Modifying the Odyssey crossover - my adventure

MartinLogan Audio Owners Forum

Help Support MartinLogan Audio Owners Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

spectral

Well-known member
Joined
Dec 27, 2007
Messages
327
Reaction score
1
Location
NE USA
So I've been unhappy with the Summit X for a variety of reasons, I can't fit the CLX in my room, and I really like the O's... which led me here, exploring the possibility of upgrading the O's crossover, of this very worthy speaker. This is a continuation of this thread... the comments from other fellow mod'ers have been extremely encouraging.

First off, the O's have rewarded me for over 9 years now, and have been the reason I spent so much money on the rest of the electronics. They do have their own limitations, which I am hoping to address with this mod. But they are sensational to boot. When did I first started thinking about this? Well, 9 years ago, when I asked the dealer - so what's not so high end about this speaker? Answer: the crossover.

Attached to this first post is the crossover schematic; comments and pictures to follow...

View attachment martin-logan odyssey crossover schematic.pdf


Edit Dec 6, 2011: Attaching the woofers' specifications
View attachment Odyssey-8-inch-driver.doc,View attachment Odyssey-10-inch-driver.doc
 
Last edited:
In this second post, I'll describe what I saw in the schematic... first off, two crossovers - one for the woofers, the other for the panel. Let's take the woofer xover first. You will notice there are two resistors in parallel (RX13 & RX14) giving a 1.5ohm impedance to attenuate the output of the woofers - this is actually related to the -3dB switch in the back of the speakers, and the switch is right above them in the diagram. If you look at the description of my system, years ago I plugged in a 50W Caddock resistor in series with the woofers to attenuate bass output even more, and have settled for 4 ohms for my room (the switch is at 0dB - bypass).

Replacing the capacitors requires a lot of research... I posted before three main sites I have found useful, and here they are again... audiocircle, humblehifi and laventure. A lot of the mod'ers on the net play with capacitors in their tube circuits, so some of the comments they make relate to how capacitors sound in various pieces of electronics. This doesn't help when it comes to crossovers.

The last link is the most helpful of all, though still subjective; but at least, they've tried the capacitors under test in various applications, from power supplies, to active amplification circuits and crossovers. This is a nice start. One thing stands out - you can't go wrong with Mundorfs and Duelunds; well, this is something I am already very familiar with, and these are the choices of most true high end manufacturers - and then this web site put the stamp on it; this is all I needed to get me going...
 
Part three, and it's time to look for replacement capacitors... partsconnexion.com and madisound have them all (madisound is a little cheaper). So then I read up on Mundorf's web site - the MXT's are to be used as the parallel-to-the-woofer capacitor, while the MXP's for midranges - notice those two 100uF and 170uF caps in the woofer xover that run parallel to the woofers.... this would be the first testing ground for me.

To make a long story short, MXT is out of the picture, and manufacturers appear to use MXPs for their woofers, so I order the 100uF for the rear woofer. Bingo... tests with warble tones prove that the woofer that has the mod sound cleaner than the corresponding un-mod'ed one in the other speaker. We are off to a good start...

But how do you get to the crossover? Well, just detach the rear panel carefully, and the panel's crossover is looking at you; the woofers' is on the opposite side and you need to take off the front woofer to access it; but once you do it's easy to work with that crossover.

And what do I see? Those woofer capacitors are electrolytics. Really? REALLY? And I paid so much money for these speakers? Clearly, a product made to a price point, and the crossover designed to just fit nicely in the enclosure... but not high end.

So next come the first pictures...
 
First off, the replacement of the 100uF electrolytic - the Mundorf MKP 100uF/400V (+-3 tolerance, better than the 250V version):

IMG_4271-2.jpg

Next to it, you see four Mundorf M-Resist Supremes; they are 1.5ohms each, arranged to give me two 3-ohm resistors in parallel, i.e. what the schematic shows.

Next, here's the capacitor installed to the right (notice the empty space where the electrolytic used to sit, and where the other 170uF electrolytic still is):

IMG_4261-2.jpg

I LOVE point-to-point wiring
 
Next up, the panel's crossover; to the right, I have already installed those resistors you saw above - not perfect for sure. Notice how ML keeps their resistors elevated to dissipate heat - nice! Theirs are 5% tolerance, mine 2% - no real difference. And they are both non-inductive types.

Mostly Solens and one horrible Bennic - yikes!

