BTW- Michael Percy will match his products to 1%(or closer) for you, which will keep your crossover points equal. Most high-end caps are manufactured with a(typically) 5% tolerance(or better) to start with. (http://www.percyaudio.com/Catalog.pdf) |
Wavetrader
don't forget that your option is not just 2x8 or a single 16.
A 12mfd and a 4mfd would also work.
For that matter, 16x1mfd gets you there, too, though space probably won't allow this. Anything that adds up to 16.
My Magnepan X-over uses 4 caps in the tweeter totaling 22mfd.
The smallest is a 0.1mfd bypass.
If/when I rebuild them, I may skip the bypass and go directly to the value, the combination depending on whose caps I choose.
One other minor point. Caps have a build tolerance. A 16mfd cap of 20% could be anywhere from 12.8 to 19.2mfd.
In the class of component you are thinking about, 5% is more common while I have seen 3%.
It is probably also possible to buy 'hand selected' parts of measured values. |
For loudspeaker use, my opinions mirror Undertow's. |
Undertow, yes, thanks for first recommending Claritycap MR & for the tour of your modded Jolida phono stage. |
Dgarretson,
I think you emailed me on the Clarity MR's once?
Anyway if your interested as I see you have been doing some tweaking I have yet found another application that the MR's did not fail.. They are actually perfect to be honest :-)
See link
http://www.audiocircle.com/circles/index.php?topic=62730.0 |
"Remember that caps used in passive crossover see high voltages and therefore they change shape along with the signal flowing through them - this can add significant distortion in some cases"
Both Claritycap MR and Duelund reportedly have design improvements that minimize this physical movement of the capacitor plates(resonance). Clarity has a white paper in which they claim that in extensive listening tests, low capacitor resonance was a greater factor than capacitor material, in determining listeners' perceptions of a capacitor's quality.
Bob, when I rebuilt my Merlin crossovers I found that adding a .015uf russian FT-1 teflon bypass cap in parallel with the stock midrange/woofer main cap made a small improvement in speed & precision of the driver. When the cap is fully broken in, I will remove it and listen carefully for the mid-range bump you describe. |
Precisely matched smaller capacitors in parallel may work better than an equivalent large capacitor. A lot depends on geometry as well as the choice of dielectric. Remember that caps used in passive crossover see high voltages and therefore they change shape along with the signal flowing through them - this can add significant distortion in some cases and can affect dynamics. A better solution would be to go with active crossovers (using low level line signals) and consequently reduce this form of distortion and power loss substantially. |
I suggest bypass on the tweeter from practical experience. I tried them on my mid/bass crossover caps and found that there was a pronounced forwardness and over prominence on the upper midrange that threw the whole speaker out of balance.
Now, YMMV, but it makes sense to me electrically also. Bypasses enhance upper end frequencies, and so make a difference at the upper end of the frequency spectrum. For a mid/bass driver this could be over emphasizing the upper crossover point and causing anomalies there between it and the lower end of the tweeter. Bypasses on a tweeter affects just the upper end extension and air, without any other driver interaction. At least that's the theory I'm working with to explain what I encountered.
As in all things, I recommend trying it for yourself in your system, and learning from the experience.
Enjoy,
Bob |
Lets back up.. If in fact they are 2 8 uF caps equalling 16 uF in a parallel design, yes you could go to one cap.. however the reason they were probably in there was that whatever that manufacture is they used might have been cheaper to use 2 of the 8's... opposed to a 16.. or simply that manufacture does not even make a 16...
However if you can get a very good 16 uF that is guaranteed better quality than the current one in there so be it and go for it. Or you can go with 8 uF's again due to its more rare to find exotic caps in exact large sizes, for example you might find that mundorf might only make a 12 uF and an 18 uF... So you can't even get the 16 uF unless they offer you to make a pair custom which will normally cost you a lot more... Thats why then you can choose to keep the dual 8 uF's.. which is called "Battery together" type config... or parallel.
Anyway technically you do have a little higher voltage capability with the dual 8 uF's over a single 16 but that has no real application in a speaker accept that some manufactures might argue that the 2 caps are still better with 2 small reserves and faster than one big cap.
Finally the best caps are for the cheap side Jantzen superior...
Mundorfs are in the middle, but I feel in most applications the Jantzens have better bass and just as good for less money..
And on the High side for the better pricing is the Clarity Cap MR's which none of them equal in sound, or you have to go into very costly and much time to get(so have patience) Duelund capacitors. |
Hi Bob, I'm curious why you recommend limiting small bypass cap to tweeter. |
I would also suggest bypassing the caps in the tweeter circuit, each with a smaller .1uF cap. Don't do this in the mid or woofer sections though, just the tweeter.
Enjoy,
Bob |
Keep in mind: If you replace the caps, the higher the quality, probably- the longer the burn-in. They may sound absolutely raunchy at first(been there). Polypropylene/polystyrene may take 200hrs and Teflon caps 400+ before they reach their target performance levels. |
Rodman,
Yes the 2 8's are in parallel 200 vdc . That's what I was hoping for.
Thank you |
As long as you are certain the two 8's are paralleled, and you replace them with the same voltage rating in a 16uF: You should experience no problem. The value/replacement will be exact. |
The safest path is to do a one to one replacement with original values of better quality. All audio designs are complex compromises and it is impossible to know when a change will effect the sound. I have read several postings in which the substution of unquestionably better parts of different values degraded the sound and the modifier eventually went back to the original values. |
