Best electrolytic capacitors

Hi Sam BG N fantastic for bypass but only a sprinkle, superb dynamics & bold sounding. My fav now is ES bipolar very very good transparency & depth with very even sound from top to bottom. FG a little too hi fi sounding. FZ also very nice & even sounding, bass is it's forte quite dynamic but also depends on which circuit your using on cause it may sound a little too up front. Overall caps have their sonic signatures, you may know the sound but it does not mean that it will sound good on any location that you put in. That means you've got to try different caps in that position, listen & then tweak from there. Also it's a big mistake to use the same caps to populate the boards. One should learn to mix & match with other brands or types of caps to arrive at the sound that your after.

Cheers
 
BG N can only use 1 or 2 max & value cannot be high for SS tube another story
You have to experiment yourself. There's no universal answer to this. Like Jean
Paul says this sort of tweaking becomes an Art. Local decoupling Im using a mix
of ES bipolar & BG std. FG not for me KZ better for main supply but as I mentioned
all these are circuit depandant. What's really import at is after you find the right
combination that you like, the next thing is to try different values of capacitance.
Contrary to what many have said more capacitance the better frankly its hog wash
to me. Either they're poor listeners which really is the fundimental to good tweaking
or they have not tweak enough.
 
Keep in mind that we're discussing analog side for these tweaks & this applies
dac as well. Panasonic FC is also not bad it brings realism to instruments but
what I dislike about it at least from my equipment & room is that the presentation
is too up front for my taste. Again a mix & match might bring you good results

Cheers
 
combination that you like, the next thing is to try different values of capacitance. Contrary to what many have said more capacitance the better frankly its hog washto me. Either they're poor listeners which really is the fundimental to good tweaking or they have not tweak enough.

I agree, but some exceptions : you have to play with the voltage values as well; sometimes i saw a higher capacitance, let say : x2 is not worst with some increased values in voltage ! But again it's not a general conclusion, take again the same capacitance value but increase this time even more the voltage in relation to the last test and the result will be not good anymore ! Of course at iso perimeter, i.e. with the same cap model ! What changed each time : it could be the space between leads (inductance), diameter of the leads, also the ESR,ESL : but here again I was not able to conclude a rule from there !

So all said Sumotan : all is about to try & experiment, of course expérience matters at this game/art; all is also about sound compensation between the hifi stuffs you have : amp, speakersn dac, etc....

I prefer the FC over the FM : these last can have a subjective feeling of odd harmonic in the mid-treble despite a warm/soft presentation... often ! FM, hould not be alone on a pcb, or at rectifier position imho ! FR are not better than the FC, but some exceptions: try & error is your friend again ! I use it on my dac : same value/voltage : but not exactly the same FR cap body size : a difference can be heard at powering a voltage rail of a dac when I swap between each :surprising no, me the first here : but all true...:scratch: (same esr !)
 
Last edited by a moderator:
Yes Eldam I forgot to mention contrary to what has been said I generally find that higher voltage of same brands caps & value to sound better .
Why I don't know. Yes agree this tweaking stuff should be done for our whole system & the speakers are the ones that reflects our tweaking.What is correct for my ML speakers may have different presentations with box speakers. Off course ML being bipolar its far more sensitive especially towards soundstage presentation. Use the wrong cap sound will be in your face which I dislike most. Again this is personal some may like this kind of presentation. This is the real beauty of DIY,
just like cooking, we mix & match components to suit our taste.


Cheers
 
Last edited by a moderator:
Yes Eldam I forgot to mention contrary to what has been said I generally find that higher voltage of same brands caps & value to sound better .
Why I don't know.

While looking for some caps for a current restoration, I've found that higher voltage varieties of the same cap will often have lower dissipation factor (tan theta) and lower ESR than a lower voltage version of the same cap. So, the higher voltage cap is basically less lossy, and should sound 'better' in the clean sense.
 
Hi Monte,
I've found that higher voltage varieties of the same cap will often have lower dissipation factor (tan theta) and lower ESR than a lower voltage version of the same cap.
What you are looking at is called dissipation and is the inverse of "Q". The higher voltage parts generally have a better dielectric and will also tend to fit the old mounting location of a capacitor better if you get the same lead spacing. If you don't form the leads properly, you can damage the seal when you jam the caps in and force their legs apart. This is almost universally how I see capacitors installed by folks doing a "recap job". Anyone familiar with my posts know that I dislike the idea of people diving into a working product (or good non-functional device). Often their soldering skills and equipment are not up to par and damage to the PCB or parts is commonly done. Even the act of pulling a PCB can cause damage.