IMG_4263-2.jpg

IMG_4269-2.jpg

Solens are good in power supplies, not crossovers. They are probably used here because of their high values and small size.
 
first off, the replacement of the 100uf electrolytic - the mundorf mkp 100uf/400v (+-3 tolerance, better than the 250v version):

View attachment 14529

next to it, you see four mundorf m-resist supremes; they are 1.5ohms each, arranged to give me two 3-ohm resistors in parallel, i.e. What the schematic shows.

Next, here's the capacitor installed to the right (notice the empty space where the electrolytic used to sit, and where the other 170uf electrolytic still is):

View attachment 14530

i love point-to-point wiring

good thread
 
Last edited:
Selecting the capacitors

So here's the conundrum... what do I buy? Mundorf MKP for the woofers is enough - they run parallel after all. The panel is more interesting... The uber-performing Duelunds VSF's cost something like $250-$500 each for the values we want. The Mundorf Supremes have more values to choose from, while the Mundorf Silver-in-Oil don't go very high in value (10uF max), they cost 4X the Supremes and are huge due to the very high voltage rating (impossible to fit in the cabinet).

So for the high values we are after, the Silver-In-Oil's are cost prohibitive, for my wallet at least. Then I keep reading that they aren't really that much better than the "basic" Supremes, and a lot of people seem to think they tip up the treble (I don't want that either) - I really want to keep the excellent tonal balance of the Odysseys. I am thinking they'd be overkill for the O's and just plain wrong, at least in my all-Spectral system.

So I have settled for the Mundorf Supremes with the Vishay 1837 or 1839 bypass capacitor - around $50 each for the values we want - for the panels. They are being shipped this week, to be installed over the Columbus weekend, so more to come later...

But right now, a word on selecting the right replacement values - not all caps are available as a single piece - the 10uF 100uF are the exception. So here's the breakdown right now:

  1. 100uF woofer electrolytic -> 100uF/400V Mundorf MXP
  2. 170uF woofer electrolytic - 3x 56uF/400V Mundorf MXP
  3. 10uF panel -> 10uF/800V Mundorf M-Cap Supreme
  4. 13uF panel -> 2x 6.8uF Mundorf M-Cap Supreme
  5. 47uF panel -> 2x 22uF + 3x 1uF Mundorf M-Cap Supreme
  6. 51uF panel -> 5x 10uF + 1uF Mundorf M-Cap Supreme

Only if the Supremes don't produce the improvement I am hoping for will I consider the uber-expensive Silver-In-Oil (and I'll need a lot of those). Finally, the VISHAY-RODERSTEIN MKP-1837 (or the axial version 1839) bypass cap will be used on a as-needed basis, by trial and error.
 
Last edited:
Have you considered Black Gate capacitors at all? I didn't notice any difference replacing replacing one of the caps in my crossovers with a Black Gate, but I did notice a huge difference doing so in a DAC that I used to have. They are now manufacture-discontinued and expensive, but Parts Connexion does have them in stock. BTW the difference was not immediately apparent. This was a case where I had dismissed the mod as having no effect, but then a few monthe down the road noticed that it was sounding great.
 
Bernard,

funny... I posed the same question to Spectral a while ago, and I got smacked in the head - their response? "Capacitors have improved significantly since the Black Gates were spec'ed out"... and there you have it. More in the next post...
 
Last edited:
So there's been plenty of debate around the Mundorf Silver-In-Oil caps... To further qualify comments I made earlier, I picked up the following from one of the web sites I cited:

# Mundorf MCap Supreme
Metalized Polypropylene Film
10uF/800VDC = $60.00

All of the MCap Supremes are actually two higher value capacitors wired in series and placed inside a single wrapper. We found the original MCap Supreme to be the best reasonably priced capacitor available for loudspeaker cross-overs, today. It is rich, detailed, and full-bodied, though a bit softer sounding than the Duelund VSF. Relatively free of glare. Available in values up to 22uF.

# Mundorf MCap Supreme Silver & Oil
Metalized Polypropylene Film with Silver Flashing
10uF/1200VDC = $148.00

This is an oil impregnated, metalized paper capacitor, with the same series configuration as the original MCap Supreme. High-purity silver is used for the capacitor coating, and the winding is impregnated with a special oil. Despite the hype, we found the sound to be similar to the regular MCap, above. The major difference is that the sound spectrum shifted upwards, adding emphasis to the highs, at the apparent expense of the bass. While this creates a greater sense of air and transparency, the impression of a diminished bass response may not make it the best choice for a full-range speaker system.

From my perspective, I'll just talk dollars and sense - the 10uF/800V Supreme is $57.37 while the 10uF/800V SIO costs $207.32 at partsconnexion as of this writing. Now, I am not going to spend $400 on two caps just to be... ahem... on the "safe side", w/o any indication how safe that side is... Perhaps I am wrong...

