Output Capacitor Upgrading Questions

[williaty]

Well, I have a slightly different version of this question. I'm having a fair bit of trouble finding a place that has 100uF caps in stock. Heck, not that many different kinds of "audio" 100uF caps seem to exist at all. Here's what I've been able to find so far:

Mundorf M-Cap MKP 100uF-250V
Mundorf M-TubeCap 100uF-550V
Obligatto Film Axial 100uF-400V

Axon True Cap 91uF-250V
Solen PB-MKP-FC Metalized Polypro 100uF-400V
Solen Fastcap PPE 100uF-630V


Given that the rating is supposed to be around 200V and I've seen a couple of comments on this board about higher voltage ratings potentially sounding worse, that really leaves me with the Mundorf MCap MKP and the Axon True Cap. Are either of these any good? Given that the Axon (which at least one member has used) is only a 91uF, should I use a 10uF cap (which are MUCH more available) in parallel to it?

 

[InfernoSTi]

Hmm, interesting. Is there a magic ratio between the body value and the bypass value? I was looking at a 100uF Solen and maybe some value of a Mundorf Supreme or ESA Clarity Cap between .47-2 uF.

I have heard the general rule is 1/10th for the bypass cap.  My understanding is that this can be reduced if the cap has a lower tolerance: if it is a 10% tolerance, then the 1/10th rule would apply but if it is a 2% tolerance then a much smaller bypass cap could be used, say 1/50th or some such figure.  Morgan Jones in "Valve Amplifiers" uses the figure 1/100th for modern, high quality caps.  I have also heard the ESR can be lowered by using parallel caps (two 50 uF caps in parallel would have a lower ESR than an equivalent 100 uF cap).  Morgan Jones also recommends "starring" the bypass points.

Just a thought or two about bypassing.

John

 

[deltaunit]

Has anyone experience of bypassing the 100uf output capacitors with a much smaller .1uf or .01uf capacitor?

I was considering going with a 1/100 1uf bypass but I don't think I have the room in the case for the type I want, a high quality AmpOhm Polyester Film. I could fit a lower quality 1uf bypass and bypass that with an even small high quality .01uf or I could just use a single high quality .1uf bypass.

Any opinions on which would be the better option? My output caps are already polyprops.

Cheers

 

[Beefy]

Axon True Cap 91uF-250V

This is what I used in my build. I can't compare to any of the more exotic/expensive options, but *definitely* better than the stock electrolytic caps - and they are dirt cheap to boot.

 

[jrebman]

Any of these will be an improvement over an electrolytic IMO, with a couple of exotic exceptions.  The actual voltage is much closer to 100v than 250, but the difference in dissipation factor between  a low voltage electrolytic and a higher rated film cap is pretty much always goinng to go to the film cap.  That said, I don't know that I'd use the m-tube caps as these are really meant to be power supply caps and they may or may not sound as good as the others.

You should also be able to get the axon in 100 uF 250v from Michael percy audio, and at least one other person here used the erse caps.

That said, even the 91 uf should be just fine -- the value herre is not super critical so a bit more or a bit less won't make muc of a difference.

As for bypasses, you'd really just have to expreriment and any of the values mentioned should work fine.

Even a good film or teflon film cap may also be used in combination with the stock electrolytic and should also show some benefit.

If I were doing this, I'd try an axon, erse pulsecap, th mundoorf MKP, solen or obligato -- in no particular order and thenbypass wiith a .1 uF sonicap platinum teflon, or even maybe a .01.

I plan to use a pair of now unobtanium blackgate 150 uF 350v and may bypass them with either a sonicap platinum or a vishay ERO -- if I can hear a difference.

Again, in this application, these values (except for the minium voltage rating) are all that critical -- so 901 uF, 150 uF, no biggie.

Of course there are other cap and bypass choices, but these are what I have experience with and or prefer over some of the others.

Good luuck,

Jim

 

[Laudanum]

What about bypassing the stock electrolytic output caps with a largish film cap?  Maybe 10 or 20uf or even a bit higher value. I would think that would have a larger influence on sonics than a small bypass and wouldnt cost as much, if staying within the same cap brand/range as a full out film replacement.  Im kinda skeptical that a high quality, small value (.1uf or lower) bypass cap is going to have much influence bypassing a 100uf electrolytic.  I do have some Audiocap Thetas I could try but I believe that they are .01uf .  And some larger value Dayton Film and Foils (all under 1uf, .33 may be the highest value I have).

