Question about old electrolytic cap tolerances

[Marcel_D]

I'm going through the parts list of an HP 333A Distortion Analyzer (~1985) and I'm reading something odd.

Cap: Fixed aluminum electrolytic 50uF +100% -10% 6vdc

Capacitors back then really had such bad tolerances? I mean +100% sounds pretty unreasonable in something like a distortion analyzer.

This would lead me to believe recapping the beast with modern caps would significantly reduce its drift and improve performance? Thanks.

[sreten]

Hi,

Yes they did have such poor tolerances in the past. Basically to guarantee
say 66uF 90% of the time you would build say 100uF caps, sold as 66uF.

Selection seems obvious, but values drift over time, initially quite quickly.

However the issue totally depends on what the cap is doing. If its a
coupling cap for say 5Hz roll-off then -10% is 5.5Hz, + 100% 2.5Hz,
and the wide specification barely effects the analyzers performance.

Also the spec for the part may not remotely reflect the part used.

rgds, sreten.

I'm not sure recapping would improve an analyzers performance that
much, but it depends on what you use for and how you push limits.

[Apex Jr]

The standard back then was + 80% -20% didn't know there
were other call outs...

Steve @ Apex Jr.

[Marcel_D]

Thanks sreten, this unit is turning into a little history lesson for me.

@Apex: It seems to have caps of like...every tolerance. The ceramics seem to be 80,20, mica are +-5, tantalum +-10...best I've seen is a +-2% mica. Not bad.

[Conrad Hoffman]

-20 and +80 is common and no doubt looser tolerances existed. My usual advice is to prove a cap is bad before replacing it, but I've had such poor luck with 6V electrolytics, I'd automatically replace them with new. If there's any way it would fit, I'd go with 16V, as the 6V jobs are generally inferior right from the start. It seems like almost every piece of test equipment I've worked on that used 6V caps, needed them replaced. They can also fool you on measurement, as a crazy high dissipation factor will show up as a normal or even higher capacitance measurement; you won't know this if your meter doesn't read DF!

[stratus46]

Originally Posted by Conrad Hoffman

-20 and +80 is common and no doubt looser tolerances existed. My usual advice is to prove a cap is bad before replacing it, but I've had such poor luck with 6V electrolytics, I'd automatically replace them with new. If there's any way it would fit, I'd go with 16V, as the 6V jobs are generally inferior right from the start. It seems like almost every piece of test equipment I've worked on that used 6V caps, needed them replaced. They can also fool you on measurement, as a crazy high dissipation factor will show up as a normal or even higher capacitance measurement; you won't know this if your meter doesn't read DF!

Through hole or SMT? I've found the SMT electrolytic caps to be the most crappy things EVER in the capacitor world. For older gear upgrading the Voltage in the same package size is pretty easy as the newer caps are WAY better. Panasonic FK or FP for SMT. FM or FC for through hole but keep in mind I'm not into subtle nuances as these are restoring video tape decks.

G²

[Marcel_D]

I've got an ESR meter on the way. I'm at a point where I have a bunch of vintage audio stuff that probably needs recapping but I figured why change them all if some are all right. Saves me time and money. Or at least it would make me feel better to know the caps I'm changing are actually in bad shape. If I'm lucky I might hear a difference too.