For further reading, there is this thread on audiogon that many of us have perhaps already seen...
 
I just noticed that Parts Connexion has a sale of 25% off "Alexander by Duelund" caps.

Interesting info about Black Gate.
 
Spectral,
for the panels - I have great experience with Obbligatos. Relatively cheap and small compared to competition.
It requires some nerves to wait for delivery from Hong Kong but otherwise I can only recommend them.

http://stores.ebay.com/diy-hifi-supply50
 
bernard,

funny... I posed the same question to spectral a while ago, and i got smacked in the head - their response? "capacitors have improved significantly since the black gates were spec'ed out"... And there you have it. More in the next post...

i don't care how much electrolytic capacitors have improved over the years, black gates are still the king - still a significant improvement over any other electrolytic available today, and i mean any. However film capacitors are another kettle of fish where black gates cannot match up to. For example mundorf silver/oil is definitely more transparent than any black gate model available. The only problem is that there are thousands of fake BG's flooding the market.
 
Last edited:
I have edited post #8 above with the values of the 47uF and 51uF I have ordered; here they are again:

47uF panel -> 2x 22uF + 3x 1uF Mundorf M-Cap Supreme
51uF panel -> 5x 10uF + 1uF Mundorf M-Cap Supreme

The 1uF's were preferred as to be as far way as the others. Finally, I have ordered a bunch of the Vishay-Roederstein 1839's as bypass capacitors.

I am sure people are itching to hear the cost - about $1900, within the $2000 budget I set off with. Silver-in-oil-based combinations would have easily pushed the cost up to $8000 not counting the woofer caps.
 
Last edited:
As the caps begin to arrive today, I thought it prudent to mention that I will be replacing one cap at a time, pausing for a lengthy audition. It is very likely that a mod may end up with deleterious effects, so one needs to be very careful; not to mention that I will keep the original panel caps around for quite some time, just in case.

In the meantime, I found this white paper from ClarityCap on why capacitors sound different very interesting...

And for reference purposes in this thread, I just scraped the following list off the humblehomemadehifi web site - it presumably represents the cream of the crop according to that popular site:

Amp Ohm FE-XAL-AL alu-foil / polyster in oil*
Audience Auricap MKP *
Clarity Cap MR *
Duelund VSF Cu-foil / paper *
Duelund CAST Cu-foil / paper in oil *
Intertechnik Audyn Cap Reference *
Jantzen Audio Superior Z-Cap *
Jupiter HT BeesWax *
Le Clanché PPM MKP *
Mundorf M-Cap 630VDC MKP *
Mundorf M-Cap Supreme Cap *
Mundorf M-Cap Supreme Silver/Oil *
Mundorf M-Cap Supreme Silver/Gold/Oil *
Obbligato Aluminium Foil *
Vishay MKP 1837 * (bypass; 1839 is the same cap in axial configuration)

Finally, let me clearly state my goals with this mod: keep the tonal balance of the Odysseys, and try to improve on overall bass and panel speed and articulation.

More pictures and initial listening impressions next week...
 
spectral,

Keep us posted on how thing's progress, i'm very interested in this, and am about to begin work on my Spire xover, and am hoping that this makes the difference between an excellent speaker and a truly engaging one.

Are you also looking at new inductors, as it seems that this area can be tricky to get the values needed and keep the DCR required.

Best Regards,
Paul.
 
spectral,

Keep us posted on how thing's progress, i'm very interested in this, and am about to begin work on my Spire xover, and am hoping that this makes the difference between an excellent speaker and a truly engaging one.

Are you also looking at new inductors, as it seems that this area can be tricky to get the values needed and keep the DCR required.

Best Regards,
Paul.

All the panel capacitors are in as of last weekend, with the front woofers' having arrived yesterday, while still waiting on the Vishay bypass, with the epoxy setting coming in last... So here are my initial impressions and how I approached this mod...

