RSIIb Outboard Crossover Update

Maxamillion

Super Member
I've received all the parts for the x-overs: Northcreek and Solen Heptalitz inductors, Mundorf Supreme and Solen caps, Cardas copper litz wire, 5-way binding posts and mounting plate, etc.

After much sketching and thinking, I came up with a design for the x-over cabinet that's the same width as the back of the RSIIb's (8 1/4"), 24" high and 19" deep. That width will allow the cabinet to be placed behind the speakers and look like an extension of them - if I can get the veneers to match. In that position, the x-over cabinet will completely cover all the fuses, speaker jacks, and level controls on the back of the speakers, giving a nice neat appearance. Because of the very large size of some of the components, and the fact that there are 10 inductors, 9 capacitors and several resistors per speaker, I decided to place a partition in the cabinet so that half the components could be mounted on one side and the rest on the other side. This should allow for plenty of room between the big inductors and other components to prevent interference.

Below are pictures of the boxes. They are constructed of 1/2" MDF with 1" square hardwood glue blocks in the corners (which double as stops for the side covers), brass wood screws and urethane glue. The first two pictures are of the boxes without covers, showing the partition in the middle, and the third shows one box with the side cover on. I intentionally left a 1" gap on the sides of the partitions for wiring purposes. I included one of the Northcreek inductors in the pictures for scale (4.3mH, 12 ga.).

Next up is to fill the screw holes, plane and sand everything smooth and veneer the boxes in oak to match the speakers as closely as possible, then wire everything up! It's getting there....
 

Attachments

  • RSIIb Outboard Crossover 001.jpg
    RSIIb Outboard Crossover 001.jpg
    29 KB · Views: 316
  • RSIIb Outboard Crossover 002.jpg
    RSIIb Outboard Crossover 002.jpg
    26 KB · Views: 185
  • RSIIb Outboard Crossover 003.jpg
    RSIIb Outboard Crossover 003.jpg
    60.7 KB · Views: 172
Hi Max,

I guess your going to need the room in those boxes :yes:

:scratch2: I have tried to find some pictures of the front and back of your RSIIb's but haven't had any luck if you could post one that would be great.

I did find your schematic in one of your former posts.

http://www.infinity-classics.de/infi...ical_sheet.pdf

Thats a real monster to be sure!

If your project takes off be prepared for a ton of questions from e-mail and posts as I have from my outboard crossover thread it has really been a surprise (10,000 plus views) in the interest of such project.

http://www.audiokarma.org/forums/showthread.php?t=159340

Also what I thought was going to be good for parts and what was good really surprised me.

What I'm saying is something may measure well objectively but may not sound good subjectively! :yes:

This was the case with the Solen Heptalitz inductors for my tweeters. The "Q" factor of the Solen is superior but unfortunately they were way to bright! For my project.

It doesn't mean they won't work for you as its all in the voicing that you and you alone prefer.

Interestingly I see you have a lot of Iron Core inductors, Which are very audible as I have had a number of people comment on my Klipsch Forte's bass being more punchy, fuller and tighter than they had recalled the speaker capable of.

Here is DeanG of the Klipsch Audio forum making a comment on Iron Core vs Air Core.

"I didn't expect to ever come back to this thread again, but have had a surprising experience. I finished my ALKs a few days ago, and as part of the build I went with some 10AWG Solen air cores which I mounted on separate boards because of their size. The DCR of the 10AWG air cores is .16, the same as the Erse Super Q steel laminates I've been using from Parts Express. FWIW, there's a pretty big difference in the bass. After an hour or so of listening I put some Super Q's into the circuit to see what it would sound like, and though still very good, sounded lean in comparison, and not nearly as full bodied and tight as the air cores. I've had the beasts laying around for a long time, and I'm glad I finally found use for them. I didn't expect to hear a difference, wasn't even listening for it, but it was pretty obvious."

And heres the original thread very interesting reading but very technical as well.

http://forums.klipsch.com/forums/t/38976.aspx

I also believe along with DCR the sheer size of the core has something to do with the sound as though there is a greater breath and greater sense of ease. I haven't also had a real need to be concern with DCR in series with a driver after all room response is far more varying IMO. Perhaps 4 ohm drivers are more of an issue.

