Electrolytics.

[Diabolical Artificer]

I'm replacing some caps on an amplifier, I was about to put a 22u cap in, when I noticed the originals are actually 0.22u. I measured it to make sure it really was a .22u. I've ordered some, but notice that on a few, it says available whilst stocks last. So it looks like low value electrolytic's arn't going to be that easy to get in the future. What's the lowest value electrolytic caps you've come across?

Which brings me to my main question. could I have used a non polarised poly cap to replace it, and why are electrolytic's polarised in the first place. I know they explode if reversed, but if a 1000u poly cap was cheaply available, what would happen if I used it in a power supply?

I'm also puzzled why a capacitor has a polarity in the first place, as they block DC and an AC current surely doesn't care which is poss and neg. Odd that I hadn't thought of this before, it's a big hole in my "database".

One last observation regarding replacing electrolytic caps, I noticed 450v rated small value caps like 3.3u for instance, are cheaper than lower rated voltages.

Andy.

[Lucien Nunes]

Electrolytics are mainly used because they give large capacitance at low cost in a smaller package than any other technology. They have few other redeeming features. In many applications you can replace them with film caps but 1000uF would be very large and very expensive by comparison.

Electrolytics are polarised not by way of the actual capacitor function itself, which is as you say polarity insensitive, but due to the electrochemical process that creates and maintains the very thin dielectric film of aluminium oxide dielectric. In order to obtain very high capacitance density, the dielectric has to be very thin. It can't be made as a sheet material so it is electrochemically deposited on the anode foil during formation. It will decay unless maintained by electrolytic action, for which a DC bias is required. Incorrect polarity will dissolve the film instead of maintaining it. The process of reforming electrolytics is using this process to repair holes that have developed in the film due to prolonged lack of DC bias.

Some values of high voltage electrolytic are very cheap due to the vast numbers used in mains-powered SMPSUs, e.g. in low energy lamps.

[ms660]

Originally Posted by Dr Wobble

Which brings me to my main question. could I have used a non polarised poly cap to replace it, and why are electrolytic's polarised in the first place. I know they explode if reversed, but if a 1000u poly cap was cheaply available, what would happen if I used it in a power supply?

You can replace electrolytic with non electrolytic, size permitting.

You can also replace none electrolytic with electrolytic but that would depend on the circuit impedance involved due to to the inherent leakage current in an electrolytic.

So far as I know.

Lawrence.

[ukcol]

Hi Andy

Electrolytic capacitors allow a large valve of capacitance in a very small space. Capacitance is proportional to the area of the plates and inversely proportional to the distance between the plates. The material of the dielectric also is a factor. In electrolytic capacitors the distance between the plates can made very small giving the high capacitance. They need to be polarised to keep the dielectric in place and if reverse polarised the leakage current rockets creating heat and the risk of a bang.

A 1000uF poly capacitor, because of its relatively thick dielectric, would need a very large plate area to get the required capacitance. This equates to a physically very large component.

[dave cox]

Electrolytic caps are smaller (=cheaper) for a given value. For example, I have a bag of 1uF / 450V electrolytics these are tiny compared to a film cap with the same spec! I see odd ones below 1uF but its not so common. Given the dominance of low voltage circuits many low uF requirements would now use high k ceramic types which will be much more reliable in the long run.

dc

[HamishBoxer]

.47uF Electrolitic I have certainly and possibly .22uF.

[G8HQP Dave]

If all you want is capacitance then any capacitor of the right value will do. However, real life is not as simple as a simulation. Electrolytics have significant leakage, so can't always be used in place of a film cap. On the other hand, electrolytics have a little series resistance which may be used to damp an unwanted resonance so can't always be replaced by a film cap. For a given value, a film cap may be large: too large to fit, or sufficiently large to have much greater stray capacitance to other nearby circuit nodes.

Circuits generally use the components they do because someone decided that these were the best components for the job at the time the circuit was designed. Sometimes this choice can change with time, but something I often say on an audio website (where naive 'upgrading' is common) is "if you want to genuinely improve a circuit you first have to understand it even better than the original designer".

[Sideband]

Originally Posted by Lucien Nunes

the dielectric has to be very thin.

....and this is a problem with very low value electrolytics. They tend to fail much more quickly than the higher value type. If you measure the ESR of low value electrolytics, it's a good bit higher than the high value types. You might be better off trying to find some low value Tantalums

[Okto1984]

I can't comment on their suitability in all cases, but I have been experimenting with film caps to replace electrolytics. I was given a bag of miniature (5mm pitch) film capacitors of various values up to 4.7uf, and I have been using them where needed to replace the electrolytics in various devices. Currently there's about 20 of them in a TV I recapped, some in a couple of tape players, a record player, and several vintage computers/consoles. Everything has worked nicely so far, but maybe one day I'll find a situation they don't work in.

[Diabolical Artificer]

Thanks everyone, I understand a lot better now. talking of high value non -electro caps, I remembered I have a big Castanet, I think, 100u approx cap in the shed, measures approx 2" x 2" x 3/8.

The 0.22u caps in question do have a high ESR, which is why I thought about replacing them, but did wonder if they were ok, but couldn't find a datasheet at the time with a value to compare to. The cap in question is an orange Nichicon; all the other orange Nichicon caps are fine.

This subject of replacing caps is relavent to something else I have on the bench, a bit of 70's test gear full of tantalum's. With the price tantalum's are, not to mention the dubious provinance of some tantalum sources, I'd be tempted to replace deffective tant with something else.

Looking online I see Niobium oxide caps are cheap, but are only available up to about 10v, oh and theyre SMD as are film caps but they're more expensive than tantalum's.

I would'nt replace a capacitor with a different type unless I understood the circuit it's in say for coupling/DC blocking caps on the input of an amplifier stage. My knowledge of oscillator's and HF/RF circuit's is skant, so I'll have to replace the tantalum's in question with the same.

With the use of graphine in a lot more applications, I wonder if it will get used in capacitor manufacture as a dialetic? AFAIK, it's conductive, so maybe not.

Andy.

[turretslug]

Every now and then, another variant on the "carbon-based wonder material" theme appears, everyone gets a little bit excited- and then it all goes quiet again. Either, these nine-day wonders are rather less useful and rather more costly and difficult to implement in widespread real-world applications than their evangelists promise- or someone like the military or big business takes the fancy to collar it. Maybe I'm just getting cynical...

With compact, inexpensive non-polarised film capacitors now widely available in the low uF range, it struck me that one useful application would be that Achilles heel of elderly FM receivers, the electrolytic discriminator capacitor. The one that goes leaky/low value/causes distortion. At least a plastic film type would avoid the positive-to-chassis confusion that ensues in many sets.

[Oldelectronics]

fascinating post by doctor wobble particularly Lucian's comment it certainly improved my knowledge thank you.

Kev

[Diabolical Artificer]

I've learn't a lot since joining the forum Kev, but realised I had no idea why an electrolytic was polarised. Thanks to Lucian et al for enlightening us.

Suppose your right TS, electrolytics have been around since they stop putting caps in jars, can't see them going anywhere soon. You'd have thought someone would have come up with a type of plastic thin enough to replace the chemical dielectric though. Material science has come on a bit since the 60/70's. Remember the old stiff plastic looking shirts and dresses people used to wear?

Am sure I read of a new dielectric relatively recently, but for the life of me, can't remember the detail's.
I do remember the range was limited to around the low 10's of microfarad's. They looked like dipped tantalums.

andy.

[dave cox]

You are probably thinking of the high K ceramic multi-layer (non-polarized). These have some 'wild' characteristics with C changing greatly with both temperature / applied voltage and they can make quite a racket when they are being hammered by a SMPS !

There are also niobium based caps (similar to tantalum I guess ?).

dc