  1. First of all, if you can't get shiny solder joints don't bother - you will mess it all up. Second, test your solder joints structurally to make sure they are firm, and then test, test, test and test again for continuity etc. You will end up spending about 40-60 hours total. Tools I needed: cutter, tongues, two-sided tape, WBT Silver solder, variable temperature Weller soldering gun (very important - joints can be thick and you need a higher temperature there), Super Glue, electrical tape, head-mounted flash light (very critical), Fluke multimeter, epoxy glue (yet to be used), cable ties.
  2. I decided to upgrade one speaker's panel caps first, in order to be able to perform a meaningful and easy A/B between them.
  3. The results are truly exceptional and the caps have not been broken in yet: I first played a piano recording and kept going back and forth between the speakers - the upgraded one consistently exhibited more depth and dimensionality, attack, 2-3dB added dynamic range, and overall improved clarity in the entire spectrum. The highs in particular have now become as crystal clear as you'd expect from ribbons, with incredible attack and decay (the Spectrals are really enabling that). Basically, a major veil has been lifted. By contrast, the other speaker sounds a bit flatter (less depth), a bit congested - and this is really evident with choruses, where the upgraded one just shines - with slower attack and shorter decays (the latter was very easy to demonstrate with gentle cymbal strikes: dIiIiIiiiiing vs diiing - the "Ii" is trying to also depict a sinusoidal decay). The overall outcome is more vividness in the entire spectrum. Being an extremely critical listener, this is the very first time I'd say I am really wowed by the sound of the system.
  4. These tests were repeated after swapping the channels via the Revox, and the results were consistent.
  5. I believe the upgrade can be dangerous if your system is already on the bright side, so approach this carefully.
  6. After the second speaker was upgraded, it was time for some serious listening, and mind you the capacitors are still not broken in. The results above were confirmed again: large orchestral pieces are reproduced with extreme ease - vivid, sharp, with visceral attack and believable instrument decays; very precise pin-pointing; very realistic piano attack; very realistic instrument timbers; impressive voice reproduction. And the additional dynamic range helps render big-sounding instruments like mass French horns and tubas more realistic, though this is still not horn-speaker territory either.
  7. I no longer hear the glare in the 400Hz and 4kHz ranges that annoyed me so much before.
  8. I believe all of the above are the result of simply plugging in faster capacitors, and you can probably do better with even faster but more expensive ones. But for me, it's also important not to change the speakers' tonal balance at the same time, therefore I believe in my system, the Mundorf Silver-in-Oil might make the sound too bright. If I had the money, experimenting with Duelunds would have been interesting, and I may still do it with the 10uF cap, for an additional $470 per speaker; but I am awaiting a review of the Duelund caps by Myles Astor on WhatsBestForum.com first.

So far, the $2K spent is nothing compared to the gained improvement. I have absolutely no reason to upgrade to the Summit X's at this point, and in fact, I would argue the new O's might be the better speaker now for a variety of reasons: quieter cabinet, larger panel, just as fast, better tonal balance (which on the X's is really hard to get with tilting - I feel the tonal balance in the Summits and the Summit X's was intentionally tipped up in order to make them fit in a wider variety of rooms, and tilting was provided as a way to tune that; but in the end, it's unnatural to begin with. I really like panels to be tuned sitting straight up - like the CLX for example).

I will comment on the bass after all woofer caps are in place.

Upgrading the inductors is an interesting proposal, and I have given it some thought; but there are some issues: 1) They have to be custom made due to the unusual values (no idea what the cost would be); 2) I am not happy with the wire gauge in the existing ones, so any replacements would have to use 14 or 12 AWG wire, making them very heavy and big - fitting them in the cabinet might be an issue; 3) The point-to-point wiring is still making use of 16 or 18 AWG wire, so that would have to be upgraded as well; 4) heavier gauge might increase magnetic fields with potentially detrimental effects. And I haven't researched yet how inductors really affect the sound and to what degree.

I will be posting more pictures later today.
 
Some more pictures...

IMG_4285-2.jpg
The 51uF replacement

IMG_4295-2.jpg
All the panel's caps ready to be installed

IMG_4311-2.jpg
Everything in place; 10uF top; 13uF upper left; 51uF bottom left; 47uF bottom right
 
Great job there, excellent write up. It appears that swapping the caps has had a substantial effect on the signal, (for the better), which is good to hear. Out of the caps replaced, which one's are in series with the load?

As i mentioned on Agon, i have placed an order for Clarity MR for my Spires, and have divided the capacitances equally between the originals, ie, there are 2 x 47uf for the mid band, so i'll use 4 x 23.5uf. Clarity will make these to order, but will be around 5 weeks, then i'll put them on my Audiodharma for a few days.

Let us know how things alter during the first 200 hrs with the Mundorf's, it sounds like it wont take long to get there..;)

Another question is how do you measure the FR, do you use software/mic, or sig gen etc. I will want to check mine before removing any part to keep as a reference, but will need to get a test method together. All very exciting, for us at least!

Regards,
Paul.
 

Latest posts

Back
Top