|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
73.226.116.199
In the audio path circuit there are two 47uf caps. One were the signal comes in and the other when it leaves.
This is only temporary but, what should I expect if I increased the cap values from 47uf to 100uf? Sonically speaker.
Again, these are in the audio path caps. Not power caps.
thanks
charles
Follow Ups:
Assuming both caps are equivalent construction and material quality DA, esr, esl seems to go up with package/capacitance value size in a given voltage, temp range and tolerance, though I'm sure there are probably plenty of exceptions to that rule. I'm thinking the differences due to the above will be mostly negligible.However.
This configuration is a passive single pole high pass filter. It may indeed have a real affect on system performance depending on your situation.
There will be about a fifth of a decade difference between the rolloff frequencies, at equivalent loading, with the 100uF cap obviously having the lower rolloff frequency. The rolloff frequency heavily depends on loading. More loading give higher 3dB frequency.
If your load is a high impedance pre amp input for example the differences in the audible band are very negligible however if your load happens to be something like 8 ohms speakers you'll ruin the sound entirely....no more bass.
To be fair the 47uF ca; would already be ruining the bass if that's what you are doing so I'll guess your load is higher impedance?/?
Edits: 01/19/16
Hello,
Standard electrolytics are not great for this application. Changing the cap to a bi-polar electrolytic will cut the distortion to about 25% as much. Search for articles by Cycil Bates if want to get gory details, believe it was articles 4 & 5 that address electrolytics. Believe they are still posted on Linear Audio magazine website.
As audio signals are AC, using a bi-polar helps. Also if size can be fit in a larger uf cap will give better low end response. 47uf is large enough that low end isn't being effected much with usual in and output impedances. (10k ohms and up)
My rule of thumb for rapid estimation of low frequency cut-off, without doing the math, is that a 1.0uF capacitor will give you an ~2Hz LF cut-off (-3db point) with a 100K parallel resistance. This is sufficiently low to prevent phase shift at 20Hz. Therefore, 10uF would be more than adequate for a 10K parallel resistance. That's why I wondered (see below) why this unit would need 47uF in a coupling topology.
And yet we usually find that the bigger the capacitor (keeping capacitor type unchanged) the better it sounds, until ultimately it is removed and sounds much better still (provided offsets are handled). So I'm not sure it is so much about phase shifts, maybe more about low frequency noise in the playback making it sound more realistic.
However, in coupling the output of a phono stage to a linestage, one must be aware of rumble caused by LP warps and off-centeredness of the spindle hole. The spurious ultra-low frequencies thus generated can wreak havoc on the amplifier and speaker, sucking up lots of power for nothing, at best. Some older circuits used a hi-pass filter set at about 10Hz or 20Hz, for that reason for phono. I'm sure you know this; just saying... The custom of using such a filter seems to have disappeared in commercial products, like the "mono" switch and other features that used to be common in full-function preamplifiers.
Why do you want to do this? What are you trying to accoumplish? Is there a problem with it as is?
Could you be more specific? Are the caps in an amp or pre-amp, or in a loudspeaker crossover?
The universe is made of electrons, protons, neutrons, and morons.
This is a preamp section of a NAD C326BEE
I'm trying to improve the sound with a better grade of cap in the preamp section. So far replacing the original cap with a Nichicon KW series 100uf (original 47uf) cap there most definitely a sonic improvement. The original thin midrange is now fuller and the sound has more of a punch and bounce to it.
But I am concerned I may be rolling off a FR point or causing other effects 'sonically' that I am not aware of.
charles
I think there is no such thing as Nichicon KW. There is an FW and a KZ, but I'm not aware of KW.
If you have KZ, that's probably the best 'lytic cap readily available. Others to consider are Elna Silmic (which I find to be somewhat dull on top) and Panasonic FM (which can be a bit hard and cold). Some people really like the Nichicon non-polarized, ES. Jensen also makes 'lytic caps. Everyone has their favorite. Cost is hardly a factor for that size, so experiment. Electrolytic caps do require some break-in time, usually a few dozen hours.
The uF value of the cap is what determines the lowest frequency that will pass, and has very little effect on highs. In your preamp, going from 47 to 100 probably has no effect at all on what you hear, but the brand of cap does. A higher voltage cap may pass high frequencies somewhat better. If you are really concerned about high frequency roll-off, you could experiment with small value film bypass, but I have never found a suitable combination of 'lytic and film.
Peace,
Tom E
Googling. Easy...
I know nothing about this particular piece of gear, but 47uF seems like a gigantic value for a coupling capacitor in the circuit of a piece of solid state gear. So, I am wondering what that capacitor is really doing in the circuit. Perhaps it is a decoupler, to isolate the power supply to a gain stage from other stages in the circuit. The Klyne preamplifier that I own uses electroytics of similar value for that purpose.
Thanks for that. I have not heard of, or heard, the KW. Nichicon makes a confusing array of caps, and I can't be familiar with all types. I wonder how much difference there really is.
My advice stands: KZ is among the better available. That one I have heard, and it's very nicely balanced between detail and fullness, well worth the modest increase in price from lower grades.
