Shipping

The big speakers travel strapped to a pallet, the grills facing inward. Inside the carton are sheets of 1.5" solid styrofoam full width and length protecting sides and back. You can load pallets on top, drop them, or otherwise abuse without damage.

The only assault not prevented is the old forklift-tyne thru the carton. Even wooden crates don't stop that. Hasn't happened in a long, long time.

Factory break-in?

We use a function generator for at least 24 hrs (48 hrs on large systems) that sweeps 20Hz to 2500Hz every two seconds at the 2W level, a very strenuous breakin. It's the equivalent of 10 times that time with music. But I agree all parts breakin: caps form, wire sets up a conductive path, and there is the loosening of suspensions from motion and excursion.

Did I mention the damping adjustment is VERY IMPORTANT?? I thought so.

Do you also burn them in after assembly (floorstanding loudspeakers)? For how long? 

They get a minimum of 48 hours on the function generator which includes a 20Hz to 2500Hz sweep repeated every 3 seconds. Plus they are then listened to for several hours and a preliminary tuning of the PR and setting of the level controls is done. The rest is up to you.

I don't think I understand the meaning of Q and how it is affected and believe it would help although I love my speakers the way they sound now.

"Q" stands for "quality factor" and is a measure of the system bass characteristic. Generally a "Q" of .7 is considered ideal if the entire system (woofer plus enclosure plus vent if any) can achieve it. Ported systems have high Q (above .7), sealed systems have low (Q) (below .7). A tunable passive radiator system like ours is the best of both worlds, since it adjusts to various Q's and even compensates for the series resistance of your speaker wire.

In the past attempts have been made to produce enclosureless bass systems with very high Q's designed to overcome the severe rolloff caused by the dipole effect. Carver's ribbon speakers used an array of open-baffle 10" woofers with extremely high Q's (around 10). It didn't have good extension. Some people prefer overdamped bass (Q of .5 and below for the system). I believe in the middle road, system Q of 0.7.

Does anyone know if these speakers (ribbons) sag or relax over time like some others?

The ribbons don't sag. They are clamped at the edges all around, and the diaphragms weigh about 1.5g. No mass to speak of, so no weight to pull them down. They should last basically forever. The tweeters have a moving mass of 0.1g, same thing. Longevity should be excellent.

Foam surround ageing

If the foam appears intact leave it alone. If you can poke your finger through it you'll need to replace the woofer. We have a new 38cm driver with a double thick surround guaranteed 20 years not to rot. The 30cm is rubber.

Spikes

Spikes both couple and decouple the cabinet/speaker output from the floor.

Bass wavelengths are quite long and, below about 200Hz, boundary dependent. Without a surface to travel along they dissipate somewhat rapidly. A woofer would ideally be as close to a boundary (floor) or multiple boundaries (side and back walls, and even ceiling) as possible, or at least a constant distance from them. By elevating a cabinet from the floor with spikes, you reduce the propagation efficiency of bass wavelengths. So, you decouple bass from the room, even if ever so slightly. The effect is quite audible.

Spikes couple cabinet output to the floor, turning it into a transmission medium. Soundwaves travel through many solids much more rapidly than through the air. Instead of "moving the floor", cabinet output is transmitted to the listener ahead of the music, through the floor (made usually a good carrier of sound like wood or stone). This is why I'm no fan of spikes, and the Sunfire people aren't either.

Try some damping compound between the spikes and the cabinet (not between the spikes and the floor) and let me know if you hear a difference. I've seen composite spikes that were metal only on the tips, otherwise rubber. Should work better.

Since spikes do two things I don't like--diminish bass propagation, and transmit or even amplify spurious cabinet talk--I never recommend their use.

As Sunfire recommends, rubber or other absorbent materials can be used as feet for speakers or subs.

Since a lot depends on the height of the stand and the materials from which your floor is made, why not experiment? Personally I like Dynamat.

How are you able to incorporate ribbons and dynamic woofers in a flat frequency response speaker, where so many have failed?

This calls for a booklength response. I'll give you a few short pointers:
1. Use push-pull ribbons. Single ended waveform fidelity is terrible. You only build a ribbon single-ended to save money.
2. Use low crossover freq. Right now no ribbon mid we have crosses over higher than 166 Hz.
3. Have the ability to adjust levels with great precision. Our electronic crossover permits 0.01dB increments per channel.
4. Build passive crossovers to very tight tolerances. For us that means four decimal places, or 1/2000th of 1%. We laugh at the tolerances (5%, 1%) even the best competitors use.

How much of an improvement (if any) would you get by using the active crossovers as opposed to the passive crossovers? Are active crossovers for the ribbons more complex than the standard 24db/octave units you can get from some place like Marchand?

