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

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The one and only
Joined 2001
Paid Member
Re: O'Brian works for me too...

Tosh said:
Linkwitz is concerned with the basics: getting the loudspeaker to have a uniform polar response. (See his recent lecture here)

That's an interesting piece, much of it agreeing with my own
experience, although I don't think it addressed one of the most
interesting issues - the differences in polar response of a dipole
versus a small source. The ideal dipole has no lateral response
(90 deg off axis), and of course the small source does. What they
have in common is little or no diffraction off enclosure boundaries.

:cool:
 
Being the owner of Magneplanars, I suppose I fit handily into the dipole category, even if they follow a slightly different approach from other dipoles.
I don't think there's much question that speakers are one of the weaker links in the chain, but that can become a problem if someone chooses to take that as meaning that the source and electronics aren't important. At which point you've thrown out the bath water, the baby, the towel, and for good measure, the bar of soap. Whether Linkwitz fits into that category I'm not sure, but it's certainly not impossible.
Don't misunderstand--the idea of open baffle speakers is good. If you use good drivers you can come up with excellent results. I'm just a little concerned about his apparent lack of interest in anything else in the system. If, for instance, we assume that he used chip amps to voice the speaker, there might be...issues, shall we say, when it's hooked up to other electronics.
If I get a chance, I'll look into the lecture thing and come back later tonight. (My fingers are crossed that it's a light night...)

Grey
 
Grey,

regarding the HF filter, I suppose it's just kind of a safety issue, because in a typical HiFi environment you can't be sure that there is not a siginificant amount of HF around, given unperfectly shielded devices, unshielded cables used by some people and so on. AM-broadcast stations nearby, cell phone (TDMA) poison, SMPS in all sorts of other equipment... btw, do you know the extremely severe HF/EMC test conditions for mic preamps and other studio gear to be used in german broadcast etc? Not a single unit will ever pass anything above 70kHz if it aims to get the approvals.

So IF you know your system is free of HF, dispose of the filter. If not, better to loose a dB or so at 20kHz (we should consider phase also, more precisely L/R phase match) than to hear Radio Eriwan in the background... and, last not least, SL is a seasoned RF engineer -- look at the list of RF analysers he has (co-)designed while at HP.



I onced listened to the Orion and they blew away, even with "that dreaded op-amp" XO and a cheap AVR for power amplifier duties.

One could think of many ways of improving the electronics, successively. Install the best available op-amps (say, National's LME series or one of the TI THS types... or even discrete JE-990's or so), maybe bias them to class A, improve the power supply, make the shelving filter passive. If one doesn't want feedback of any sort one must go for passive LRC filters which then require low impedances to be doable (uhm, them coils), the phase shifters could be done with xformers (or align the speakers better)....

