Metal film cap sizes.

[Fluoroscope 5000]

Not sure if this is the correct section for this question, but I've been reading that Metalized Film caps color the sound least in audio circuitry. If this is true, I'd want to replace any cheap electrolytic audio path caps in my A/I unit and powered monitors with good quality MF. I am comfortable enough with my soldering skills to be able to do this, and I think I can figure out the caps to be replaced and all the the correct values, but I have some questions:

I'm not sure where to find MF caps that will fit in the PCB lead spacings of under 1cm, and if whether they should all be polarized?

I also question whether replacing neg feedback bleed caps would have an affect on the sound, since the signal passing through them goes to ground? One of those caps in my monitors is 100uF, and the MF caps I have found at that value are rather expensive.

Finally, are all film caps actually MF, or should I stick with the specific denotation?

I know how to use the filtering system in Mouser, but I'm not sure there are many small enough lead spaced MF caps there.

Thanks in advance for your time.

-Greg

[J M Fahey]

You wonīt find many electrolytics in the sound path, if any at all, in modern designs.
The very few you do, as in that 100uF one in the power amp feedback path, are there because they are the sensible option.
As in: yes, you can replace all steel bolts in your car with Titanium ones, and the bolts themselves will be "better" (and definitely 100X more expensive) .
The Million Dollar question is: will your *car* be better?

Will a 100uF metal film cap be "better" than a similar electrolytic?
Well, it will probably measure better in some aspects .... doubt any difference will be audible and thatīs the point.

Doubt (in fact I am quite certain they donīt) any Pro stuff, any brand, uses 100uF metal films there instead of electrolytics, maybe thereīs some reason for that.

Now on "audiophile" stuff, be my guest.
"Way more expensive=better perceived sound" .... I wonder why

In a nutshell: donīt overthink it.

[Enzo]

And something else to consider: a 100uf 50v electrolytic cap might be the size of the last joint of my pinkie. The same thing in a film cap might be the size of my whole thumb. Always check the data sheets for dimensions. You might find your "better" cap doesn't fit the space, and would have to be mounted sticking up in the air or some other offense.

[The Dude]

And something else to consider:

The "whatever" was designed this way. Engineers tweaked the circuits to sound good with those parts. Changing to "different" parts, if at all perceptible, may or may not be "better". My personal view is that the end result of changing out any caps in the audio path will be near imperceptible- that is, unless some of those changed are bad or near bad.
That aside: Consider an amp designer designs an amp to sound good with a particular speaker. You replace the speaker with a "new and improved" more expensive speaker. In many cases, the amp will sound worse/not better because other components in the amp were adjusted for optimum sound with that particular original speaker.

[Fluoroscope 5000]

Thanks for your advice guys. I really don't know enough about it all to know if I'm overthinking or not. If I do decide to do anything, I may just swap the input coupling caps. True, it may not be worth the trouble and there's always the chance of destroying something. I actually doubt the cap types were compensated for in either piece of equipment, but it's a good point. It's all mass produced stuff, and I seriously doubt even the digital EQ correction in the monitors goes that far. I certainly won't be doing anything without more research. I'm still interested in finding reasonably priced MF caps and/or good places to buy. I've read that Panasonic are good bet, but the lead spacings look to wide.

[Timpanogos Slim]

The lore about capacitors is endless, but there is only one real truth: The only way to know what a capacitor sounds like in a given circuit is to actually try it.

The Accepted Wisdom is that paper-and-foil caps sound best, followed by film-and-foil, followed by whatever, etc.

Some of the best sounding capacitors i have are a bunch of 400v 3.3uf metallized polyester aerovox caps i got in a "mystery box" from goldmine electronics probably 10 years ago. Like a pound of them. They are amazingly transparent. They were intended to be used in switching power supplies or AC motor controllers or something.

There are a lot of great sounding caps that are metallized polypropylene. I have a bunch of 5uf 100v metallized polycarbonate caps (with shielding foil!) that were explicitly designed for high frequency switching power supplies - they sound excellent.

Most other capacitor compositions have come a long way in the last few decades. You've heard that ceramic capacitors are terrible for audio? Not so if they are C0G MLCCs. But they aren't available in high values.

Electrolytic caps have made huge advances, and a high quality low-esr or for-audio modern electrolytic may sound as good or better vs. a film cap -- but there's a fair chance it will only do so for 5-6 years.

WIMA, AVX, Kemet, few other companies make boxed film caps with 5mm or 7mm lead spacing that can be squeezed into most circuits where a narrow electrolytic used to be. Get them from Mouser, Digikey, Newark-in-one (or other In-One outlets like Farnell), Allied, etc. WIMA has a really good reputation and i am reasonably confident that it is predicated on how attractive their boxed capacitors *look. All matte finish and soft corners, mmm, baby . . . .

If we're talking about equipment that was made more than 10 years ago, and has been in regular use or stored somewhere hot and dry, there's a reasonable possibility that you can improve the sound with a re-cap. Values larger than 1uf don't get replaced with films at my house. I've had very good results with pretty much any "for audio" electrolytic i have bought from a reputable major electronics distributor. Also, fwiw, I think that Nichicon UPW low-ESR power supply caps sound good in the signal path, but nobody believes me.

[J M Fahey]

If you can hear what a cap does, then itīs grossly abused or very poorly designed.