I only like to use steep filters in lowpass applications. 24dB/oct is in phase but has about 10msec delay at 100Hz. Also, steep filters ring, and the high pass section of the filter reveals that ringing.

What I do is an active 24dB lowpass and a passive 6dB highpass to the ribbons, with polyprop caps (Kimber, Axon, plus our own 400V polyprops). An extra filter pole is added below out of band to accelerate rolloff at bass frequencies. This combination gives me a lot of flexibility: I can stagger poles, tweak tolerances (the aforementioned 1/2000th of 1%), experiment with different caps and wire etc. No need to upgrade our crossovers--they're executed at a much higher level than any commercial design I know of.

Voicing equipment?

We have two setups to voice crossovers/speakers and a system must sound good on both. First is Krell MD10 CD transport, Wadia 27ix DAC, SST Son of Ampzilla amplifiers and Kimber Select IC's in a completely treated LEDE 14x31' room.
Second system is a $99 Philips CD changer, Parasound 2200, Ratshack IC's and Belden 16 gauge speaker wire in a completely untreated, very large room.
Between the two it's very easy to hear flaws and problems. We optimize xovers to 1/2000th of 1% tolerance, which makes the 1% and 5% parts found in commercial designs look pretty silly. It's important to us however.

How do you make your cross-over to 1/2000 of 1% accurate? I find this statement quite overwhelming. Is that mean the each cross-over you make is within that tolerance to each other in electical behavior? Do you hand-tune each cossover?

Yep, and I do it myself. Have a B&K Precision cap meter accurate to four decimal places. Each xover is trimmed to exact value, no tolerances. This means, for example, 1.600uF, not 1.6uF. We laugh at 5% and 1% parts. You only get repeatability with high levels of precision.

How can you tell the polarity is reversed?

Inverted polarity means the leading edge of the music waveform is ongoing negative. Trumpeters suck their instruments, singers inhale their notes. It's an amusical, dull sound. Try both polarities to discover the correct one. Requires you reversing the speaker leads at both speakers.
Don't bother if your speakers are the ordinary kind with inverted polarity drivers. It's wrong both ways.

Is that mean you can only hear the polar difference using ribbons but not the cone/dome?

You can easily hear polarity on all our speakers. The "MP" in "VMPS" means "minimum phase". All drivers are electrically in phase, and first order networks perserve the phase coherency of the envelope.
Common practice is to wire the drivers in a multiway system alternatively out of phase, to maintain good amplitude linearity through the crossover region. Unfortunately in the passband the midrange (of a 3way) is out of phase with the woofer and tweeter. Most manufacturers do it this way because it measures better. The practice destroys the integrity of the music signal and I hate it. You can cover the "in phase notches" in a crossover network by various means. Although the in-phase speaker will measure less flat (particularly in the crossover region) than the non-minimum phase design, its sound is superior. Some designs in addition are completely phase random, with slopes claimed to be approaching 100dB/oct. You can't tell whether the music is coming or going with such speakers, which I also dislike.
Speakers that have out-of-phase drivers include the Magneplanars, Newforms and virtually all cone dynamic two and three ways.

You talk a lot about appreciating the sound of speakers where all drivers are in-phase. It goes without saying that this is a very desireable thing. I understand why it isn't done in a conventional passive crossover (to cover up the notch at the crossover frequency). However, if the only cost involved in going to an in-phase design for a speaker manufacturer is the addition of a few small components in the crossver, why don't more manufacturers do this?
You mentioned a while back that very few manufacturers have in-phase crossovers. Would you be kind enough to name a few, just for educational value?

Bud Fried was a real advocate of series first order filters and in phase wiring of drivers. For years he and I were the only ones do so. Recently I understand Shamrock (Mike McCall) wires his drivers in phase.

I think the main reason most designers invert polarity of alternate drivers in a multiway system is that it measures better that way. No one wants to be embarrassed by a measureable suckout in the FR curve. Although there are means available to minimize problems in the crossover region caused by in-phase design, truth is an in-phase configuration will never measure as flat through the crossover than the inverted-polarity wiring. Lord knows we want speakers to measure well.

The highly audible fact that inverted driver polarity destroys the integrity of the music is of no concern to these people.

I've heard that one can't compare the sound of ribbons/planar to dynamic/cone speakers because their sound "characteristics" are different. To be honest, I don't know what that means. I thought music is music and that accurate sound reproduction should be the same regardless whether the speaker is cone or ribbon. Can you/anyone offer any insights into this?This is a long and involved subject, particularly since "ribbons" are not generic where you can easily compare one to the other.

The best known "ribbons" weren't ribbons at all, but single-ended planar dynamic. They only reproduced half the waveform, THD averaged 30% below 5 kHz, where a push-pull tweeter came in. To add to the problems, the tweeters and midwoofers were wired in opposite polarities. There were huge differences between the Apogee sound and convential cone dynamic speakers because the Apogee's errors were so huge. Not to say they didn't sound good, they just had serious design flaws.