So for starters, if you are not using a soldering station (around $90 ~$120 in Canada) and a solder sucker (the big blue ones) along with some solder wick, then don't even consider attempting this work. Finally, and this is just too hard for some of you to resist, do not work on anyone's equipment but your own. The main problem is that you don't know enough to understand what you don't know. I've seen all kinds of horror stories created by some well meaning "friend". Just because something works after you're done doesn't mean you haven't caused damage. Technicians are actually trained in proper soldering techniques. It's a learned skill. Some "technicians" haven't been trained in soldering, anyone can do that - right? No, the better technicians are getting really tired of cleaning up after the cowboys out there.

So if you want to do your own equipment, please think about this first. Then, take pictures before, during and after your work and get set up to do this in a careful, methodical way. Anyone who rushes will make mistakes or destroy some copper. If you do, don't be angry that it might take a couple hours to fix the problems - or be told that the equipment is too damaged to be economically repaired. (BER). I've had to tell more than a couple people over the years that they destroyed a piece of equipment. Some even still in warranty they were so new! It never fails, then they are angry with me! Me, the guy who makes some money (or loses it on a job) by repairing equipment. It isn't in my best interests to write something off (in that I don't make any money) unless that equipment will either be too expensive to repair, or it will never be reliable again. Those are my main limits, and I think they are in the best interests of the equipment owner. You should see what happens if a "friend" is the one who did the work!

Anyway, just some observations from many years in the business.

Now for those who think I'm being one-sided about this, I recently met a guy without training who does a wonderfully clean soldering job. I've told him that I will work on his stuff if he gets stuck and that his workmanship is cleaner than many technicians I know. But people like him are not common.

Last parting thoughts. For those of you who like to experiment with different capacitors. PC boards will only take so much work. Do not replace components any more than is necessary. That might mean that you remove the component and "tack solder" the part in question on a pair of solder joint close to the location and only solder them in once you have decided for certain what you want in there. If a capacitor is too large for the original mounting location, you can't use it! Be careful desoldering that you don't pull out the plating in the holes, or drill them out when forcing a larger diameter lead through that hole. Something else I see a lot of. So, if the part doesn't fit due to lead or body size, DO NOT USE IT !!!

-Chris
 
Last edited by a moderator:
Thanks for the advice, but this isn't my first rodeo 🙂 I have a Hakko FR 300 instead of a Soldapult - the plunger solder suckers always seemed to cause the iron to slightly nick the pad or the PCB, and they're useless for multilayer PCBs with lots of copper. For SMD reflow, I have a Hakko FR 810 hot air tool and an FR 872 infrared pre-heater.

For handheld and through-hole work, I have two different Weller SMT tweezers and a few Weller temperature controlled stations with various handles from 40W to 80W and tips from about a half mm to a half cm, so I can handle anything from hand soldering an SC-70-6 package or soldering a shielding can to a ground plane.

Flux is also quite useful, especially for rework. I have a few Kester flux pens as well as a huge jug of Kester 959 liquid flux. The 959 is nice to use, since it's isopropanol based, so it can serve both as a quick clean-up and re-flux. It's a no-clean flux, so I don't have to worry about residues unless I'm working on something like a condenser mike impedance converter. By cleaning off old crusty 44 flux from 50 years ago, you can help to prevent that icky white stuff on the PCB.

Fortunately, this restoration is of a couple of RCA BA-6A, and all of the electrolytic caps are mounted in octal plugs. Out of 3 units worth of caps, I found a total of only one set of caps that actually look like caps anymore, and there is no reason to find NOS or even modern replacements. So, I'm building modern, high performance caps into new octal plugs, and it should go pretty well. New caps can be found with 10,000 hour ratings and very low impedance, which should remove the need to replace the caps every few years. There is one dual cathode bypass cap that gets cooked by a nearby 6V6 output tube, but again, it's mounted on an octal plug, so it can just be swapped out without touching a single joint on the BA-6A itself. This limiter was designed for broadcast, to be easily and quickly serviced, so the original QC markings on each solder joint will remain intact!! 🙂

Thanks for the tip about the cap lead and the rubber bung. Yes, it's important not to stress the lead wire as it enters the capacitor! I like to use a round jaw pliers to hold each lead below the base of the cap and then bend it to fit. If you instead hold the body of the cap, you'll transmit stress from the lead wire to the rubber bung, and that could shorten the life. There's no 'original replacement' parts here so artful lead prep (and some bus wire) will be needed to get the new caps to fit the octal plugs. It should go well though.

The most annoying part of this part of the restoration is finding attractive covers for the octal plugs! I could leave the new caps exposed, and it'd be mostly safe, but it'd look too 'modern' and colorful. Trying to find the right diameter cardboard is actually sorta difficult! There is a half dollar coin sorting tube that may have the right diameter and an attractively crimped end, but I need to wait for the octal plugs to arrive to see for certain if they'll fit. It's annoying, but I may end up winding my own cardboard tubes. Anyone have any suggestions for 1-3/16" ID sleeves to cover some caps?