Heres the North Creek Inductor site.

http://www.northcreekmusic.com/COILS.html

I am also using Duelund Graphite resistors these have a negative temperature coefficient everything else is positive this to is very audible even when being used in parallel with my tweeters to keep a rising impedance. They are the very best I have ever heard I was using North Creeks Ohmites which they favored out of a field of 20 or more brands. But with the Duelund's its not even a contest.

As for your pots once a level is found I personally would replace the pots with two resistors but thats me.

All in all you have a lot of components and from a budget stand point choices have to be made no doubt!

Most of the music is of course in the mid-range so that were I put my best capacitor (Duelund Copper Foil) along with an upgraded Autotransformer. Caps are something else they really are system dependent I was using Hovlands and after the Mundorfs Silver/Oil substitutions the Hovlands were oh so wrong! I was using bypasses and they took forever to break in but even still the Mundorf's out of the box were clearly the better choice.

For those interested in cap comparisons take a look at this site

http://www.humblehomemadehifi.com/Cap.html

Fortunate for me parts count are low and so are my values. But on the other side of the coin the quality of your speakers drivers is of no doubt a higher quality than mine.

Your journey into the Infinity should be very interesting to many :thmbsp:

Cheers

SET12
 
SET,

I am replacing all of the iron core inductors with custom wound North Creek 12 gauge, except the 9.5mH inductor on the parallel leg of the high midrange. It's just too big to replace with an air core - I talked with George of Northcreek and he said that to make a custom air core of that size, with a DCR approaching the iron core (~.15 ohms as measured by me) would cost ~$1000!! I'm crazy, but not THAT crazy! It's not in series with the driver, so I'm hoping its effect will be minimal.

It turns out that by using the 12 gauge Northcreek and 14 gauge Solen inductors I am able to match the DCR of the stock inductors very closely. Since I have level control pots on the UHF tweeter, low tweeter and high midrange I'm not too worried about the Solens being bright since I can always dial them down a bit. I am planning to let everything break in for a while, set the pots to where I like the sound best, then remove them and replace them and all the rest of the resistors (except for the .7 ohm 50W on the woofer) with Duelunds as they are not that expensive (as compared to their caps!). The pots are straight variable resistors, not L-pads, so I do not need two resistors per pot, only one. That will also allow me to combine some of the resistors/pots into single values to reduce the parts count.

As far as caps go, almost all of them will be Mundorf Supremes - according to the Tempoelectric capacitor review, they are second best to the Duelunds, and are MUCH less expensive (I need a lot of them). The 200uF NPE cap on the woofer will be replaced with a Solen PP, as will the 25uF NPE on the low midrange (bypassed with a Mundorf Supreme). Maybe some day in the future I will try Duelunds on my tweets, that's the beauty of outboard x-overs - much more tweakability! The high midrange uses a 23uF cap - it might be a bit too expensive to go with a Duelund there.

I am also replacing the wood screws on the woofers with T-nuts, damping the woofer baskets, adding braces to the cabinet (see attached picture) and replacing all the wiring. Still lots to do...

Here are some pictures of the RSIIb (click on RSIIA(B) on the left side of the screen):

http://www.infinity-classics.de/inf...ries-1+b-2-10+A+B-1981-86/index-Reference.htm

You can see how the cabinet tapers to a width of 8 1/4" in the back, the shape was presumably intended to minimize standing waves inside the cabinet. With my design the outboard crossover cabinet is the same width as the back panel and hopefully will look like an extension of it. It will also cover all the fuses, level controls and connectors, none of which will be used any more.
 

Attachments

  • Infinity RSIIb Rebuild 001.jpg
    Infinity RSIIb Rebuild 001.jpg
    26.9 KB · Views: 166
Just a quick update on this project. The x-over cabinets are now veneered in a nice red oak and are awaiting staining/varnishing (golden oak followed by Wipe-On Polyurethane finish). Once that dries, it's time to install all the x-over components - Yeaahh!

I have gutted the second speaker cabinet and braced it like the first one. I also added squares of self-adhesive Soundcoat between the side braces of both cabinets, on the angled back panels for just a tad more vibration damping. Finally, I sealed all the inside seams with Loctite construction adhesive (the kind in the pressurized can).