- Klaus
 
I found that transcript to be interesting, both for what it said, and for what it did not say. There were lots of pretty graphs and things, but honestly, I didn't see anything groundbreaking. The points he made are at least twenty years old and quite possibly older.
Some specific points:
--It is implied, if not stated outright, in the transcript next to Slide 8, that the photo is of his listening room. I have two quick comments. One is that the room, while an interior decorator's dream, has vast expanses of glass. Sorry. Beautiful it may be, but that's not a good listening room. I'm sure that Linkwitz's response would be something to the effect that the speakers are far enough away from the glass that the reflected sound is ignored by the listener's ear. But he, himself, admits that the room is hard, in the acoustical sense, with a long reverberation time. He rationalizes this by saying that reverberation times are not important for small acoustical spaces. I vehemently disagree, but will leave it at that. My other observation is that he is clearly breaking his own dictum to have a symmetrical listening room.
--I agree with his stated guidelines, but not necessarily for the same reasons he advances:
1) Symmetrical listening space (even if he breaks his own rule). I went so far as to move a door in my listening room during rebuilding so as to make it match one on the other side of the room. Why not? I already had all the pry bars, hammers, etc. out...might as well take an hour or two and reframe the silly thing. Besides, they'd done a lousy job of hanging the door the first time. By the time I got done with it, it hung perfectly. But I digress...
2) Delay reflections by at least 6mS. Honestly, more is better, but I won't quibble.
3) Off-axis sound (and by implication, the reflected sound) should approximate on-axis sound.
This is all pretty much old news, which isn't to say that it isn't important or correct, just that I'm not sure how timely his revelation is in the overall scheme of things.
--He repeats several times how similar the two speaker systems sound. As a matter of fact, it's the entire foundation on which his presentation rests. There are any number of possible conclusions that could be drawn. One is that he designed both speakers, so it's no surprise that they should sound similar. Assuming that he has a fairly clear idea of what he wants in a speaker, he's going to keep fiddling with it until it sounds the way he thinks it should sound. If he can't get the sound he wants, he will presumably try different drivers/cabinet configurations/crossovers/etc. until he succeeds in getting something that sounds "right" to him.
To use a commercial example, all B&W speakers have an identifiable house sound. From the largest to the smallest, they are all cut from the same sonic cloth. If you like the B&W sound, you'll like all their speakers to greater or lesser degrees. If you don't, avoid them like the plague, because you will not find anything in their lineup that will satisfy you.
--Several of his conclusions are, at the minimum, open to question. He states that imaging is dependent on the three things listed above. Fair enough. But all it takes is one counter-example to burst the bubble, and I have just the pin.
Behold, the Rogers LS3/5a.
The original mini-monitor. The one that started the entire orchestra-in-a-box, sub/sat phenomenon. I still have mine and I still listen to them sometimes. They have obvious flaws; the bass is uneven (on purpose) and they flatten dynamics something fierce (a particularly complex crossover sporting [ugh] iron core inductors), but they are justly famous because they are true to the essence of the music and they image like a bat out of hell. Believe it or not, there was a time when there weren't any mini-monitors. With the advent (ahem) of the acoustic suspension woofer, speakers began to get smaller (and less efficient--something the LS3/5a took to extremes), but to build a speaker less than the size of a shoe box and have so much music come out of it was a near miracle.
So what's my point?
Well, you see, the LS3/5a is a quirky little thing. Pull the front grill and you'll see a KEF B-110 mid/woofer and a modified KEF T-27 tweeter. But then you notice something odd...thick felt surrounding the tweeter. That felt is close to the dome. Very close. As close as the exposed leads of the T-27 will allow. And it's so thick (on the order of 3/4") that it reaches from the front of the speaker to the speaker grill. That's on purpose. The felt is there to absorb any high frequency information that tries to go across the face of the speaker to re-radiate from the front corners, which are quite square, thank you very much (this being long before rounded corners came into vogue). Now, one of the things about this felt is that it enforces a pretty strict radiation pattern on the tweeter--it is an unapologetic point source. But note that it's not an omni-directional point source, it's a uni-directional point source. Closer to a laser than a candle flame in radiation pattern. And that, right there, is the counter-claim. The LS-3/5a's woofer is not so-treated, and can be considered omni-directional, at least in the lower frequencies (crossover ca. 3kHz if I remember correctly), but that tweeter...that tweeter is different...and in doing so it lays waste to Linkwitz's idea that the off-axis sound should be a close approximation of the on-axis sound for good imaging, because the LS3/5a was anything but similar in off-axis response.
So is Linkwitz wrong? Well, yes...and no. In that he indicates that the only (or perhaps the best) way to achieve good imaging is to obey his three principles, he is wrong, clearly. You only need one good counter-example to prove a point. But that even off-axis response is a good thing, even if not necessarily for imaging--now that I can agree with. It brings fuller, richer sound to the table. But not necessarily better imaging.
Indeed, one of the more bizarre speakers ever to hit the market, the Walsh, and its spiritual descendants, the Ohm speakers, are omni-directional to a fault. They image widely, and I mean wiiiidely but with next to no image depth. The effect is not unlike that spaced-out image you get if your speakers are out of phase. But this only serves to emphasize that a uniform off-axis response does not of necessity give you a good image.
...Just in case you didn't consider the bubble sufficiently burst before, that is...
--One thing that I don't feel that he addressed particularly clearly is the reduction in side-radiated acoustic energy from dipolar speakers. Perhaps he didn't emphasize it because it ran counter to his thesis. I believe I saw where he noted that the dipole had a deeper image, but he took it no further. Suffice it to say that if there's less sound hitting the wall, then there's less sound to reflect to your ear and confuse the image.
Particularly in a room with twenty-seven acres of highly reflective glass.
--I also have to wonder if his impressions of a flat speaker having too much high frequency energy were also influenced by that room. No, I'm not necessarily arguing that his essential point is wrong--others have reached similar conclusions--but that he might have gone further than necessary. As evidence, notice the distinctly lowered upper high end in his graphs. A bit overdone, perhaps?
--Finally, I'd like to advance the possibility that the reason that he feels that the two speakers sound the same is that he has a terrible listening room, literally awash in reverberant sound. 450 mS by his own measurements. That's enough reverberation for a largish jazz club or a small church. With that much reflected sound zinging about, it would be difficult to tell an aardvark from a zebra. Yes, he says that others reached the same conclusion about the speakers that he did--but in that same room. I have to wonder if the same conclusion would be reached in a better listening room.
I won't insist on these last two points...just something to think about.

