bypass cap on input film cap?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
before spending money on high quality input caps I experimented by adding bypass cap to existing input caps.

The current input cap is a 2.2uf polypropylene.

the polypropylene bypass caps used were :
Rifa 10nF 400V
Arcotronics 4.7nF 1600V
WIMA 0.1uF 100V MKP2

The 4.7nF sounded the best, the 0.1uf WIMA just sounded worse.

After settling on the 4.7nF and having a proper listen it is sounding like a solid improvement . better transient response and overall cleaner sound.

There was no control so no idea what factors are important.
the lower value the cap, the better it sounded. its also better to keep it low to avoid changing the filter characterisitics
the higher the quality, the better it sounded:
The Rifa and Arco have copper leads, Wima is not copper (steel?)
Arco had highest voltage rating.
 
Last edited:
There is no need to bypass coupling capacitors. At best it makes no difference at all. At worst it could (just) insert a slight suckout at the frequency where the inductance of the big cap resonates with the capacitance of the small cap; this will normally be inaudible, but if heard some people may misinterpret the change as an improvement.

laserscrape said:
the lower value the cap, the better it sounded.
This is because a low value bypass moves any resonance out of the audio frequency band, so the bypass does less damage; at worst you just get some HF phase shift. The ideal bypass value for a coupling capacitor is zero.
 
There is no need to bypass coupling capacitors. At best it makes no difference at all. At worst it could (just) insert a slight suckout at the frequency where the inductance of the big cap resonates with the capacitance of the small cap; this will normally be inaudible, but if heard some people may misinterpret the change as an improvement.


This is because a low value bypass moves any resonance out of the audio frequency band, so the bypass does less damage; at worst you just get some HF phase shift. The ideal bypass value for a coupling capacitor is zero.
appreciate the explanation. the 4.7nf bypass did seem to sound better but the least time was spent comparing the 4.7nf to no bypass cap at all because I thought it could only help, will have to check again.


also is there any potential benefit to adding a bypass cap even though its bad practice? like lower ESR?
 
Last edited:
did some tests and the 4.7nf definitely offers the best sound, there is no difference in the level of detail or tone. with the bypass cap the sound is faster, bass and treble are more independent from one another. with a very fast and bass heavy electronic music the differences are most clear.


To compare one channel had the bypass cap, the other didnt. the headphone's detachable plugs were shuffled blindly and reconnected.
for every track I would judge which sounded better then swap the plugs and judge again, this eliminates driver inconsistencies and hearing inconsistencies. The amp is op amp based so channel differences, if any, are very small but I made sure to use the opposite channel from the first test to guarantee that its the bypass cap making the difference.

the test tracks were converted to mono aswell.
all 5 tracks I prefered the bypass cap and was able to guess which was which
 
I recently used a 10uf Nichicon Muse for an input cap, and found that the Wima MKP10, in .22uf 400v works very well in unison.

It is very easy to produce interactions between the parts, that doesn’t sound very good, just have to search for combinations that work.
 
would you say the MUSE is colouring the sound in a positive way or do you consider it the most accurate cap aswell?

That's a good question. I'd have to say that it's hard to tell, the sound doesn't seem coloured to me and neither was it my aim to make the sound 'nicer'.

The MKP sounded accurate but also 'hard' sounding. I didn't like it and I would have called it coloured except that this didn't make sense to me how a film cap could sound coloured so I left it as a mystery.

In my TGM3 where I used a Tantalum (an old blue drop) the sound was definitely coloured slightly, and for the better.
 
That's a good question. I'd have to say that it's hard to tell, the sound doesn't seem coloured to me and neither was it my aim to make the sound 'nicer'.

The MKP sounded accurate but also 'hard' sounding. I didn't like it and I would have called it coloured except that this didn't make sense to me how a film cap could sound coloured so I left it as a mystery.

In my TGM3 where I used a Tantalum (an old blue drop) the sound was definitely coloured slightly, and for the better.
thats really interesting. I always used PP where possible just because they're considered to be the best on paper but in situations where the values were too big the improvements from standard caps to Elna silmics were really significant, much more significant than any film cap upgrade.
Ill definitely try replacing some PPs with these MUSE caps. thinking about it of it all these PP caps are general purpose but the MUSE are top end audio grade.
 
The only electrolytic coupling cap that I am familiar with is the Muse, and they are overall better than many of the pp parts. I’ve salvaged parts from different power supplies, other appliances over the years and they go into a tin for later use/experiments.
I try not to buy audiophile film capacitors, so the two that I have we’re done so after overwhelming reviews and a killer sale price.
I would be interested in comparing the Muse to the Mundorf bipolar, or the Kasei bp, but haven’t made that effort.
There are other electrolytic parts but they are better used in power supplies, sound dull in signal coupling, like the popular Silmic. Some companies like Marantz use them as coupling caps however, plenty of happy customers...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.