacitors play a role in the feedback loop stability. Change it too much and you risk instability. Granted, most current-mode control (and similar) power supplies (these days) are pretty robust and easy to keep stable, but it is still not a good practice.

[wizard69]

I see this "lets buy a kit of parts" mentality in all sorts of technical pursuits.   One example I just ran into  is a guy ordering O-ring kits for pneumatic repair and only using three sizes out of the kits.    I say kits because he had 5 kits of assorted O-rings with every kit missing the same three sizes of O-rings.   The fact is he could have ordered several hundred of those 3 O-Rings and not spent the money that one kit cost him for 10 of the needed O-rings.

So the issue of kits is not something that is a problem just in the electronics repair industry.   It is wasteful in just about any repair activity.    Note: that I'm saying this in the context of a repair activity or business.   It is a different story for an R&D operation but even there you need to be careful.   

Beyond that if your income stream involves repairing common item it may pay to buy a know failure item in bulk.   Often with a bulk purchase you are still paying far less than the inflated prices seen in kits.   

Maybe I'm triggered by this subject but I don't see the point in kits, often they are ordered by lazy people or people that don't know any better.   You are almost always better off ordering the correct part for item to be replaced.

[CatalinaWOW]

I see this "lets buy a kit of parts" mentality in all sorts of technical pursuits.   One example I just ran into  is a guy ordering O-ring kits for pneumatic repair and only using three sizes out of the kits.    I say kits because he had 5 kits of assorted O-rings with every kit missing the same three sizes of O-rings.   The fact is he could have ordered several hundred of those 3 O-Rings and not spent the money that one kit cost him for 10 of the needed O-rings.

So the issue of kits is not something that is a problem just in the electronics repair industry.   It is wasteful in just about any repair activity.    Note: that I'm saying this in the context of a repair activity or business.   It is a different story for an R&D operation but even there you need to be careful.   

Beyond that if your income stream involves repairing common item it may pay to buy a know failure item in bulk.   Often with a bulk purchase you are still paying far less than the inflated prices seen in kits.   

Maybe I'm triggered by this subject but I don't see the point in kits, often they are ordered by lazy people or people that don't know any better.   You are almost always better off ordering the correct part for item to be replaced.

If repair is your business, and if time to repair is not an issue I totally agree.  For those at remote locations, or homeowners, or hobbyists the answer is no where near as compelling. 

The O-ring kits are a great example.  I have a couple, in different materials for different purposes.  And as you correctly state I have over a decade or two of ownership used less than 5% of the kit contents.  By the time I die I am sure that more than 85% of the O-rings will not have been used.  But the gasoline saved by not having to drive down to a local store (in my case about 40 kM away) those few times a usable O-ring was in the kit paid the entire price of the kits.  And the time saved not having to make that drive or order over the internet when local sources don't stock the correct part (more often than not) has its own value.

For professionals where time to repair really matters (just about anyone supporting an assembly line, a mine or something like that) it makes sense to ignore the general purpose kits and figure out what parts are used in the machinery you are supporting.  And then stock up a kit of those parts, being sure to re-order as they are used.  Your associate buying the pneumatic kits would have done well to take that approach, assuming that his salary in doing the research didn't cost as much as the kits.  Probably the best economic answer is to buy a few kits to keep things up and running.  Those kits will then teach you what you use and you can then order those parts in quantity.

[perieanuo]

hi,
big no no
I just keep revolving capacitor stock from serious suppliers, tme farnell & company
if you have ONE returning repaired TV in warranty cause one new&non-conform capacitor, the time and gas money you loose undo all the profit and pleasure gained

[Zero999]

I see this "lets buy a kit of parts" mentality in all sorts of technical pursuits.   One example I just ran into  is a guy ordering O-ring kits for pneumatic repair and only using three sizes out of the kits.    I say kits because he had 5 kits of assorted O-rings with every kit missing the same three sizes of O-rings.   The fact is he could have ordered several hundred of those 3 O-Rings and not spent the money that one kit cost him for 10 of the needed O-rings.

That doesn't mean kits are always bad, just that that person was a fool for buying a complete new kit, rather than just ordering the O-rings to replenish it. Take my kit of E3 capacitors for example. When I run out of a common value such as 10µF, I don't buy an entire brand new kit, just a pack of 10µF capacitors! 

[wizard69]

I see this "lets buy a kit of parts" mentality in all sorts of technical pursuits.   One example I just ran into  is a guy ordering O-ring kits for pneumatic repair and only using three sizes out of the kits.    I say kits because he had 5 kits of assorted O-rings with every kit missing the same three sizes of O-rings.   The fact is he could have ordered several hundred of those 3 O-Rings and not spent the money that one kit cost him for 10 of the needed O-rings.

That doesn't mean kits are always bad, just that that person was a fool for buying a complete new kit, rather than just ordering the O-rings to replenish it. Take my kit of E3 capacitors for example. When I run out of a common value such as 10µF, I don't buy an entire brand new kit, just a pack of 10µF capacitors! 

There are many factors to consider here.   In this case I was addressing what appeared to be somebody getting started in the repair business apparently on a shoe string.   Buying a whole kit of electrolytics just makes no sense to me in that context.   It is a big expense for stuff you may never use.   Repair is a business where you have to carefully watch where your money goes to have any prospect of staying in business for very long.

As for other activities that is something that each engineer will have to decide for themselves.   A kit of parts might make sense if you are engaged in R&D, product design or the like.   Even there I'd have to suggest that one be careful as you can easily end up with a big out lay of cash better spent elsewhere.   

I'm not against kits nor against the idea of organization in some contexts I just think kits of electrolytics can be wasteful starting out in a repair business.   That comes down to the idea that your chances of having a good fit for parts replacement is slim.   If you are going to have inventory, it is best to build that inventory around the stuff that you commonly see on your bench.