Grey
 
KSTR said:
Grey,

...regarding the HF filter, I suppose it's just kind of a safety issue...

... and, last not least, SL is a seasoned RF engineer -- look at the list of RF analysers he has (co-)designed while at HP.



Indeed, and a safety issue that any number of other pieces of equipment have managed to do without. If you think it's cool, leave it in. Won't bother me a bit.
As for Linkwitz and his design credentials, I have degrees in geology and psychology, I'm a pretty good bass player, a whiz with bees, and I brew excellent beer. None of which are in any way relevant to my abilities--or lack thereof--in the audio field. They're separate and distinct. At best, I may drag them in when making an analogy but that's about the extent of it. Linkwitz's RF designs do not necessarily make him a good AF designer, any more than my ability to cook gourmet-level steaks means that I'm a crackerjack asparagus cook (trust me...I run from asparagus...wouldn't have a clue how to cook it).
I am saddened--but hardly surprised--to hear that German radio stations bandwidth-limit. Stations here in the US are no better. Of course, it's been eons since serious audiophiles cared to use FM broadcasts for source material.
In all fairness, I will muster a lukewarm defense for the Public Radio folks. They at least let some low end through. On some of their stations. From where I am I can get acceptable signals from three different PBS stations...and none of the three sound the same, despite broadcasting the same program material. But even the worst of the three sounds better than the best of the other local stations.

Grey
 
Grollins, good example that one of the restaurant, and I agree. Let me put another one:
"Look at this guy, he's just made a car using a lot of hours just on it's shape, he says it's the main point of a racing car. Oh my, he just throw the first motor he got on ebay!. I know how to make motors, and for sure I know of aerodynamics. I'll try his car with the motors I make, maybe we got something there for a Christmas present."
 
Linkwitz's RF designs do not necessarily make him a good AF designer, any more than my ability to cook gourmet-level steaks means that I'm a crackerjack asparagus cook (trust me...I run from asparagus...wouldn't have a clue how to cook it).

My speakers don't seem to use Rollins crossovers.

BTW, public service message: just trim off the bottoms from the asparagus, toss them with some olive oil, salt, pepper, garlic, and a touch of balsamico, and throw them on the grill. Cook, rolling them around occasionally, until they're done.
 
SY said:


My speakers don't seem to use Rollins crossovers.



I have a love/hate with any filtering system, be it RIAA or crossover (passive or active). I find them perversely fascinating, but frustrating to work with. Every time I finish one, I swear that I'll never do another filter. Six months later, I find myself scratching out a new schematic on a piece of scrap paper. Oh, well.

Grey
 
diyAudio Editor
Joined 2001
Paid Member
Well, Linkwitz certainly understands filter systems, which is why his experience is more relevant that if his expertise was how to cook a steak.

I find the comments about his listening room interesting and fun, but to comment negatively about any speaker design that has such a good reputation without listening to it is kinda silly- especially when it is different than most others.

His ideas WERE more radical back in the day of the Audio Artistry Beethoven speaker, which had a great reputation -when open baffles were rare. I think a lot of the current open baffle interest is due to the AA speakers..

Clearly the guy is doing many things right, but when you are a groundbreaker, eventually the world will catch up to you. In fact though, few speaker makers use his circuit intensive approach even today. Why? well we can find out. I think that is the point of this thread- Can the Linkwitz circuitry be kept intact, using less components and discrete components, yet still be true to the Linkwitz approach? If a bunch of opamps can be eliminated while still performing the same exact functions, I think that that would be interesting to hear - how much better would it sound?

The idea would be to be able to compare the two I would think. Another idea is to modify the circuitry itself to be "better" , but that's anothe project.

My son thinks that Jimi Hendrix isn't that radical, because lots of people play like him now.....
 