I took all my EMIMs apart and discovered that two of them were SmCo EMIMs, while the rest were NdFeB! Also found the NdFeB magnets had their nickel coating peeling off (there is another thread about this ordeal in this forum). Anyhow, I obtained two more complete NdFeB EMIMs (after two tries - don't ask!) and fixed the peeling problem by coating the magnets in polyurethane (see pic below). I'm stalling putting them back together to see if/when Graz comes through with the new diaphragms for us.

That reminds me, time to check in with Graz and see what progress he's making...
 

Attachments

  • RSIIb EMIM Rebuild 007.jpg
    RSIIb EMIM Rebuild 007.jpg
    31.4 KB · Views: 169
Last edited:
RSIIb LF Equalizer Recap

So I've reached a bit of a lull in my RSIIb upgrade, waiting for Graz to come through with the new diaphragms before I reassemble all my EMIMs. Last night, I came up with the idea that I should recap the LF equalizer for the speakers, so I opened it up to have a look (see pictures below). It does not conform to the Infinity Classics schematic, even though my speaker matches their schematic exactly! Curious...

Anyway, I ordered all the parts today, here's what I'm going to replace:

(2) 5000uF, 50V Mallory filter caps (the shiny metal cans) will be replaced with Panasonic TSHA 4700uF, 50V.

(2) 1000uF, 16V caps (the two dark ones in the pictures) will be replaced with Panasonic FM of the same values.

(2) 220uF, 50V Xicon output coupling caps (the blue ones in the pictures) which are bypassed with a 3.3uF PP (yellow ones) and a 0.033uF PS (silver ones) cap, will be replaced by 220uF, 50V Nichicon Muse KZ electrolytics bypassed with 3.3uF Mundorf Supreme PP caps - I'll probably take the old .033 PS cap out.


That ought to take the sound up a notch when the LF equalizer is in line.
 

Attachments

  • RSIIb LF Equalizer Rebuild 002.jpg
    RSIIb LF Equalizer Rebuild 002.jpg
    33.7 KB · Views: 213
  • RSIIb LF Equalizer Rebuild 001.jpg
    RSIIb LF Equalizer Rebuild 001.jpg
    26.4 KB · Views: 194
  • RSIIb LF Equalizer Rebuild 003.jpg
    RSIIb LF Equalizer Rebuild 003.jpg
    92.9 KB · Views: 249
Last edited:
Update on the Crossover Boxes

The boxes have been veneered, stained and varnished (see pictures). They are little lighter in color than the RSIIbs, but I'm counting on the polyurethane varnish (Minwax gloss Wipe-on Poly) ambering over time, so they should get closer and closer as they age.

All in all, they came out pretty nice (my wife thinks they are too big!). They are big, but those Mundorf Supreme caps and Northcreek inductors are huge! They need space...
 

Attachments

  • RSIIb Outboard boxes 003.jpg
    RSIIb Outboard boxes 003.jpg
    26.6 KB · Views: 140
  • RSIIb Outboard boxes 002.jpg
    RSIIb Outboard boxes 002.jpg
    27.2 KB · Views: 145
  • RSIIb Outboard boxes 001.jpg
    RSIIb Outboard boxes 001.jpg
    22.9 KB · Views: 115
  • RSIIb Outboard boxes 004.jpg
    RSIIb Outboard boxes 004.jpg
    28.6 KB · Views: 116
The equalizer recap is done (see pics). I had to drill out the holes on the PCB to a bit larger diameter to fit the leads from the Mundorf Supreme caps - they use very large gauge leads on those.

I tested it out using the preamp/amp jumper circuit on my Marantz 2230 (very handy feature for this sort of thing) and it works as it should.

Does it sound better after the recap? Hard to tell in that system with the Marantz receiver and Pioneer HPM-40 speakers - the true test will be in the main system with the RSIIb's.

P.S.: In case you're wondering why I went with such premium caps in a low frequency equalizer, it's because I do not intend to bi-amp the RSIIb's. Instead I will be using my 500WPC Gilmore Raptor amps full-range, so the full signal will be passing through the equalizer; when you bi-amp you can run the higher frequencies directly from your preamp, bypassing the equalizer box altogether.
 