Peace,
Tom E
I was pretty sure I'd heard of Nichicon KW before, so I did a search. Then I remembered that I actually have two sitting here, waiting to be installed in the filament supply of my tube preamp.
One of the two caps I had in there started leaking. I replaced both of them with a couple of caps I had on hand, but they were only 3,300 µF, and I've found the sound improves with more capacitance. I bought two 22,000 µF Nichicon KW for that position.
By the way, thank you for your take on the KZ and Elna Silmac. The original poster was asking about signal path, but have you done any comparisons of these (or other) electrolytics in power supply use?
I'm not sure Silmic's are available in PS sizes. Local decoupling, yes, which is where I use them sporadically in my amps. For main power reservoirs, I stick pretty much with Panasonic TSHA. They are rated at 105C and have low ESR and sound just fine. I have used the Mundorf AG 'lytic caps in a power amp, and, while the bass seems a bit more energetic and the mids a bit more tangible, I'm not sure they're worth the substantial premium. For ultimate fidelity or the financially privileged, perhaps, but not for the working class budget.
I don't believe that any capacitor is the best at everything, which is why I intentionally mix them up (and resistors) within a single component. Of course, channels are matched, but at different places within a channel I employ different brands and types. Inside my power amps there are Panny TSHA, Elna Silmic, Panny FM, and BG Std. In another pair of amps I use Nichicon KZ's instead of BG's, and they're nearly as good. Resistors: Vishay naked Z-foil, various Caddocks, Shinkoh tants, Susumu SMT, PRP, Vishay metal film. Film caps: Jupiter copper foil/beeswax paper, Amtrans film/foil, ERO styrene/foil, WIMA FKP, MKP, silver/mica, etc.
Peace,
Tom E
The problem with implementing ELNA Silmic caps in tube gear is more their voltage rating than their size in uF. I think the upper limit is 50V, but perhaps some lower uF values are available at 100V rating, tops. This is a problem, of course, for tube PSs but not usually for solid state PSs. In uF value, they do go up to at least 220uF/50V, which is quite handy in a PS. I've never searched for higher value ones, but maybe...
Edits: 01/14/16
220 µF, 100 Volt is the highest value avalable, according to the Elna® RFS (Silmic II) spec sheet (link below)
"Suddenly, I'm not half the man I used to be. 'Cause now I'm an amputee" J. Lennon
Edits: 06/30/18
That all depends on the power supply. I recently built the Pete Millett LR phono circuit, which uses a +/- 15V bi-polar power source. I'm currently powering it with a supply I already had, while I make a decision on the power supply. If I go with the PS board I bought from Pete, the Silmic's come in the correct value, 1,000 µF, 25V, but they'd physically never fit his board.
I do have room to squeeze in two Silmics at one end of the board, after the output of the regulators. After doing my homework on the regulators Pete calls for, LM2991T and LM2941T, the spec sheet calls for not using too low an ESR cap on the output. For that reason alone, the Silmics will be a better choice than the Nichicon HE caps I'd use before the regulators. It will also follow your "mix them up" advice.
I agree with you about the Panasonic TSHA. I've also used the TSHB. They both measure very well on my ESR meter, are reliable, and reasonably priced.
Thanks for the feedback on various components you've had success with. Much appreciated.
Below is a link to a comparison of the "sound" of various electrolytic caps in the power supply of a SS Marantz 2220B receiver. It's only one man's opinion, but he likes the Silmics over Panasonic FC, Elna Cerafine, and Nichicon KZ.
Here's another shoot-out , where the tester liked the Silmic even over the legendary Black Gate caps for power supply use. These posts, along with the Nelson Pass comment I linked to in another post in this thread were leaning me toward the Silmics. But, as always, there's no substitute for personal experience.
Thanks again.
I have not used these in the PS. But I have used them in the audio path.
Generally I found the Elna Silmac to be soft on top to even a little rolled off. High praise aside I didn't care for it.
But it does have a great tube like sound - otherwise.
I appreciate the feedback on the Elna Silmic caps, although Nelson Pass, for one, would disagree with you. Or he did, in 2009, anyway (link below). I have read similar takes to yours before, that their top is soft.
I have an old, James Bongiorno designed SAE power amp, which I've owned since 1976. Given its bright sound, Silmic caps in the signal path seem like they might be a good choice.
Thanks again for your take on the Silmics.
A cap acts as a high pass filter and f=1/(2*pi*R*C). So a larger cap lowers the cutoff point of the filter.
If the cap is in the audio path then I would think a film cap would be a better choice than another electrolytic.
Edits: 01/05/16
47uf is going to be a physically hard cap to stick in there.
So I would like to stick with the electrolytics.
but again how much is the difference effecting the sound or rolloff?
If you increase the value of the cap it will allow more low frequencies to pass through the circuit. So if a piano sounds somewhat this with the current cap, increasing the cap value will make for a fuller sound by passing more low frequencies.
The universe is made of electrons, protons, neutrons, and morons.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: