advice or parts source

That's an odd one for sure! I'm not used to seeing 'lytics with so low a capacity.

Lower distortion, longer lifetime. Unless you feel that those are bad things, go right ahead and stick in the 0.22 filmcap.

Here, also, I'd suggest using a plastic-film capacitor.

These are almost certainly filtering/smoothing caps. As you can see, the tolerance is extremely loose. You can safely move up to the next working-voltage rating and to the next-higher standard capacity, with no worries at all. Modern caps are generally more compact than the older ones and so you shouldn't have a problem finding a fit.

If possible, source 105-degree-C caps rather than 85-degree-C caps. They may have a longer lifetime.

Do you need radial-lead caps, or the nowadays-less-common axial-lead caps? If radial, the Panasonic NHG or FC series caps from Digi-Key ought to work out fine for you - they're available in the capacities and voltage specs you cite above, with tighter tolerances, and you could move up to the next capacity in the series to improve the filtering a bit if you want.

If you need axial-lead caps, Digi-Key has the BC Components ASM 021 and AHT 118 and AML 138 series - the latter are long-life parts which are available in exactly the values you want.

--
Dave Platt                                    AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

perplexed

polypropylene

comparable

So, what you appear to be saying is that for a given capacitance and rated voltage, film caps are a better choice if they are available. The electrolytics are used (in general) where higher capacitance is needed. Why, then, is my board liberally sprinkled with 1uF, 2.2uF, and 3.3uF electrolytics if film caps last longer and create less distortion? I'm talking about the FM signal path caps, decoupling caps, etc. Was it not possible in 1981 to get a 1uF 50V non-electrolytic capacitor?

What would be the drawback(s) to replacing the electrolytics with their

1000-5000h life with film caps of identical rating with a 200,000h MTBF?

Thanks a lot for your insight.

Dave

Sorry, I should have been more clear. I mean what would be the ramifications of replacing the 0.15uF electrolytic with a 0.1uF or 0.22uF ELECTROLYTIC?

Ah, cost. A 3.3uF film cap is >$2 whereas the identical electrolytic is about $0.30.

For signal-coupling applications, I believe that this is an accurate statement.

I $uppo$e that it ha$ to do mo$tly with the relative co$t of the different type$ of capacitor$.

Electrolytics are cheap, and physically small for their capacity. Filmcaps are expensive (relatively speaking) and are physically larger per capacity.

As a crude example: using Digi-Key single-quantity prices, small electrolytics suitable for signal coupling purposes might run in the $.20 range and would be quite small (under 1/2" long and maybe 1/4" in diameter?). Metallized-polyester caps of similar capacity cost about a dollar and are at least twice as large in each dimension. The compact "stacked metallized film" capacitors can rival electrolytics in physical size, but a 1 uF 50-volt part will cost you $3.63.

Certainly. However, they were physically a good deal larger than electrolytics of similar capacity, and quite a bit more expensive. The small stacked-film caps available today compare favorably in size to the small 'lytics, but they're probably ten times the price.

The only technical drawback I can see, would be that you'd need to make sure you're buying filmcaps which can fit into the physical space available.

The filmcaps are significant more expensive than 'lytics.

Note that all of the above applies to signal-path caps. I would *not* recommend replacing electrolytics with filmcaps in power-supply bypassing applications. Power supply traces which are bypassed only using low-ESR, high-Q caps (e.g. stacked film) can suffer from some nasty resonance effects - the inductance of the traces interacts with the high-Q capacitance to form a tank circuit - and this can result in oscillation, ineffective bypassing at certain frequencies, and other nasty effects.

--
Dave Platt                                    AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Little to none. Depending on the source and load impedances, it's possible that going down from 0.15 to 0.10 _might_ result in a premature rolloff of the low frequencies. I doubt it, but it's possible. Going _up_ to a 0.22 should ensure that this does not happen.

Given the loose tolerances of most 'lytics, it's unlikely to matter, but I'd recommend usign the 0.22 uF anyhow.

--
Dave Platt                                    AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Having also read your subsequent posts in this thread:

At

, you can find some nice metallized polyester caps by AVX, such as the 0.15 uF 63V 10% for $0.08, qty 1 (or 5% for $0.18). They are SMALL, too: 2.5mm D x 7.5mm W x 6.5mm H, with

0.2-inch lead spacing. They go up to 2.2uF, although the size at 2.2uF rises to 6mm D x 7.5mm W x 12mm H.

The Mouser.com part number for the 0.15 uF 10% one is 581-BQ014D0154K. The 5% version's part number is the same, except for a J instead of K, at the end.

The 0.47uF 63V 5% and 10% metallized polyester AVX caps in the same series have Mouser.com part numbers 581-BQ074D047J and ...K, and go for $0.25 and $0.23, qty 1.

For the 0.15uF cap, I would tend to prefer the WIMA MKP2-series metallized polypropylene, especially if small size is important: The Mouser.com part number 505-MKP20.15/100/5 is rated 100vdc/63vac, 5%, and sells for $0.87 for qty 1. It's SMALL, too, at 7.2mm D x 7.2mm W x

13mm H, with 5mm lead spacing.

If you can tolerate a slightly-larger size, you can get some metallized polypropylene caps with "< 0.1% dissipation factor (DF)": Mouser.com part number 1429-1474 is 0.47uF 100v 10%, for %0.70 qty 1, and measures

10mm D x 18.5mm W x 15.5mm H. And there's also the 1429-2475, 4.7uF 250v 10% for $3.49 qty 1, measuring 37mm W x 18mm D x 27mm H. And once you get up to sizes that large, there are quite a few alternatives available, in film caps.

There are film caps in the larger capacitance values, at Mouser.com, too, although they tend to get a bit pricey, in most cases. For example, check out the ones in the AVX "FFB" series (which seem to be intended for DC filtering).

Note that you may have to solder wires onto the leads, if you are replacing physically-larger caps with some of the very small ones that I mentioned above.

Regarding whether you can always replace electrolytics with similarly-rated film caps (and then get better performance and reliability), the answer might be "usually". But note that, sometimes, maybe especially in power-supply bypass/decoupling applications near ICs (or, e.g., similarly, on a per-board basis), the designer might have relied on the higher ESR (equivalent series resistance) of an el-cheapo electrolytic (or a tantalum) cap to be a necessary part of the circuit (maybe, e.g., for damping). If I ever had to guess, then I would guess that, for an audio signal path cap, a film cap would almost always be better than an electrolytic (and I would then almost always prefer polypropylene over polyester).

Sorry for the long post. Gotta run. Have fun.

Tom