Hi, everyone!
Well, for some time, I have been thinking about how to refurb the crossover in the 2.5s. After all, those huge electrolytic caps are approaching 30 years old.
Of course, you can't buy an 1100uF bipolar 100 volt cap anywhere at any price (or a 700uF for that matter!).
So, the solution is to make your own. But... the best way to do it.... hmmm.
I finally decided to parallel 11 100uF caps to do the job. I chose the Nichicon VP series for a couple of reasons. One was that they had numbers I liked. The second was cost. In these quantities, they cost less than a buck each. There were other contenders, but the only others that came close in area like ripple current were four times the price. These come in around 450 mils a piece, so multiplied by 11, we get almost five amps of current. That should do it!
I mulled over geometry and construction technique, and finally decided on the configuration in the photos.
The parts are all hot-melt glued together, and what I did was take the old capacitor clamps, invert the tabs and Dremel off the sharp parts. The modified clamp is also hot-melted to the inside of the capacitor ring. OFC copper and silver solder topped off the construction. It turns out that the cap mounting rings for the old 700uF capacitor works for the inside of the 1100uF replacement. It goes on the inside. You can use the ring for the 1100uF capacitor as the ring for the 700uF replacement. In this case, the parts go on the inside.
These can just be screwed down in place of the old parts, and you can solder anywhere on the two conductive rings.
I will put up additional photos up when the parts are installed. And, after they have broken in for a week or so, I'll report back with any results.
I did the 125uF parts using 12 10uF VP parts, and they are constructed differently. 120uF raises the crossover frequency of the EMIMs by a smidge, but I think the EMIMs are crossed over too low in the 2.5's anyway. Adding a high-quality film bypass can take the value right back to 125uF.
Now, this begs the question: Suppose I designed a PCB for this and had it fabricated. Would anyone else want one in order to make this project a lot easier? We could probably get decent pricing on a run of 25 from a proto board house.
Could also use the board to make any other value (like the 700uF), just by not populating everything. Or perhaps I could come up with a slick design to make 1100 or 700 and look nice, too. I'm even thinking about a combination board that can be snapped apart, with the 120uF assembly included. We'd have to see what's cheaper- two boards or one trick board.
The VP series can be had from Mouser, BTW.
Lemme know what ya'll think.
Well, for some time, I have been thinking about how to refurb the crossover in the 2.5s. After all, those huge electrolytic caps are approaching 30 years old.
Of course, you can't buy an 1100uF bipolar 100 volt cap anywhere at any price (or a 700uF for that matter!).
So, the solution is to make your own. But... the best way to do it.... hmmm.
I finally decided to parallel 11 100uF caps to do the job. I chose the Nichicon VP series for a couple of reasons. One was that they had numbers I liked. The second was cost. In these quantities, they cost less than a buck each. There were other contenders, but the only others that came close in area like ripple current were four times the price. These come in around 450 mils a piece, so multiplied by 11, we get almost five amps of current. That should do it!
I mulled over geometry and construction technique, and finally decided on the configuration in the photos.
The parts are all hot-melt glued together, and what I did was take the old capacitor clamps, invert the tabs and Dremel off the sharp parts. The modified clamp is also hot-melted to the inside of the capacitor ring. OFC copper and silver solder topped off the construction. It turns out that the cap mounting rings for the old 700uF capacitor works for the inside of the 1100uF replacement. It goes on the inside. You can use the ring for the 1100uF capacitor as the ring for the 700uF replacement. In this case, the parts go on the inside.
These can just be screwed down in place of the old parts, and you can solder anywhere on the two conductive rings.
I will put up additional photos up when the parts are installed. And, after they have broken in for a week or so, I'll report back with any results.
I did the 125uF parts using 12 10uF VP parts, and they are constructed differently. 120uF raises the crossover frequency of the EMIMs by a smidge, but I think the EMIMs are crossed over too low in the 2.5's anyway. Adding a high-quality film bypass can take the value right back to 125uF.
Now, this begs the question: Suppose I designed a PCB for this and had it fabricated. Would anyone else want one in order to make this project a lot easier? We could probably get decent pricing on a run of 25 from a proto board house.
Could also use the board to make any other value (like the 700uF), just by not populating everything. Or perhaps I could come up with a slick design to make 1100 or 700 and look nice, too. I'm even thinking about a combination board that can be snapped apart, with the 120uF assembly included. We'd have to see what's cheaper- two boards or one trick board.
The VP series can be had from Mouser, BTW.
Lemme know what ya'll think.
Last edited:
