[mewrei]

I'm planning my build of one of Alex Cavalli's CTH amps and I'm loving that he left enough room for proper electrolytics (like Elna's) in such a small board. However, he did design in bipolars (non-polars) at several stages. None of these bipolars look like they're in the signal path (correct me if I'm wrong), but I'd like to get pretty good ones anyways.

I've been using Nichicon ES Muse caps, are these pretty good in this context (I'm guessing a feedback loop but I'm unfamiliar with the other application being used there), and is there a better bipolar for the job (aside from Black Gate N-series, can't find any of those in the size I need)?

Bipolars are in positions C2, C5, and C6
Input Stage Schematic
Output Stage Schematic

Outside of this context, is there a better bipolar electrolytic to use when in the signal path (outside of film capacitors where applicable)?

 

[tomb]

Quote:

Originally Posted by mewrei /img/forum/go_quote.gif I'm planning my build of one of Alex Cavalli's CTH amps and I'm loving that he left enough room for proper electrolytics (like Elna's) in such a small board. However, he did design in bipolars (non-polars) at several stages. None of these bipolars look like they're in the signal path (correct me if I'm wrong), but I'd like to get pretty good ones anyways. I've been using Nichicon ES Muse caps, are these pretty good in this context (I'm guessing a feedback loop but I'm unfamiliar with the other application being used there), and is there a better bipolar for the job (aside from Black Gate N-series, can't find any of those in the size I need)? Input Stage Schematic Output Stage Schematic Outside of this context, is there a better bipolar electrolytic to use when in the signal path (outside of film capacitors where applicable)?

I'm not the authority on the CTH, but there's no reason to use audio-quality bipolars outside of the signal path. I love ES's as much as (or more than) anyone, but there's no comparison between one of those and a good power cap in a non-signal path application. Panasonic FM and Nichicon UHE's are generally as good as it gets, with Panasonic FC's and Nichicon UPW's not too far behind. The Panasonic and Nichicon bipolars are probably a better choice, too - or even film caps if they're not too big.

You will note that with audio-quality, signal path electrolytics, they very seldom even publish the ripple and ESR data. That's for a reason.

EDIT: If they're in the signal path - or if there's even a remote chance that they may be in the signal path, then that's a different matter altogether. You'll have to hunt far and wide to find a better bipolar than the ES's.

 

[mewrei]

So in feedback loops and such, a lower-ESR capacitor is best? Theoretically, wouldn't an ultra-low ESR organic polymer cap be even better than a standard electrolytic?

 

[tomb]

Quote:

Originally Posted by mewrei /img/forum/go_quote.gif So in feedback loops and such, a lower-ESR capacitor is best? Theoretically, wouldn't an ultra-low ESR organic polymer cap be even better than a standard electrolytic?

Theoretically, yes - but organic polymers don't go to very high voltages and high capacitance at the same time. I'm not so sure about feedback loops, per se, but often in decoupling and power cap scenarios, you're not going to find an organic polymer with the voltage and capacitance rating needed - at least not in a typical headphone amplifier.

 

[jcx]

there's really no DC path to the output Q4,5 base??

I would expect >10s % base current imbalance and for highly variable results over multiple instances of the circuit with normal part variations

yes the servo will find a bias point but it will vary from channel-to-channel, amp-amp

 

[mono]

Organic polymer caps have made good strides over recent years. Now they're becoming quite common on PC motherboards, video cards, in values like 16V/330uF. I think they're a bit overkill for an analog audio amp though, except if you're building something to give to the grandchildren some day, far enough into the future that 'lytics might've dried out.

 

[cfcubed]

Couple things:

* jcx - there were of course compromises to fit CTH features on less than 3" x 5" PCB. Scale print the PCB & have a magnifying glass handy

Runeight discusses OB decisions on the site. We do find the arrangement working quite well across some 40-odd(?) builds & feedback/reviews are overwhelming positive for the design.

* mewrei - It's commonly thought, as only the C4s are in the direct signal path, those are ones people consider boutique parts for. Not much space to work with tho. See also CTH tube & tweak thread for others' experiences.

Won't say much more about this except that space is very tight & be sure any component substitutions fit (& meet to exceed ratings, etc).