Todays push-pull midrange and treble ribbons are a completely different animal. They sound a lot like conventional speakers (if you can find such with all drivers wired in phase, still a great rarity). Properly executed modern ribbons just sound more like live sound.

With the right amplification VMPS ribbons sound extraordinarily lifelike, even the smallest model. Normally I am not great enthusiast for my own stuff (all I ever hear was the problems) but two nites ago, listening to the RM 2 Neo's through a pair of CJ Premier 12 tube monoblocks, the sound was sublime, without defect. The music (Beethoven Pastorale, B. Walter on CD from 1958) haunts me as I write.

One thing I noticed about VMPS - every magazine I have has rated them "Best sound of show" at one show or another. 'phile, listener, positive feedback, TAS, T$$, etc. I decided it was either.....

1. Brian sets up better for a show than anyone.
2. The speakers are just incredible.
3. Both #1 and #2
I'm hoping it's #3.

Question to Brian and other RM2 experts out there: is the RM2 (or other VMPS spakers in general) picky as far as placement goes? What about amplification? Thanks.

I always thought a speaker must be flexible in placement and setup. It should adapt to its environment and associated equipment, not the other way round. After all, your chances of finding the exact right sources, IC's, and speakerwire by trial and error are about zero, which leads to the frustrated audiophile who owns good equipment and is getting bad sound, wanting to chuck the hobby completely.

The RM 2 Neo may be placed as close as 4" to a back wall, tho I use about 2.5'. Ditto the side wall. It can go in the middle of the room. It can be adjusted for rooms as small as 8x10' or as large as 20x50'. Level controls and the bass damping adjustment of the passive radiator permit all these things.

Too many owners ignore the setup instructions and get less than optimum results. A VMPS floorstander sets up like no other speaker and you have to train your ear to what is better and what is just different. It's worth the effort.

Passive biamping?

A very good compromise would be to get an SS amp that has good current and biamp your speakers. You can use the tube amps for the mids and up and the SS for the bass. I think you might really like the results. Of course you need a good electronic crossover.

Actually passive biamping with the builtin crossover works very well and no electronic crossover is necessary (though you can use one once the woofer coil is bypassed provided the xover offers 6 dB or 24dB slopes, don't use 12 or 18 dB).

More on biamping

You don't have to worry about the power mismatch, it is input sensitivity you worry about. Typically tube amps have higher input sensitivity than solid state. If the level difference is small you can compensate with the speaker level controls. Otherwise you'll need a volume pot on the tube amp, or use a tube amp that has level controls, or an integrated tube amp. 

If your pre has one set of outputs get a Y connector/splitter to turn one output into two. Run fullrange signals into each amp, the speaker does the crossovers. Once you have adjusted relative levels you're ready to biamp. 

If you don't do the level matching all you'll hear is the mid/treble amp, with very faint bass.

Level controls can't increase output, only cut it. Therefore you need an adjustment on the louder (more sensitive) of the two amps, unless you're using identical amps. Tube amps are invariably more sensitive than SS.

RM2 placement setup

You can sit fairly close to the RM 2 and get a wide soundstage. Simply toe in the speakers more severely towards the center the closer you sit so that they crossfire a few feet in front of you.

My room is 31' long and I sit about 25' back. I toe the speakers in so I can see the front and back outside edge of each speaker cabinet, and a good expanse of the outside baffle. I toe the speakers in until the center image is very firm. More toe in makes the image bloat, less toe in makes it diffuse. If I move say, to 10' away I make a sharper angle towards the center of the room, always so I can see the front and rear outside cabinet edges. Try it and see.

Why does mass loading affect the midrange? And what exactly does it affect? I also find it interesting such a simple idea is not used on any other speaker I have ever heard of in recent time.

The idea is simple and necessary. The bass loading affects the entire spectrum right up to the treble. I adjust my speakers all the time, like when trying out new equipment. Especially with the Analysis Plus wire (which required considerable undamping) I'm glad the adjustment is there. 

Most adjustments have been taken out of speakers since manufacturers concluded customers are too stupid to follow directions.


Cable lengths?

I just did an installation with the RM 40 where the owner, for reasons of his own, had a 5m run of M350 interconnect to a subwoofer and another 5m run back to the power amp from the woofer line level out. In the bypass mode that meant 10m of interconnect between preamp and amp. The signal loss was amazing, including virtually the entire bass range. The level was down about 2dB according to the preamp volume control which had 1dB increments. Restoring the system to a 2m pair of IC's also restored all the music. 

I am firmly in the short interconnect, long speaker wire camp.