I dont quite understand the potential resonance issues, would larger film bypass caps exacerbate them more than small value bypass caps?  What does is "sound" like?

Also, Jim ... I see that you're considering the Black Gate electrolytics, what about the Jensen Electrolytics?  In 100uf or even 220uf, they are a more compact alternative to the large value film caps and less expensive than all but the cheapest film caps.  I have used Jensen oilers (1uf) as output caps in a CDP and also to cap a "helper" tweeter in loudspeaker but that is my only Jensen cap experience, no experience with their electrolytics.    Im not that big on Solens and not so sure that the Axons are all that much different. I have used both in speaker crossovers in the past, as well as Dayton PP's so i should say that I never really liked them much for tweeter applications but I think some are still in service on drivers other than tweeters.  Also used them non-critically in car speaker crossovers. 
A Dayton film and foil bypassed with an Audiocap Theta did work out well in a phono stage I built a decade or so ago but that was all small value, under 1uf total and again, not a standard Dayton PP. 

This is more food for thought for me for a later date but I know this much, I cant put a load of money into trying different large value film caps.  If a cap "upgrade" is going to be subtle, especially with the music I listen to, it just becomes a diminishing return thing for me right off the bat.  I would probably do Speedball first regardless, since most say this is the biggest improvement. 

A lot of questions, I know.  Thoughts?

 

[Grainger49]

From Parts Express:

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=027-616

Solen 100uF/400V $34.55, now $41 each.  Higher values are rated at 250V.

 

[jrebman]

Laudenum,

A small, high-quality film cap bypassing a large electrolytic is fairly common practice and is not so much meant for "flavoring" the electrolytic to the sound of the film cap, but more along the lines of lowering the dissipation factor and/or shunting the self-noise of the electrolytic.  Go ahead and try the .01 -- certainly can't hurt, but if somehow it does, just take it out.  There is another cap rolling thread on the Crack, and somewhere in there I was discussiing the possiblity of bypassing the stock electrolytics with a .1 or .01 vishay mkp-1837, but as PB pointed out, the working voltage of the vishay is too low for this application.

Elsewhere in this forum others have spoken of bypassing psu electrolytics with smaller, as well as larger film caps -- the smaller, again are more for shunting out self-noise and the larger ones are more for adding capacitance and or flavoring.

I don't know anything about using the jensen electrolytics as coupling caps, and as good as they are, they are not the same kind of chemistry as a black gate, and besides, I've had a pair of closely matched 150 uF 350v BG NH caps for a while and will put them to use in this amp.

There is a "rule" that says you should not bypass a black gate, but with certain types (NH, NX, and others), and in certain situations, this can be done to god effect.

I built a millet minimax headphone mp last year for my dad and used a couple of black gate N types as output coupling caps in much the same way as they are used in the Crack, and indeed, bypassing them with a .1 uF vishay mkp-1837 was a very nice improvement -- and the caps were either 470 uF or 1000 uF if I recall correctly.

With all that said, the stock electrolytics are supposedly quite good as they are, and bypassing them with a small, quality film cap may be all one needs to do.

Unfortunately, I don't have access to the kind of instrumentation (scope) that would let me see what happens exactly when bypasses are used and/or when a resonant situation happens, so I just have to trust my ears, and I've never had a situation where bypassing a big ps electrolytic with a small film cap did not make a positive difference.  Sometimes the differences were fairly small, and in other situations, very, very good.

If you don't want to go for the sonicap platinums, then also look at the Gen IIs, as well as others with sufficient voltage ratings.  Sonicaps, IMO, just happen to be quite nice, especially for the money.  Unfortunately, most of the larger value caps, and all of the platinums are generally not suitable for BH gear as their voltage ratings are somewhat lower, but they do have the basic sonicaps up to at least 3.3 uF in 600v.

Hope this helps somewhat, but as I said, without any way of measuring or looking at what is going on, I'm basically just following the crowd and hearing what I hear.