At what point does it cease to be an Orion?
This is not just philosophical musing. If, for instance, someone were to take my earlier suggestion and use a physical offset to time align the drivers rather than do it electronically, would that still be an Orion?
If the intent is to re-imagine Linkwitz's speaker, it would be useful to set some ground rules as to what does and does not constitute fair play. I suppose it's clear that I'd go back nearly to square one, others seem to feel that a simple opamp swap is the limit.
No, there's no reason that people can't do various levels of modification, according to their level of experience and ambition, but then the thread will get badly fragmented, with some going for a full rehabilitation of the crossover, others swapping opamps, some opting for a physical rebuild. To keep it coherent, I'd suggest laying out a clear program, just to keep things organized.
Also, just out of curiosity, how many people out there actually own one of the Linkwitz designs? And of those, how many are seriously interested in modifying their speakers?

Grey
 
Modifying orions

Hi Grey,

me, for instance. I own a pair of Orions and I am just about to reproduce the XOs in the digital domain (oh, how shocking, my dear).

I would be very keen on comparing the original, the digital, and the "passed" XOs. Unfortunately, too stupid to do the latter on my own.

Cheers,
Thomas
 
Account Closed
Joined 2001
Folks,

I think "the line" with SL is very clear. And he notes this in the "caution" at the bottom of his front page.

http://www.linkwitzlab.com/#CAUTION:

Any of the design modifications being discussed here would cause the Orion to cease being an Orion....in SL's view.
However, I think he enjoys/welcomes new thinking on any of his efforts. But you better have your A-game ready if you find yourself in a discussion with him about it.

SL does not participate in most discussion forums....which I guess kind of makes him an easy target, but you might find him at the OUG from time to time.

Here's an example of recent "modification" to one of his designs and the discussion that ensued.

http://orion.quicksytes.com/viewtopic.php?f=12&t=784

Cheers,

Dave.
 
I'm interested in the Orion because I am getting ready for a speaker upgrade, and it seems like a good choice: It seems to be a credible, accepted design. Lacking carpentry ability (and inclination), I need something simple that I can talk somebody into helping me build. I like the idea of the open baffle. And finally, I am a firm believer that biamping/triamping is the way to go, and there aren't very many DIY designs out there employing it.
My interest in this thread is because I would like to use a different version of the electronics for the Orion. Also, I need an excuse to keep building amps. I'm not particularly interested in just stuffing the Linkwitz board and knocking off a few gainclones. I would like to see it with simpler but high-end DIY electronics. I may end up using a DSP-based system to tinker around with it a bit, but in the meantime it would be nice to build something where I am reasonably confident of the results.
 
Hi all,

I think it is ok, if anybody tries to improve Orion desing.

But. Anybody who is saying Orion is from stone age etc. should first "sit back and listen" properly built Orions with electronics Linkwitz recommends.

If one is not like what he hear, then it is OK to say what he think is reason to bad sound.

OP amps could be bad somewhere in poorly designed system, but Orion loves op-amps IMHO ;)
 
diyAudio Editor
Joined 2001
Paid Member
If, for instance, someone were to take my earlier suggestion and use a physical offset to time align the drivers rather than do it electronically, would that still be an Orion?
If the intent is to re-imagine Linkwitz's speaker, it would be useful to set some ground rules as to what does and does not constitute fair play. I suppose it's clear that I'd go back nearly to square one, others seem to feel that a simple opamp swap is the limit.

First, I'm pretty sure that offsetting the drivers is something that would occur to Linkwitz and he rejected that. As I pointed out, he had a manufacturer making his (quite expensive) speakers in the past- he isn't some guy that just made some speakers for himself in his living room. I don't think that he would be concerned with their possible lack of woodworking skills!

Offsetting the drivers is like just about everything, a compromise, as if you offset the drivers you can end up with baffle diffraction problems, if you slope the baffle, then the response changes depending on how close you sit, etc. So I think that that choice is a fundamental SL choice and shouldn't be tampered with.

I think that the scope of the project is pretty clear:

This project is to leave the Orion panel and drivers alone, and confine any modifications to the circuitry. This would involve substituting discrete opamps for chip opamps, and combining circuits
as is possible to avoid redundancy and eliminate maybe half the opamps. Of course something like leaving off the high filter would be something that could also be easily explored...