I actually doubt the cap types were compensated for in either piece of equipment,

please give the designer some credit.
Every part type is chosen for a reason and is part of a whole system, which must perform accordingly.
Say, maybe a piece of equipment has 0.2% distortion and is -3dB down at 50 kHz.

Replacing a cap with an "improved" one might change that to 0.195% and -2.95dB down, a 0.005% improvement in distortion and 0.05dB flatter response ... and that at 50 kHz.

It's all mass produced stuff,

So what?
I trust much more stuff made *this* way
https://youtu.be/155Ps75_y6s?t=4m59s
than hand inserted and hand soldered stuff which at least allows for human error.
As in: Iīm certain that a machine built Smartphone or Notebook is 1000 times more reliable than same handmade, any day of the week.

and I seriously doubt even the digital EQ correction in the monitors goes that far

.
So they are smart enough to design and program Digital EQ but not to choose the proper capacitor?

Maybe because I can do the second but not the first I respect those who can. A lot.

[Enzo]

The problem here is the term "better". Saying caps sound better than others is meaningless unless we add a context. Caps in the tone stack? Cathode bypass caps? Filter caps? Power supply bypass caps? Stability caps? And even in a specific case, say interstage cap from plate to grid in a Fender something, what does better even mean? Presumably someone likes it better, but what about the sound makes it so?

And when side by side comparing cap types, one must not stop at the materials. Have we also gone through our test caps to make sure all are at the exact same capacitance value? That ESRs match up? Caps are not know for their precision. So comparing say a Mallory and an Orange Drop at some value - say 0.047uf - is not fair if one of them measures 0.042uf and the other 0.052uf.

[Fluoroscope 5000]

I have no doubt JBL picked the best caps for the job, while keeping within a budget. When I say the digi EQing is probably not set up to address cap anomalies, I mean it probably isn't programmed to level off every minor variance in the signal path. There simply aren't enough bands, if that part of the chip is even utilized. My definition of better sounding in this case is: coloring the signal the least. When I mean mass produced, I mean consumer level. I'd hardly compare cheap JBL monitors to something by B & W, but they sound very good to me. I'd really have to treat my room b4 making any changes to the speakers. It really does seem like more trouble than it's worth, but thanks again for the recommendations. If anything, this thread may be educational for others.

[Enzo]

I appreciate your answer, but it is a bit circular. What amount of colorization is happening now and at which components?

[Timpanogos Slim]

If you can hear what a cap does, then itīs grossly abused or very poorly designed.
please give the designer some credit.
Every part type is chosen for a reason and is part of a whole system, which must perform accordingly.
Say, maybe a piece of equipment has 0.2% distortion and is -3dB down at 50 kHz.

Replacing a cap with an "improved" one might change that to 0.195% and -2.95dB down, a 0.005% improvement in distortion and 0.05dB flatter response ... and that at 50 kHz.

So what?
I trust much more stuff made *this* way
https://youtu.be/155Ps75_y6s?t=4m59s
than hand inserted and hand soldered stuff which at least allows for human error.
As in: Iīm certain that a machine built Smartphone or Notebook is 1000 times more reliable than same handmade, any day of the week.
.
So they are smart enough to design and program Digital EQ but not to choose the proper capacitor?

Maybe because I can do the second but not the first I respect those who can. A lot.

I give designers plenty of credit, but I'm aware that the budget of time and materials for the design phase can be very restrictive, and that a lot can change between design and manufacture. Particularly if you outsource your manufacturing, and particularly if you don't have much of a quality control budget.

A friend tells me that one of his buddies from college designed home entertainment electronics at RCA back when RCA actually did that. He said that they would design it the best that they could, and then they would make several versions of the product - from the very best version with all the circuit tweaks to just barely working and as cheap as possible.

They would then bring in groups of people off the street who were paid a few bucks for their opinion, and the version just slightly better than the one where most people complained about the drop in quality was the version that went into production. This is one of the ways that you end up with boards with a lot of unfilled component positions.

You'd hope that things are better in pro audio, but who knows? They are still designing to a price point.

Maybe the people in charge of managing your outsouruced manufacturing are very open to parts substitutions that shave a few dimes off the per-unit cost. Maybe someone at the outsourced factory is selling the good parts out the back door and buying counterfeits. This is actually a huge problem.

There is also the consideration that consumer oriented electronics are usually designed with a short life span in mind. They don't imagine or desire that you'll still want to use it in five years. Maybe not in two years. So who cares if an electrolytic signal coupling capacitor dries out and starts leaking current after 6 or 7 years? I may still be using the home theater receiver i bought in 2007, but onkyo never intended that. I'm even still using the one i bought in 2001. Throwaway culture is very real - for example, some research suggests that in the USA the average article of clothing is worn just 7 times after it is purchased.

You'd hope that pro audio is better than that. But it isn't always.

Another consideration - my exposure to working EETs to date, and conversations about audio electronics with them, has informed me that a majority of them don't think much of the needs of audiophiles. That in the old tradition of the 70's and 80's, the "For Audio" part is actually a lesser spec than "general purpose". You mean it only has to work to 15khz? Then we could throw anything at it.

I open up a chassis and see 10% carbon film resistors throughout, and a raft of very low dollar chips, and while i do give the designers some credit, I have to be aware that they are working within constraints.