HTH,

Jim

 

[Laudanum]

It does help, thanks Jim.  It cant hurt to try what I have on hand.  I can just clip them in temporarily and see how it goes.

I dont understand all the tech that made the Black Gates different but I do know that they are different internally than Jensens and other electrolytics.  I have some in use in one of the kits or rebuild I did years ago which means I have spares somewhere.  But they werent the N series and Im sure they arent the right values/voltages.  Probably the standards because I was always a budget type. 

 

[Grainger49]

Jumper leads and a pair of KK Teflons are what got me started on the capacitor bypassing trek.

 

[Laudanum]

Yeah, I should probably pick up some of the affordable, smaller value russian teflons that show up on ebay all the time.  That way I'll have them.  Ive never used a teflon cap before.  Forget the big brand name teflons, too expensive for me.  But I suppose it would be worth trying some of those russian cheapies.  What about polystyrene for bypass?  I assume they arent a great choice since I havent read much about them.  But I think I may have some from a couple lots I picked up really cheap on ebay years ago that I figured would work as guitar tone caps.  I think that they were advertised as polystyrene anyway, if memory serves.  They may just be mylars but they indeed make for pretty nice tone caps in guitars.  I have 2 or 3 different lower values but not sure of voltage rating or particulars.    I'll dig them out and see what they are exactly if the experts here think a polystyrene bypass is worth a listen.

 

[Laudanum]

I may have a line on a couple of 150uf solens (used/upgraded).  Would there be any reason to change the stock resistors at the headphone jack with lower value resistors?  I know it was suggested by Paul B to do so if going to a 330uf cap.  Just wondering if it would be considered necessary or a good idea if jumping to a 150uf and if so, what value would be suggested?

Thanks

 

[hopeful]

I finally dropped my Axon 91uF caps in last week - after spending quite a bit of time with the stock crack (except for a tkd pot), and then the speedball upgrade.  I would say the Axon's improve on the stock output caps, but the improvement is pretty subtle.  I would characterize the sound as being the tiniest bit more lush, and overall, a little smoother than before - but without any perceptible loss in detail.  This makes me really curious what, if any, difference I would perceive if I had a chance to try out some obbligatos or mundorf mkp's.

 

[Laudanum]

I have two of the Mundorf MKP 100uF 250V on the way.  I havent been all that impessed by the usual suspects ... the lower cost MKP's in other applications in the past, both electronic and speaker crossover.  I seriously considered the 91uF Axon regardless however as PC has them cheap.  Also considered the 100uF Erse as they are also inexpensive relative to most other MKP types in large values.  But I just didnt want to go down and already traveled road with the Axon or take a chance on the Erse.  Just seemed like the Mundorf was generally viewed as just a small step above the usual suspects and aside from the Axon through PC or the Erse, they arent all that much more money than the others.  I was hesistant to spend the money here because the amp sounds fantastic stock.  Speedball was probably the best bet, money wise.  But speedball is on the to do soon list. 
Im not expecting a big difference with the Mundorf.  Im hoping that they are a neutral sounding cap, with maybe a bit smoother upper mids and treble and a hair more detailed than the stock cap.  I suspect that it will be fairly subtle.  I have some Blackgate 150uF vk's in the parts bin, leftovers from old project.  I was going to try those at the output and still may if the Mundorf is a loser.  But jrebman gave me the idea of using these in parallel replacing the last PS cap.  I probably will do so of it's not too much trouble.
I'll report on the Mundorf's in the not too distant future. 

 

[Grainger49]

I am a confessed capacitor junkie, not in a 12 step program yet.  But the Speedball will bring you more in changes than tubes and capacitors.  So it might be the first upgrade.  Then...

After tubes and capacitor changes there are cables and speaker wires, tube sockets, resistors... the list goes on.  It can become a neurosis or a joy. 

But you will get an improvement over the stock caps.  As you say expect better, clearer, cleaner highs and improved presence and detail in the midrange.  It has been my experience that the better the cap the longer the break in period.  So there will be some time you will want to go back to the stock caps, persevere! 

