tantalum cap tolerances

31 views
Skip to first unread message

Douglas Ehrenclou

unread,
Dec 24, 2020, 4:56:01 PM12/24/20
to WB...@googlegroups.com

Hi all, I have a bunch of WBS modules: 460/470/490 modules, 464 HP/LP module 466 compressor modules, and 462 eqs that I’m ready to put to use. After reading posts here, I understand it would be good to replace all the tantalum caps. Been pricing stuff on Mouser/Digikey, and for some caps 5% tolerance caps aren’t in stock and super long lead times. Wondering if 10% tolerance acceptable? Thank you!

warren beck

unread,
Dec 24, 2020, 4:59:02 PM12/24/20
to Douglas Ehrenclou, WB...@googlegroups.com

Good evening and Merry Christmas from Canada!!

Tolerance is not at all important here.  Many people get caught up in 1% resistors, etc… so your 10%, more readily available components are just fine!!

 

Good luck, and enjoy!!!

 

Let us know how it goes.

 

 

 

Regards,

 

Warren

--
You received this message because you are subscribed to the Google Groups "WBSps - Ward-Beck Systems Preservation Society" group.
To unsubscribe from this group and stop receiving emails from it, send an email to WBSps+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/WBSps/CAF6tt%2B9Z-CNjjitF0W6mfnPAU2%3DVvPChmPwHvu8KTrii%2B1qqsg%40mail.gmail.com.

John Maizels

unread,
Dec 24, 2020, 7:55:00 PM12/24/20
to Douglas Ehrenclou, WB...@googlegroups.com
Your biggest problem with tantalum caps might well be working voltage rating rather than tolerance. 

I've noticed that it's getting harder and harder to get tantalums with high voltage ratings (eg: over 35VW), and a few I've replaced recently have most likely failed due to running on the edge of their rating.

Theoretically you could replace (x) uF (y) VW with two caps in series and get (x/2) uF at (2y) VW.  I don't know whether that's smart in practice for either audio or power supply - and presumably  you also get 2x effective internal resistance, which might take the shine off using a tantalum in the first place. 

John

Ken McKim

unread,
Dec 25, 2020, 12:55:27 PM12/25/20
to Douglas Ehrenclou, WB...@googlegroups.com
Hi Douglas,

First, unless you're experiencing a high rate of failure, or you're replacing them for precautionary reasons, I don't see the wisdom in a wholesale replacement of every tantalum in the unit. I favor an "evaluate and replace where necessary" approach. By evaluate, I mean, if possible, do technical measurements (freq. response, square wave response, distortion, noise); then, keep your eye, or your ear, on it. Speaking of ear, if you don't have the equipment to do the aforementioned testing, then identify a module that "appears" to be exemplary and subjectively compare (listen to) other modules to that one.

Tantalum caps gained popularity in the sixties, seventies and eighties because they were physically small for a given value. They are not inherently superior in an audio application. In fact, if you look at the specs on some tants, the dissipation factor is far inferior to some of the more mundane aluminum electrolytics. The other thing that recommends tantalums is a typically low leakage spec. A lot of designers put them in the power filtering part of their designs for that reason: if you've got rail filtering going on using caps at each amplifier stage, times X number of stages, times X number of modules, the leakage currents can add up to something significant. They used low leakage tantalums to reduce the overall load on the power supply. There are low leakage aluminum electrolytics readily available (Mouser / Digikey, etc.) in a wide range of values that will do the job of any tantalum quite nicely.

The tolerance spec is not indicative of the quality of the cap, and most off-the-shelf caps are rated 20%. That's fine. When a designer evaluates the demands of a circuit he has designed, he overbuilds. Unless the cap is going to be used in some kind of timing circuit, the only pressing need is to keep the cap's reactivity well out of the audio bandpass. A 10uF cap, placed at the input of a solid state amp stage, will typically not become reactive till down around 1 hertz. That's way out of the factor of 10 rule of thumb. So, a cap tolerance error factor will only swing that a few tenths of a cycle at most. I think we can tolerate that.

Lastly, if this is about preserving the integrity of preserving the "original design", then I must respectfully drop out of the discussion. My daily job is one of taking practical steps to keep equipment functioning well. Angsting over the original nature of the physical component puts us in the group inhabited by electric guitarists that won't replace the scratchy volume pot on their instrument because then it won't be original (and might knock several hundred off the resale price in the collector's market) - and they can't fantasize that Leo Fender personally soldered that pot in. In the end, I dare you to perceive the difference, in a double blind test, between tantalum and aluminum electrolytics. The only human I know of that can do that is Jimmy Iovine. He claims to be able to hear the difference between silver plated and gold plated XLR connectors. The rest of us just grab a mic cable and get on with it.

As ever,
Ken McKim
Trouble Report





Ken McKim
Owner, Trouble Report

 


On Thu, Dec 24, 2020 at 4:56 PM Douglas Ehrenclou <dehre...@gmail.com> wrote:

Hi all, I have a bunch of WBS modules: 460/470/490 modules, 464 HP/LP module 466 compressor modules, and 462 eqs that I’m ready to put to use. After reading posts here, I understand it would be good to replace all the tantalum caps. Been pricing stuff on Mouser/Digikey, and for some caps 5% tolerance caps aren’t in stock and super long lead times. Wondering if 10% tolerance acceptable? Thank you!

--

Anthony Kuzub

unread,
Dec 28, 2020, 12:45:46 AM12/28/20
to WBSps - Ward-Beck Systems Preservation Society
my nickel (no more pennies)
I once asked why capacitors explode when they are plugged in backward... there is a film and it's coated and when the charge is inverted electrons puncture holes in that film causing shorts exciting the soup inside the can to explode in an exciting fashion. 

I take that logic and apply it to low-frequency audio.  And it explains (to me) why low frequencies destroys Audio caps... near DC condition in constantly changing polarity in AC signals.

My hunch is "tants are expensive" and fixing all takes more time than fixing the one that's broken and buying little precious metals. 

don't put them in backward and don't do more damage than there was when you got there. 

if they are yours have at it and swap them all gives you a "now moment" 20% is okay, better to go up in value than down.

Be well

Anthony 
Reply all
Reply to author
Forward
0 new messages