Attachments

  • RSIIb LF Equalizer Rebuild 004.jpg
    RSIIb LF Equalizer Rebuild 004.jpg
    30 KB · Views: 133
  • RSIIb LF Equalizer Rebuild 005.jpg
    RSIIb LF Equalizer Rebuild 005.jpg
    34 KB · Views: 128
  • RSIIb LF Equalizer Rebuild 006.jpg
    RSIIb LF Equalizer Rebuild 006.jpg
    37.8 KB · Views: 146
Last edited:
Recapped Equalizer Update

I let the newly recapped equalizer play overnight and I just took a listen to a few very familiar songs with and without the equalizer in-line with my Marantz 2230.

I could detect no ill effects on the midrange and treble with the equalizer in-line; it sounded pretty much the same either way, which is great news to me since I could easily tell when the equalizer was in-line before the recap, as the sound became less clear and a bit fuzzier. The Muze KZ/Mundorf Supreme combo seems to be doing a great job letting the music come through!

Overall the system actually sounded a little better than normal with the equalizer in-line because of the nice little mid-bass boost it gave the sound; it was kind of like hitting the loudness button, but with a bit more kick to the boosted bass. The boost rounded out the sound of the HPM-40s very nicely.

I think I'll be much happier now with the equalizer in-line with my RSIIb's; I did not like the effect it had on the mids and treble pre-recapping.
 
Mundorf Supremes

I happened to have two leftover 0.56uF Mundorf Supreme caps from the crossover build (I was going to use them to bypass larger x-over caps but decided against it), and that just happens to be the exact value my Supratek Chenin preamp uses for output caps. As much as I tried to resist (not too much, actually), today I opened up the Supratek and swapped the Mundorfs for the stock Auricaps.

The Mundorfs only have ~ 3 hours on them but they are clearly superior to the Auricaps in this application: better bass depth, better treble extension and clarity, reduced upper-end glare, much more detail - pretty much a slam dunk. If break-in improves them, then I'll be a mighty happy fellow indeed. These caps have an interesting construction, being composed of two caps of double the listed value connected in series in a way that is supposed to be inductance canceling. I don't know if that's the reason, but they sure do sound good.

I can see why these were rated second best (to Duelund $$$) in the Tempoelectric cap shootout! Now that I've heard them in my RSIIb equalizer and my preamp, I can't wait to hear what they do to the RSIIbs!! :yes:
 
Last edited:
Outboard Crossovers are Complete!

I finished up the second crossover today, so now both are complete. I have attached pictures below. The first picture shows the completed box. The second shows the low side of the box - on the bottom is the woofer crossover (<150Hz), on top is the low midrange (150 to 800Hz) and in the middle is the high midrange (800 to 4000Hz).

The third picture shows the high frequency side of the box - on top is the rear tweeter crossover (not sure of the range of that one), in bottom is the low tweeter (4000 to 8000Hz) and in the middle is the ultra-high tweeter (>8000Hz). It looks like a lot of extra room on the high side, but remember that on the other side are the big old inductors, so I needed to keep the sensitive tweeter x-overs away from them.

The fourth picture is a closeup of the wiring at the binding post. I wired all the positive connections directly to the + lead from the 4.3mH woofer inductor, and all the negative connections to the - lead from the 3.5mH low midrange shunt inductor. The binding post holes were drilled out slightly to fit the 12 gauge inductor leads snugly, and those connections were then soldered.

The 200uF woofer shunt cap, 70uF low midrange series cap and 25uF low midrange shunt cap are Solens (the 70uF is bypassed with a .33uF Mundorf Supreme). All the rest of the caps are Mundorf Supremes.

The big amber-colored inductors are Northcreek 12 gauge, while the reddish ones are Solen Heptalitz 14 gauge. I should have plenty of room in the box if I ever want to upgrade to a 10 gauge woofer inductor.

All internal crossover wiring is done with Cardas copper litz wire - 11.5 gauge on the low side, 15.5 gauge on the high side.

The fifth picture shows the binding posts and plate I used - they are the Dayton Premium binding posts and aluminum plate, both from Parts Express.

I re-used one laminated iron core inductor (on the midrange shunt circuit) because it was too big to get in a large gauge air core (9.5mH). I also re-used the stock cement resistor for the woofer shunt circuit (0.7ohm, 50W).

Everything is held down with Amazing Goop RTV adhesive - the heavier components are also zip-tied as well.

The stock resistors and pots from the high midrange, rear tweeter, low tweeter and high tweeter circuits are in there for now, as I intend to let everything break in for a while and adjust the pots until I get everything just the way I like it. Then I will remove the pots and measure their values, and replace them and the resistors associated with them with Duelund resistors of the same total value.

So, what do you think?

I'm holding off putting everything back together since I figure I might as well wait for the new Graz diaphragms - let's see how strong my willpower really is!
 

Attachments

  • RSIIb Crossover 002.jpg
    RSIIb Crossover 002.jpg
    27.1 KB · Views: 123
  • RSIIb Crossover 005.jpg
    RSIIb Crossover 005.jpg
    27.8 KB · Views: 147
  • RSIIb Crossover 003.jpg
    RSIIb Crossover 003.jpg
    30.3 KB · Views: 130
  • RSIIb Crossover 004.jpg
    RSIIb Crossover 004.jpg
    31.4 KB · Views: 202
  • RSIIb Crossover 001.jpg
    RSIIb Crossover 001.jpg
    28.6 KB · Views: 106
Last edited:
So, what do you think?

I'm holding off putting everything back together since I figure I might as well wait for the new Graz diaphragms - let's see how strong my willpower really is!

I think you did a fantastic job. Beautiful work. I also don't think there's any chance you're going to hold out until the Graz diaphragms get here - you're just gonna have to try these babies out.

David
 
In case anyone is wondering about the effect of substituting the 12 gauge air core inductor for the stock laminated steel core inductor, AK member jwalker kindly ran a simulation for me on a very expensive linear simulator program, yielding the graph below. The green trace is the stock inductor and the red trace is the 12 gauge air core inductor.

As you can see, the air core lowers the response by about 0.7dB at the response peak and about 0.3dB at 20Hz, while shifting the crossover point slightly lower. Will this be audible? I don't know, but most say that any change less than 1dB is inaudible to the average person. I suppose I could go to a 10 or 8 gauge inductor to bring it back up to the stock level, but a little tweak on the LF equalizer bass gain and contour knobs should do that just fine. Unless of course I like the changes better - time will tell.

Thanks again to jwalker for the help!!
 

Attachments

  • Inductor Simulation.jpg
    Inductor Simulation.jpg
    2.4 KB · Views: 458
Last edited:
Further Equalizer Update

Idle hands are the devil's workshop...

Now that the crossover is complete, I decided to have another look at the equalizer. Even though it is a low frequency equalizer, its effects range all the way up to 1000Hz, so it's worth it to get it as clean and clear as possible, hence my re-cap above.

It uses a pair of socketed LF353N opamps, which are early 1970's-designed dual-channel units that are very inexpensive and are generally not considered very good for audio (if you do some reading of the solid-state DIY boards). Looking at the alternatives, the Burr-Brown OPA2134 looks like an excellent replacement, and is very highly regarded in the same DIY circles. The specs of the 2134 are miles better than the LF353N, for example THD is 0.00008% vs 0.02%! Best of all, the pinouts are identical and the opamp is socketed, so it should be a simple drop-in replacement, no soldering needed.

I've ordered a pair of the OPA2134 in the 8-pin DIP format from Digikey ($4.86 for the pair). Stay tuned, they should be here by the end of the week.
 
Last edited:
I received the Burr-Brown OPA2134 opamps tonight and hooked the equalizer up to my shop system to see if I could hear a difference. I went back and forth three times between the stock and Burr-Brown opamps with both equalizer knobs cranked to the max. I couldn't hear any difference in the low bass, but the HPM-40s I was using don't go down as low as the RSIIbs do.
I thought I could hear a little bit of a difference in the mids, with the BB opamps sounding just a little bit fuller than the stock, with a little bit more detail (maybe 10% maximum). Not a huge difference, but something I think I could pick out. Of course in the real world, the knobs won't be set at 100%, so maybe there won't be an audible difference then. Was it worth $5 - yes, if only for the peace of mind that I upgraded that part and can now move on.