 

[hopeful]

Laudanum - with respect to the order of your upgrades and ease of installation, you may want to consider putting your speedball in first.  It is going to be harder to angle your iron and gain access the various necessary terminals for the speedball once your large film output caps are in place (unless you plan to simply remove and then reinstall them at the time).  Per the advice I got from Beefy a while back (which worked well for me), it is probably easiest to put in fly leads and mounts for the output caps before installing the speedball, then do the speedball, then solder in and zip-tie down your new caps at the end.

Setting the above aside, I'll be eager to see your comments on your mundorfs whenever you get them isntalled!  Are these the ones you ordered in 100uf/250 http://www.partsconnexion.com/capacitor_film_mundorf_mkp.html ?

 

[Beefy]

Per the advice I got from Beefy a while back (which worked well for me), it is probably easiest to put in fly leads and mounts for the output caps before installing the speedball, then do the speedball, then solder in and zip-tie down your new caps at the end.

Yep, and I absolutely stand by that advice. My original build post on Head-Fi for reference: https://www.head-fi.org/forum/thread/476650/crack-bottlehead-otl/435#post_7149721

I am a confessed capacitor junkie, not in a 12 step program yet.  But the Speedball will bring you more in changes than tubes and capacitors.  So it might be the first upgrade.  Then...

After tubes and capacitor changes there are cables and speaker wires, tube sockets, resistors... the list goes on.  It can become a neurosis or a joy.

Yep, I would agree that Speedball certainly made a bigger change than the output caps, but even the 'lowly' Axons were still a good improvement over the stock electrolytics.

Aside from those two upgrades, I would put a better pot on top of the list - well above the other upgrade options you mention. The TKD I used in my Crack is gorgeous compared to the Alps blue in another one of my amps. IMHO, YMMV, $0.02 and all that......

 

[Laudanum]

Yep, I know that Speedball should be first but I never do things the easy or logical way anyway.   I plan on purchasing Speedball soon.  But I will probably install the Mundorfs first.  I'll secure them with self adhesive tie wrap hold downs, atleast temporarily. And solder them in so they are easy to remove for Speedball.  Im not very good at understanding circuits or schematics but Im pretty decent with a soldering iron so I have no worries regarding removing the film caps and desoldering and resoldering anything else that may make Speedball installation easier.    

Ive been down the upgrade path several times before but Im not an extremist.  I know that the caps can take a while to break in.  Teflon sockets and expensive cables wont be in the future.  Although I do probably need to make some different cables eventually.  I used Star Quad as I had some already and it was the easiest to use for the source (mini plug to RCA's) but the capacitance is pretty high with the Quad.  Although with the short runs, it probably isnt going to matter.  But Mogami or Canare is good enough for me.  I'll get around to the cables sometime down the road.  Also, headphone cable upgrades wont happen either and neither will internal wiring upgrades, atleast not with boutique wire or cable anyway.  
Speedball, the Alps Blue pot and the Caps and I'll have about the price of the stock Amp, or maybe a few bucks more, in upgrades when it's all said and done.  Still a steal of a headphone amp at 400 bucks all told.  

hopeful ... yep those are the caps.  I wonder how much it costs Mundorf to have them things made?  5 bucks a pop maybe ??? ...   You can get a glimpse of the retail mark-up just by seeing how much less PC charges for the Axons compared to some other sellers.  I bet the Mundorfs dont cost any more than the Erse, Solen, Axon, Dayton etc. to make and if they do it's not much.  But in the audio world, 45 bucks isnt all that much for a 100uf 250V film cap.  


Good input, thanks guys.

 

[Laudanum]

Arent the TKD pots conductive plastic?  I may be a skeptic, but I just cant bring myself to spend gobs of money on a pot.  In fact, I would loved to just use the included alpha pot but the tracking at low volumes is generally poor.  I have several of those alpha pots and they are all over the place from sample to sample.  The Alps blue velvet tracks very well throughout it's range and is smooth.  I know that there are several popular pots that are supposed to be better and sound better, the TKD's being one of them often mentioned, but I just cant bring myself to spend much more than what the Alps cost.  And again, I am skeptical that two different conductive plastic pots is going to sound much different. But then again, I havent spent the money to find out what I may be missing.

 

[Doc B]

PECs are nice pots for the money.