Anyway, thanks for the advice! I know I've destroyed some gear and PCBs (>40 years ago) and I always wonder whether this also happens with surgeons fresh out of medical school... ;-) At this point though, I am happy to not have to do any solder work on the BA-6As themselves, except to undo a coupe of old 'fixes' that a previous owner had done.

Good tools _are_ essential for rework, and even then, it can get slightly difficult. I worked on an Eventide H3000 recently that had been 'on' continuously for about 20-25 years before it died suddenly. The PCB was very brittle from the high heat, so I had to keep changing the temperature of the irons for different types of joints to accommodate the varying copper losses, without just cooking the daylights out of the PCB. It went well, but that PCB was pretty tired out, and would not withstand too many more rework cycles. A complete shotgunning with up-rated Kemet Tantalum caps should prevent that failure from ever happening again 🙂
 
Unfortunately many of the parts with better specifications, or sound characteristics in their application just don't readily lend themselves to a simple installation.
A bit of structural engineering has been the normal in most of my builds in order to not stress the board, or the parts. Brackets, braces, carefull application of potting compound have gone a long way in maintaining a reliable appliance in the end. I have compromised ultimate sound or whatever when working on someone else's gear.
I have also used nicer equipment when I have been on the clock somewhere, but I whatever heat source is used, you need to know it well, and it's limitations.
Guess that's why it "DIY" audio, huh?
 
Hi Monte,
My post is more for others that will read through later on. You've obviously got a good handle on soldering skills, which is great to see. Most do not.

For lead forming, I use a set of bent tip pliers (Xcelite, but similar are fine). As you've pointed out, it's all about keeping the stress within the leads themselves. You can get quite good at bending leads after a bit of practice. That Eventide sounds like it needed an experts touch for sure. Old Neve consoles that had been left on can be stressful to work on as well. A friend of mine made the <huge> mistake of cleaning each strip in an ultrasonic bath full of ... WD-40. That killed practically all the capacitors on each, plus many transistors that failed intermittently over the next few months. He eventually just replaced every transistor as well. He didn't repeat that particular experiment ever again. Good way to kill something and leave it perfectly clean. I think he use alcohol in the bath to clean out all the WD-40, but that WD-40 went deep into everything. Mind you, once he replaced all the problem parts, the board was just like new. I wonder why? 🙂

It really takes years to become good at anything, and that includes replacing capacitors.

-Chris
 
Agreed - my post too was to help others later on prevent needless destruction of gear that really isn't all that 'finished off'! It really does take a while to get good at rework, and every device you touch has to be approached cautiously and carefully. That Eventide was a real chore, but if you 'listen' to the PCB, you can find a way to remove and re-install components safely and efficiently. But, they're each usually unique in some way, so you have to pay attention!

Some of the modern, low price Chinese PCBs that get made for DIY projects are pretty fragile straight from the factory, so a temperature controlled iron, clean components, good flux, the right solder, and patience is still needed as new. I shudder to think of re-working those PCBs a few years down the road as the adhesives start to come apart!

Ultrasonic baths seem attractive, but they can be very dangerous with old semiconductors, like those from the '60s and '70s. If you're especially unlucky, the ultrasonic vibration can stress and fail the bond-out wire connections from the die to the package. This sounds like a good reason for your friend's misfortune with that older Neve console - those bond-outs can be pretty iffy on old silicon to begin with, and if everything goes just wrong, they will further loosen and fail. Yikes!! Plus, sourcing old silicon like that is not simple at all, especially if you want it to sound 'original'. Not to mention the problems with solvents on caps like polystyrenes - that's a delicate polymer, and even alcohol can cause problems. Ultrasonic baths are pretty much just trouble unless you're working with new, mostly unassembled mechanical parts like metalwork that has tooling grease on it. Any complex module really needs to stay away from an ultrasonic bath, or for that matter, any sort of bath! That old flux on a PCB sometimes seems to act like a conformal coating anyway, and if you're not re-soldering, leave it there.

A dry paint brush to mechanically remove the dust ought to be good enough, with maybe some carefully applied solvent for specific problems. Again, I like the 959 flux for soldering prep on old stuff, since it will remove any old flux and dirt, and also put down some new quality flux. But, there's no need to hose the PCB down - just a bit right at the connection as needed. Plus, it's a good way to work through the 1 gallon minimum they make you buy! 🙂

Well cheers to all, and best of luck with your projects!
 
Vishay MKP1837 bypass

Signal path - None.
But if I have to, then Nichicon KG at one value down from calculated plus Elna Silmic at 20% of that value in parallel (plus a Vishay MKP1837 10-150nF polyprop in parallel if more "air" is needed).
Interesting... Just wandering what nF (nano) value used for a, say, 22uF cap...
Anybody ever tried Muse and Silmic in Parallel, too?
 
Last edited by a moderator: