Advice for Linkwitz Orion system with Nelson Pass-style pre-amp and amp

[GeorgeBoles]

Evening All,

I am loathe to start a thread headed something like,"Which pre-amp should I build?" but it has reached the point where I need this advice from knowledgable persons. The permutations and combinations are endless.

Currently:
I have Linkwitz Orion speakers ... an active crossover design. The crossovers are full of opamps (OPA2134 or similar, 22 per channel!). The inputs and outputs to the crossovers are all unbalanced. Currently each driver is powered by its own LM3886 chip-amp, and I have a dual rail "clone" of the NS10 as the pre-amp.

But I recently heard a system in Melbourne which makes me realize I could be striving for a LOT more, and so the audiophilia has struck again.

Power Amp Plan
I plan to build 4 channels of Babelfish-J-X (30W) for tweeters and midrange, and keep my chipamps for the woofers of the Orions. (This was decided after some agony, because I really wanted to build Aleph3/Bablefish-J for the tweeters, but everything I read makes me think the X-ed DIY amps have more presence and less "nice colouration". I might have to move up to a 60W Aleph-X DIY design if the 30W is a bit underdone for the mid-range ... shouldn't be, but one cannot tell.) I have all the expensive parts, but have not purchased the transformers for these yet, so if my thinking is woolly, please let me know now!

Options I have considered for Pre-amp are:
1. CCS-X-Bosoz: lots of discussion on this board; some apparently built and sound good . I have the mosfets for this.

2. Aleph P1.7: an original Pass design - gotta be good, even if I only get it half right . In fact a good supply of mosfets for this, but it lacks this years "cool" product ... jfets!

3. The Jfet BOZ: a dead simple design (but direct from Nelson Pass himself) which shouldn't have to try too hard to drive the Active Crossover;

4. Re-make my NS10 dual rail clone with nice parts including some Jfets and lots of care

5. The Pumpkin - I have NONE of the active parts for this amplifier.

6. Jfet-CCS-X-Bosoz: I might be the third person in the world to build one of these, which might be a bad idea . I have the jfets for this.

7. Build them all and see which sounds best .

I have also considered building unbalanced to balanced converters and balanced to balanced converters and popping them into the crossovers if it is deemed worthwhile.

I know this appears to be a huge project, but I am keen to get the Orions to sound as good as possible, and Nelson has advised me that discrete component amplification makes the Orions sound a lot better!

Any suggestions please?

And finally, an acknowledgement to Mr. Nelson Pass, without whose generosity and patience I think I would never be able to think about aiming so high in the DIY home audio game. I think a lot of us would be in a similar situation.

Regards and thanks,
George.

[Nelson Pass]

Since you need to consider balanced - unbalanced conversion,
I suggest you consider the Pumpkin currently on this forum, which
is a take-off on the UGS type circuits. This should take care of
your needs for a line stage.

The Orion is a sifficiently good and interesting loudspeaker that
it deserves any improvements that can be made in the
electronics. Linkwitz appears pretty agnostic about "audiophile"
issues in electronics, but then his focus is the speakers. He
isn't averse to using "better" active electronics.

The opamps in the Orion circuits can easily be replaced with
discrete versions, and there may be some opportunities to
simplify the system overall - I'll take another look at it.

[Zen Mod]

Originally posted by GeorgeBoles
Evening All,

...............

not that I can add much (especially not after Papa's not so as usual compendious post) but I have few thoughts........

for bass section you can build whatever you want,even Babelfish JX , but count on fact that I didn't tried it yet.....so take schematic of it just as rough idea . beta tester man steen is gone fishing lately more than usual , and I just have no time .......
for mid section plain babelfish (balanced input) looks good enough

for tweet sort of mini babelfish (mebbe even with decreased voltage and current) but certainly with disconnected CCS modulation , so you will have plain SE,not Aleph one .

regarding preamps........ my humble opinion is that Pumpkin is best of all mentioned , but in every case with promised shunt regs . in that case you'll have killer package .

you can (as we spoke in other thread) make it with almost every eeny weeny jfets you have , with minor corrections .

it's your choice ,overall and after all

but....... I'm pretty sure that I'll never go that route ..... multi amp system etc .

