Polymer recaps

[lepidotos]

I have yet to recap any of my Macs, but I'm interested in doing so -- all of them date to the capacitor plague, and I'd rather not lose any of them to corrosion. According to a guy on Vogons, using solid state caps made his Pentium 4 "noticeably quicker and snappier in everything it did", and I'm curious if that applies to, say, a 7400 as well. If not, it would still be something I'd do for longevity's sake, but I'd like to hear others weigh in.

[robespierre]

I'm afraid that doesn't make any sense from a technical perspective. Large value capacitors are used as local energy reservoirs, referred to by engineers as "bypass capacitors". When digital circuits switch on and off, they draw momentary bursts of current, which would appear as noise on the power supply lines if there was nothing to suppress them. Those bursts of noise if uncontrolled could lead to incorrect operation. The capacitors "absorb" noise because they keep the power supply lines at a constant voltage. When a device draws a burst of current, the current comes out of a nearby bypass capacitor instead of traveling down the entire length of the power supply line. This function does not have any influence on the processing speed, which is entirely controlled by a network of clock generators on each chip and disciplined to a master oscillator.

Polymer electrolytics are several times the cost of wet aluminum electrolytic capacitors and do exactly the same thing in a circuit. They have the advantage that without any liquid electrolyte they cannot leak. They also have a better ESR figure, which can be exploited by designers to pack power-hungry components closer together. This difference has no advantage in repairing of electronics that do not require it.

It reads more like the kind of audiophoolery that believes that "exotic" components (silver solder, cable lifts, braided power cables) improve sound quality.

[GaryN]

All right robespierre! It's great to see someone who not only knows what he's talking about, but can present it coherently as well.

"Audiophoolery" is a constant PITA that's been around as long as there has been recorded audio.
This is simply a variation - we'll call it "electrophoolery".
The perception (deception) that stuff "sounds better" after replacing the usual with the exotic is a combination of wanting the exotic to work and the subconscious resistance to not hearing an improvement after spending an obscene amount of money on it.

Cable lifts……    Aarrgh!

[DieHard]

Gary,

Don't be so cynical, that wire is a steal because it comes with "FREE" anti-rat coating

[GaryN]

You think? My rats ate the damn wire anyway and now they can hear my cat sneaking up on them… 

[lepidotos]

Probably the best answer to my stupid question, thanks!

[mrhappy]

it comes with "FREE" anti-rat coating

WOW... free A.R.C!!

[FBz]

“…it would still be something I'd do for longevity's sake…”

I’ll just leave this here:

https://www.rs-online.com/designspark/aluminum-electrolytic-vs-aluminum-polymer-capacitor-and-how-its-benefits-are-used-properly

[5.25 x 15 = 78.75 years]

Might not make your machine “run faster or jump higher”
but ya might get 30 years out of the polymer capacitors.

[GaryN]

…and I'll leave this here…
 The first sentence in Section 5-"Lifetime Consideration" of the quoted paper "Aluminum Electrolytic vs. Aluminum Polymer Capacitor and how its benefits are used properly" IS:

"The lifetime of electrolytic capacitors is very important in industrial applications and other application where a high lifetime is required."

That noted:
* I'm pretty sure a Mac 7400 isn't an industrial anything.
* The chart shows the general superiority of polymer caps at high temps and infers even grater improvements at lower temps
   > The low end of said chart shows "cheap" aluminums running 32 thousand hours at 65°C
   >> If the ambient temp inside your 68k Mac is 65°C or above, you have much bigger issues than capacitor lifetimes
   >>> I'm all for vintage fun but 32,000 hours (or more) staring at a Mac 7400 sounds more like Hell on Earth to me
* More importantly, it's ALL theoretical and assumes no other contributing factors like…I dunno… Quality Control?

Anyway, all of that said…
My main issue was with the sentence:
"According to a guy on Vogons, using solid state caps made his Pentium 4 "noticeably quicker and snappier in everything it did", and I'm curious if that applies to, say, a 7400 as well."

Now, aside from my feeling that there's little or nothing on Earth that can make a Pentium 4 "quicker and snappier" better than 440VAC wall voltage, there are distinct and separate discussions to be had here:

Is it worth it to put 20,000-mile platinum spark plugs in your '53 Plymouth that you only drive to car shows every now and then just because?
What IS the psychology that convinces one to hear, see, feel some kind of "improvement" after spending time, effort and money on things that don't actually provide that?… and
What makes that "reward" so internally convincing that you then install four "high poly-belted-reinforced-organic-rubber-synthetic-combination-alien technology-special edition" tires at $400 each even though the Plymouth never exceeds 40mph and yet you swear they "handle better"?

Anyway, in the end, it's all just a thought exercise and where better for that than here ?

[IIO]

i was always wondering if one could not replace these "liquid" caps with something "better".

but wouldnt other types of capacitors also have slightly different properties? ... charge time(?)

[lepidotos]

…and I'll leave this here…
 The first sentence in Section 5-"Lifetime Consideration" of the quoted paper "Aluminum Electrolytic vs. Aluminum Polymer Capacitor and how its benefits are used properly" IS:

"The lifetime of electrolytic capacitors is very important in industrial applications and other application where a high lifetime is required."

That noted:
* I'm pretty sure a Mac 7400 isn't an industrial anything.
* The chart shows the general superiority of polymer caps at high temps and infers even grater improvements at lower temps
   > The low end of said chart shows "cheap" aluminums running 32 thousand hours at 65°C
   >> If the ambient temp inside your 68k Mac is 65°C or above, you have much bigger issues than capacitor lifetimes
   >>> I'm all for vintage fun but 32,000 hours (or more) staring at a Mac 7400 sounds more like Hell on Earth to me
* More importantly, it's ALL theoretical and assumes no other contributing factors like…I dunno… Quality Control?

True, but it seems like it's doing fine so far and I'd like to be assured that whoever the next owner is won't have to deal with it either, so I'd say it's worth it for a one-time deal essentially that guarantees it won't ever start leaking on me. 30 years isn't amazingly long for electronics to last, I have an SNES that's about that old.

[GaryN]

At the risk of beating this to death for the fourth time…

My only serious issue with this entire "thing" was:

According to a guy on Vogons, using solid state caps made his Pentium 4 "noticeably quicker and snappier in everything it did", and I'm curious if that applies to, say, a 7400 as well."

1) We have an anonymous guy on a gaming site swearing his POS Pentium got "quicker and snappier" with expensive caps.
2) We have you basically believing that and passing it along.

Note that I am NOT trying to pick on you… almost everybody does it - at least to some extent even when they know better.
What I'm trying to get across is that it's a self-generated endorphin-releasing thought process by which we reward ourselves no matter whether the "improvement" is real or imagined because we want it to be true.

This plays right into the hands of every company and every salesperson in every sector. The first and hardest lesson I learned in sales was that you cannot have a conscience . You tell the customer what he wants to hear - always - and sell what you have in stock - knowing that even though the hype you just spread on them was totally untrue, they'll very likely convince themselves of the opposite.

The modern-day master practitioner of this BS is Noel Lee, founder of Monster Cable. He not only bullshitted his way into multi-millionairedom selling overpriced wire, he inspired dozens of others to do so as well. Untold numbers of wannabe audiophiles have wasted unimaginable amounts of money on "high-definition" strings of copper that in some cases, cost as much or more than the speakers they're connected to!

Wire or parts or exotic ADA converters or multi-multi-multi-core processors, it's all the same.

.                                                       .......................... end rant

[IIO]

who needs monster cables? you can just record an impulse-response of those and use that in every DAW.