With this definition, the project could be useful to any of the owners of an Orion, and actually encourage more people to build them, which I assume would make Siegfried Linkwitz more interested.
Also it would allow an easy comparison between the stock crossovers, and the new crossovers, which is one of the major appeals of this proposed project

Unfortunately , this combining of circuits involves "The Art of Electronics" to coin a phrease ;) , and will require someone of a certain level of expertise....

Also, It appears that the only way to get the current exact circuit is to have a set of his boards, which some people here have. I believe that he sells unpopulated boards. Perhaps someone with the boards can compare them to the circuit I linked to. Is it true that there isn't an exact schematic of the current circuitry
available?
 
diyAudio Editor
Joined 2001
Paid Member
the name should have ASP in it as that is what we are replacing it we agree to only focus on the "black box" Aurion is a little too close sounding to Orion so I think that might cause confusion. Nice of Linkwitz to offer a final tribute to the disappearing planet Pluto...

How about something like the "Mercury ASP" or as Homer S. would like: "Uranus ASP"
 
Variac,

Orion ASP boards and documentation are available, but a bit pricey:

http://www.linkwitzlab.com/orion_construction.htm

Unpopulated boards are also available for 1/3 the price of the Orion set of documentation:

http://www.linkwitzlab.com/pcb.htm#ASP

A complete Orion ASP block diagram with design specs is available at:

http://www.linkwitzlab.com/orion_asp.htm

SL provides all the design formulas for each stage of the ASP on his site:

http://www.linkwitzlab.com/filters.htm

A simple way to reduce the transistor count of the Orion ASP is to replace all the buffer stages built around the OP2134 with a discrete buffer stage. An interesting topology could be the one used by Borbely in his low cost starter kits:

http://www.borbelyaudio.com/eb604417.asp

A complementary cascoded JFET buffer using only 8 semiconductors.

A more radical approach to sallen and key filter replacement is described in:

http://www.essex.ac.uk/ESE/research...s/C14 A family of circuit topologies LR-4.pdf

I hope this help. I'm looking forward to Nelson's thoughts on this subject.

Regards

Giorgio
 
the name game

So Orion was the lover of Artemis. She also killed him. (oops.)

Scorpio was what Orion was chasing after that led to him being shot by Artemis.

Apollo was the one who sent Orion on the trail of the scorpion, and also the one that shined the beam of light on Orion far away at sea. And he was the one that challenged Artemis to hit the little spot of light with one of her arrows.

The Apollo ASP? The Artemis ASP?

Of course then there are the various stars in Orion--most famously Betelgeuse, but also Bellatrix, Rigel and Saiph; Mintaka, Almintaka and Anilam make up the belt. (It's been some years since Astronomy class, my apologies for any misspellings.)

eL
 
I saw a SF film from 1978 called "Capricorn One" last night. A bit like Apollo 13 meets All the president’s men. Not a bad film. I see Its Title ( minus the nefarious subplot ) as a Procrustean fit: a (science) project diverted from it’s intended aim and it’s also a star constellation.

So... Capricorn ASP? :)
 
Variac said:


Offsetting the drivers is like just about everything, a compromise, as if you offset the drivers you can end up with baffle diffraction problems...



This is why they invented felt and foam. Felt is better over the long haul. The problem with most speakers is that they cheap out and use too little thickness. It takes something on the order of a quarter wavelength thickness to achieve decent attenuation. Time, time, and again, I see felt that is, at most, a sixteenth of an inch...an eighth of an inch thick at best. Yeah, right, guys...care to try again?
As for offsetting the baffle occurring to him--don't bet on it. If he thinks that an electronic solution is satisfactory, then it may never have crossed his mind. If there's one thing I've learned over the years, it's that if someone already has a solution in mind, they are resistant to trying any other possibility that may come up. You see it everywhere, Washington D.C., the Middle East, at work, in marriages, you name it.
You can't have it both ways, either the guy is a genius at filters...in which case it's a big duh that he's going to approach this as a filtering problem that has an electronic solution as opposed to a mechanical one, or he's not...in which case the whole crossover is open to redesign (and possibly even the mechanical design of the speaker itself). So which is it? Make your choice and place your bets, folks.
The blind assumption that someone has evaluated all the options and chosen the best is antithetical to the basic philosophy of DIY.
If the idea is to keep everything as similar to the original Orion as possible, then you're limited to two options: Try different chip opamps or build discrete opamps.
There. Project planning done. Gentlemen, start your soldering irons.

Grey
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.