Tantalum decoupling caps

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Arno

Well-known member
Joined
Oct 24, 2005
Messages
454
Location
Belgium
Any specific reason why some audio company's use tantalum caps for psu decoupling close to opamps?
I know tants have the tendency to go short after some years. Euphonix for example (cs2000) had tantalums all over the console next to opamps.
It's really funny when they go short, console doesn't start. Taking all modules out one by one and than start looking for the leaky bastard.

Right now I'm reading the opa604 datasheet:

Applications with noisy or high impedance power supply lines may require decoupling capacitors close to the device pins. In most cases, a 1μF tantalum capacitor at each power supply pin is adequate

Is there any reason for this ?
(other than bugging the maintenance guy in ten years)

Just wondering...

thanks


EDIT: btw, I'm aware of WHY these caps are there just not sure why they use(d) tantalums.
 
Use X7R type multilayer ceramics, 63V--100V.
Check if everything works as expected, these have lower ESR than tantalums, so your rails are more prone to "oscillating" (ringing, singing)...
 
Tantalums tend to have lower ESR than aluminium which presumably is the reason some datasheets recommend their use. Nowadays aluminium often offers similarly low ESR, and as mentioned in the previous post low ESR is not always what we want.

Samuel
 
sometimes a flaw is a feaure

always hated tants in power supplies, but a few weeks ago i
was fixing a dbx noise reduction unit that had shit the bed.

the voltage regulator shorted and the power rails went from 15 to 23 volts.
most often when tantalums fail they shot, usually very bad.

this time the decoupling tants shorted & took all that power to ground, thus preventing extensive
component failure.  most components in that circuit would have not dealt with that.
aluminium would not have done that, as they tend to just go open when they fail.
 
A sequel of the 70's when tantalums were all the rage because their low ESR and higher value-to-bulk ratio made them appear superior. After a few years, designers (and maintenance guys) proved them unreliable and intolerant of voltage reversal, even minimal. For some reason, some designers and component suppliers still ignore these drawbacks.
 
Can't really imagine why anyone would put a device on the power rails that shorts after a decade.
Unless back then, before they had been extensively used for longer than a decade.
Luckily incandescent bulbs go open circuit instead of short when they break  ;D
Imagine unscrewing all lightbulbs till you find the last one had the short.

How many audio engineers does it... nevermind

Thanks!
 
Tants won't necessarily "go out". But it happens.
Tants are still happily working in devices I built cca 20yr ago.
 
nowadays the only remaining advantage is low leakage current.

Yes - that, and the fact that their capacitance value is very stable over time. Electrolytics often loose some over time, making them very unsuitable for critical timing.

Jakob E.
 
gyraf said:
nowadays the only remaining advantage is low leakage current.

Yes - that, and the fact that their capacitance value is very stable over time. Electrolytics often loose some over time, making them very unsuitable for critical timing.

Jakob E.

Is the long term stability fact that the dielectric is (dry) oxide rather than the electrolytic wet paste that dries out over years?
 
Back
Top