I remember once I made something like that for a guy with KHs...... toob preamp,toob xover , sand amp for bass , toob PP triode amps for both squakers and squeeks

even if that was sort of two way active system ,pretty much balanced in overall presentation, I prefer plain all passive approach- one amp for all .

but- seems that your spks are just different animal than good ole KHs......

idea in last moment - make F4 for bass . yo can hardly beat that with any sort of money or with anything different.

in that case you really need

( who sez that I can't rumble much after Papa ........ )

[Andersonix]

The Orion plus Pass amplification would be one of my dream systems (besides ESLs). But the Orions I heard (with Hafler amps) seemed 'smeared' by all the up-front op-amps, and I expect replacing them with discreet components would be a dramatic improvement. In fact, this would be the first place I'd start, before looking at better amplification. I look forward to hearing NP's thoughts on how this might be done.

[GeorgeBoles]

Dear Nelson,

Many thanks for your rapid and thoughtful responses.

I enjoy my Orion's, and I am glad that you think they are worth persisting with in order to get the best out of them. For a DIY speaker design, the parts are quite expensive, especially for a "first effort".

I had to chuckle when you adroitly characterized Siegfried's "audiophile agnosticism", but everyone would agree that his research and developments have contributed hugely to modern audiophile enjoyment.

The opamps in the Orion circuits can easily be replaced with discrete versions, and there may be some opportunities to simplify the system overall ...

Are you saying that there could be excessive distortion/noise if all the monolithic op-amps were replaced with discrete DIY versions (especially if the implementation were not perfect); that a "sweet spot" might be found by reducing the engineering perfection* of the crossover; thereby reducing the number of op-amps; and that this may allow the use of a reduced number of more "musical", but higher distortion DIY opamps?**

Since you need to consider balanced - unbalanced conversion, I suggest you consider the Pumpkin currently on this forum

Are you suggesting that:
1. I run a balanced signal from my pre-amp (say the Pumpkin) to the Orion crossover, convert to unbalanced at the front end of the crossover, and (at the crossover output) convert back to balanced for the final run to the power amps; OR
2. Are you thinking that I have balanced outputs on my CD player/other source, and as such, the signal should be converted to unbalanced within/at the pre-amp (hence the choice of a pre-amp with this capability) before being sent to the Orion crossovers?
3. While we are discussing this, is your initial opinion that there is value to be had in converting the output of the Orion crossover to a balanced signal to feed an Aleph or Aleph-X style power amp?

Whoo, boy. A long letter (quite difficult to draft) which I hope makes sense.

Again, thanks for your input, Nelson. Like Tosh, I am very interested in your input.

Best regards,
George Boles.

* P.S. For those of you who are unfamiliar with the Orion Crossover, it "corrects" all manner of things within the speaker design: time delays, baffle effects, floor effect, the roll-off of the drivers in an open-baffle enclosure - along with the "usual" features of a crossover. To do so requires lots of op-amps - 22 single opamps per channel ... and right there starts the op-amp/discrete/musicality/distortion/feedback debate.)

**I am sure that "Grammar Check" would have highlighted that sentence!

[GeorgeBoles]

Thanks to you, Zenmod, for your thoughts.

Pumpkin it is for pre-amp, given your advice and Nelson's blessing.

I have been hoarding up parts quite wisely for a while now, and have got lots of useful bits together. In fact, given that Conrad heatsinks is "only" 2100km away, I can get them quite cheaply ($50AUD for his biggest 15 cm * 35 cm), and because there are HUGE steps in voltage/current relationships between the various amplifier topologies discussed in the Pass Labs threads, transformers are the dearest components, and "hardest to re-use in a different project" that I need to purchase. (12V for A-X 30W, 18V for Aleph 30W, 22V or so for a 56W chip amp - but to use 18V for an A-X, then I would need a HUGE VA rating and be making a humungous amplifier ... you get what I mean.)

