Source for 4412 crossover capacitors

[ted01]

I'm having a heck of a time finding caps for the 4412 xover network. Looking for 40 uf, 17 uf and 14 uf. Any specific vendors/brand names?

I've checked the usual suspects; Digikey, Newark, Mouser, Parts Express... I can find maybe one value listed but not finding all values, particularly not within the same product line.

Any help would be appreciated.

ted

[Robh3606]

Why??

Caps add in parallel so you can use 4 x 10uf or 2 x 20uf as examples to get the 40uf.


Rob:)

[ted01]

I understand that. Here's what I'm wondering... If caps are additive in parrallel, why is there a .01 mfd cap bypassing the larger value caps? It's there as a high freq shunt, right? So, if that's the case, then it seems reasonable that two 20 mfd caps will be seen differently than one 40 mfd cap, in terms of freq. cutoff and assuming a complex waveform. Is there any truth to this? I'm just guessing, not saying I know anything. I also know that the caps JBL used are at least 10% tolerance. So, it seems reasonable that I could use a 16 mfd and assume that the variation (in terms of tolerance) would bring me "close enough". However, since caps are available that feature as low as 1% tolerance, and the JBL engineers that designed the network specified a value of 17 mfd, doesn't it make sense to use not only the correct value, but also the closest tolerance?

If I'm completely off in left field on this, please don't hesitate to correct me...

Thanks,
Ted


P.S. I'm not quite good enough to figure out the math; for some of you, this may seem a trivial question, but just how much difference in crossover freq is there if I use nearby values? For example, the schematic calls for a 3 mfd cap in the section of the xover that feeds the high freq. driver. If that were a 2.2 mfd cap, how much would the rolloff point shift?

Thanks a second time!

[Robh3606]

Hello Ted

I just got back to this tonight. The bottom line is to not over think this. If you want a 40Uf you can use 2 20u's and 4 10u's. You are not going to have any issue doing this.

The bypass caps do make a difference but they are seperate from the basic value used. They are also small enough in value that they do not effect the actual crossover points.

What I do is go exactly on value within 5% of tolerance which is very easy to do with film capacitors like Solens as an example.



P.S. I'm not quite good enough to figure out the math; for some of you, this may seem a trivial question, but just how much difference in crossover freq is there if I use nearby values? For example, the schematic calls for a 3 mfd cap in the section of the xover that feeds the high freq. driver. If that were a 2.2 mfd cap, how much would the rolloff point shift?


There is no simple answer to that one. It depends on many things that interact within the crossover and the drivers Impeadence curve. If you are cloning a crossover unless you have the measurement and design tools available to you don't start changing things. Just go on faith they got it right on your first time out.

Rob:)