I also put 4.7uF Solen PP bypass caps on the two 4700uF power supply caps in the equalizer; hey I had them in my spares box, so I might as well use them! :yes:
 
I received the Burr-Brown OPA2134 opamps tonight and hooked the equalizer up to my shop system to see if I could hear a difference. I went back and forth three times between the stock and Burr-Brown opamps with both equalizer knobs cranked to the max. I couldn't hear any difference in the low bass, but the HPM-40s I was using don't go down as low as the RSIIbs do.
I thought I could hear a little bit of a difference in the mids, with the BB opamps sounding just a little bit fuller than the stock, with a little bit more detail (maybe 10% maximum). Not a huge difference, but something I think I could pick out. Of course in the real world, the knobs won't be set at 100%, so maybe there won't be an audible difference then. Was it worth $5 - yes, if only for the peace of mind that I upgraded that part and can now move on.

I also put 4.7uF Solen PP bypass caps on the two 4700uF power supply caps in the equalizer; hey I had them in my spares box, so I might as well use them! :yes:

Don't get to distracted by differences caused by good quality opamps. The room placement of those speakers will be a much more critical factor and play a much bigger role in the sound quality. The equalizer is needed to compensate for a falling bass in the RS-2b's so you probably can't do without it.

Also your CD player most likely uses opamps in the current to voltage converter on your DAC's. I wouldn't lose any sleep over it ;)
 
There are no opamps in the CD player output. The Zero One Mercury CD/HD player I'm using has completely discrete output components - no opamps. Using HD playback it sounds better than any CD player I have ever heard. See link.

http://www.zerooneaudio.com/

I agree that room placement will have a larger effect than the opamps, but to me it was a no-brainer to replace the old opamps with something designed for audio, especially when it only cost $5 to do it.
 
There are no opamps in the CD player output. The Zero One Mercury CD/HD player I'm using has completely discrete output components - no opamps. Using HD playback it sounds better than any CD player I have ever heard. See link.

http://www.zerooneaudio.com/

I found my Dell Inspiron 6400 notebook PC with SigmaTel High definition audio Codec, 24 bit 96Khz DAC to sound better than any CD player I have ever heard ;)

Why, because Windows Media player has some very nice DSP effects that are not normally available in any standalone CD player. Makes old CD recordings come to life whereas my new Pioneer D6 CD/SACD player with the latest 24 bit 192KHz Burr Brown DACS sounds lackluster ;)
 
More Equalizer Stuff

I figured that since two of the three pairs of electrolytic caps were bypassed with PP, why not the last pair. So I bypassed the two 1000uF Panasonic FM caps with 1uF Solen PP caps that were in my spares box. There just so happened to be two holes in the board already predrilled through the traces next to each of the FMs, so I used them to install the bypasses - a snap!

Below are two pictures of the power supply bypasses. The first shows the 4.7uF bypasses on the 4700uF caps, and the second shows the 1uF bypasses on the 1000uF caps. With the two sets of power supply bypasses in place, I can say that the sound is a bit snappier, especially on transients. I think I'm going to repeat the opamp faceoff now that the bypasses are in place - perhaps I'll hear more differences.

I also ordered one more set of opamps to compare, Burr-Brown OPA2107, and they should arrive in a couple of days. So I'll be comparing the LF353N ($0.40 each), the OPA2134 ($2.45 each) and the OPA2107 ($12.25 each).

From what I've read the 2107 is supposed to have better definition in the bass and mid-bass than the 2134 - we'll see soon enough. :yes:

P.S. 06/22/09: I also ordered a full set of PRP metal film resistors (plus 2 Shinkohs) to replace the stock carbon film resistors, at double the stock wattage ratings. That ought to bring the background noise level down! So once I install the 30+ new resistors I'll go ahead with the opamp showdown - should be very interesting...
 

Attachments

  • RSIIb Crossover 007.jpg
    RSIIb Crossover 007.jpg
    26.3 KB · Views: 96
  • RSIIb Crossover 008.jpg
    RSIIb Crossover 008.jpg
    29.8 KB · Views: 89
Last edited:
Back
Top Bottom