Therefore I doubt that I really want to make too many "small amplifiers" as a starting point. I would rather keep everything at normal powers, so I can re-use the transformers. I only have a limited number of relatives to "give" my cast-offs to!

What interests me is your suggestion to remove the Aleph modulation for the tweeter. Will that purify the sound (at the expense of efficiency)? Is it just a matter of disconnection or do I have to re-calculate lots of bits and pieces? (I have to go home and eat now, and I will also need to re-read the Zen articles to work out what to disconnect .

I have been reading this thread so about two years now (paralysis by analysis), that I had sort of assumed that the Aleph modulation was a part of life!

I didn't think that the F4 would work for me, because the crossover is only a +/- 15 volt thing, the maximum voltage swing is somewhat less than that. The other threads are talking about +/- 30V or more swing to drive the F4 ... or have I got the bull by the tail, which I certainly could have done?

Bestest regards,
George.

[Magura]

Originally posted by GeorgeBoles


What interests me is your suggestion to remove the Aleph modulation for the tweeter. Will that purify the sound (at the expense of efficiency)? Is it just a matter of disconnection or do I have to re-calculate lots of bits and pieces? (I have to go home and eat now, and I will also need to re-read the Zen articles to work out what to disconnect .

The Aleph current source does little good for the highs. I would dare to claim that the only place it has no drawbacks is for the bass.
I have made a bunch of Pass style amps, but have ended up using a regular current source or inductor loads instead.

As for the trafo issue. A 2*18V 1000VA trafo will allow you to build just about any Pass construction with a minimum of modifications.

Magura

[Zen Mod]

Originally posted by GeorgeBoles
.............
I didn't think that the F4 would work for me, because the crossover is only a +/- 15 volt thing, the maximum voltage swing is somewhat less than that. The other threads are talking about +/- 30V or more swing to drive the F4 ... or have I got the bull by the tail, which I certainly could have done?

Bestest regards,
George.

I told you that my brain is sorta in red this days

but- I think that there is also solution for that issue.........

Magura is right for 2x18/1000

[Nelson Pass]

Originally posted by GeorgeBoles
Are you saying that there could be excessive distortion/noise if all the monolithic op-amps were replaced with discrete DIY versions (especially if the implementation were not perfect); that a "sweet spot" might be found by reducing the engineering perfection* of the crossover; thereby reducing the number of op-amps; and that this may allow the use of a reduced number of more "musical", but higher distortion DIY opamps?**

No. But my experience is that usually complex filter sets are
often composed of elements which "stand alone" without
reference to other elements, so that there is often quite a
bit more active circuitry than necessary. If you believe in the
perfection of the op amps, then you generally won't mind
running the signal through lots of them. Myself, I am willing
to put up with minor imprecision in filter curves as a trade off
for reducing the number of op amps in the gain path. After all,
the result from the loudspeaker will only be an approximation
anyway.

When I was commissioned to rework the cutterhead electronics
for Mobile Fidelity, I found that the many complex elements
involved with equalization, gain and feedback around the
cutterhead each had its own modular circuit and its own op
amp. I think it had about 11 op amps in the signal path -
although I don't recall very well. It certainly simplified the
task for the original designer.

In any case, I was able to reduce the whole thing to one op
amp and retain the measurements of the original. It was a lot
of work though, because allowance had to be made for the
interaction of many passive networks around the single stage.

[GRollins]

It's not so much a question of distortion in the conventional sense of THD...after all, with sufficient quantities of feedback, you can achieve vanishingly small amounts of any of the usual distortions.
But therein lies the rub.
Huge amounts of feedback tend to impose their own penalties.
Not to mention the sheer quantity of devices (active and passive) that the signal has to go through. One relatively easy example is the passive devices within the chips; there are few people who would claim that a capacitor made of silicon is equal to a good film or mica cap.
Yes, the music gets through opamps, but it's been processed into McSound.
Better to go discrete when possible.

Grey