Passive radiator cones

The PR cones are treated paper. Paper is fine as long as you don't have to listen to their high frequency noise and distortion products. Facing the PR down and slot-loading it out the front filters such products out nicely.

Series crossovers

Gotta agree with Bud about series first order filters for speaker crossovers. Everything he says is dead on. Which is why I use them in our speakers and have done so almost without exception since 1984, when James Bongiorno brought the circuit to my attention. By the way, Quasi-Second-Order crossovers have been with us since the mid-1950's!!

Phase plugs

Appropriately designed phase plugs are superior to any kind of dustcap, inverted or not. Haven't used dustcaps on our drivers for about 10 years. I hate dustcaps. The phase plugs are precisely the length and bluntness needed to move the woofer acoustic center forward to match the planar sections.

If you're worried about dust in the air gap blast it with compressed air. No problems in the eight years we have been making drivers with phase plugs.

Planars & cones

Cones which are driven from their apex by a single magnet depend on diaphragm rigidity (point A to point B) for pistonic motion. Unfortunately all materials flex, particularly plastic and paper. The result is a series of flexures, ripples, breakup modes etc. which disrupt linearity and worse, store energy.

A planar (electrostatic or magnetic) is driven evenly over its entire surface, and rigidity is not required. The "bowing" motion Dunlavy mentions is certainly of an amplitude no greater than the diaphragm misbehavior of an apex-driven cone, and the idea that such bowing causes reflections, while true, fails to mention that such reflections are microscopic.

I have never understood the attraction of single-ended planar magnetic drives. Everyone can hear 30% THD. However, modern push-pull planar magnetics (ribbons) have low THD, wide bandwidth, excellent linearity and very good sensitivity. They are expensive, however. You have to have the guts the spend the money and still charge a price competitive with the far cheaper cone drivers.

Having compared the best cones to the best planars, my humble opinion is that the latter win hands down, at least from 100 Hz up to beyond audibility.

Crossover tweaking

Wholesale parts replacement in a speaker crossover is not generally a good idea, for several reasons.

First is the value tolerance of the parts. We trim our crossovers to four decimal places and use the same Bennic, Solen and Axon polypropylenes... Even 5% tolerances may swing the crossover hinge point beyond the designer's intended limits. Also, highly colored caps like the MIT's and Hovlands need to be compensated for elsewhere in the circuit. Not that they sound bad, just very distinctive, and you want to accentuate their good qualities while disguising the bad.

Upgrades often make things different, not better. Make sure you can tell the difference before embarking on circuit alterations.

Neodymium

Neodymium has many excellent qualities foremost among which is its lack of a stray field and easy machining into unusual shapes which can focus the magnetic force field into unique patterns. Neo is ten times more expensive than Strontium 5 or 8. However, a ferrite magnet sends about 80% of its energy in the wrong directions, out from the pole piece front and back plus a figure 8 pattern laterally.
I haven't tried the Neodymium ScanSpeak tweeter but their designs are very good and anything which reduces the physical size of the magnet structure is an advantage. I agree the 9700 is perhaps the premier softdome but it is still a softdome with all its problems, just one that is very pleasant to listen to.
In the past 20 years speaker design has struggled with magnetics worse than those available in the 1950's. Once the Neo patents expire and the price drops I think most premium drivers will convert. Aura, a leader in Neodymium magnetics design, recently went bankrupt, probably because of the high cost of their admittedly unusual and innovative motors.

Heil speaker & ribbons

Knowing Dr Heil as I do (he lives in Burlingame, an hour away, is in our SF AES Section, and we have spoken on numerous occasions, including when I visited him at home) I can say that the Heil AMT is an interesting departure from typical ribbon design with both pros and cons from a sonic standpoint.

In the AMT the magnetic field is highly focused on a central air gap where the pleated diaphragm is suspended. This gives a much greater force field and higher sensitivity than is typical of ribbons. The diaphragm motion is unorthodox in that it "squeezes" air in a kind of accordion bellows motion. This is higher acceleration than is typical of ribbons.

The disadvantages of the AMT are, primarily: the pleats tend to slap together when overdriven, producing startling breakup noise; the supporting film has a low melting point, leading to thermal breakdown (attempts made with teflon film backings were less than successful, due to the stiffness of the material); impedance tended to be quite low, due to the short trace; attempts to make bigger diaphragms with lower resonance frequencies ran into difficulties with undamped high frequency resonances caused by the material mass limiting and cavity resonances between the pleats.

The ribbon panels in our Ribbon Monitor series are fundamentally different. They are transformerless and push-pull, with drive magnetics front and back of the diaphragm. They are low resonant, they are high impedance (3 to 6 Ohms) due to the length of the trace. They are not very prone to thermal breakdown, separation of conductor from backing film, or diaphragm noise.

Problems include mass limiting and cavity resonances caused by the magnet structure in front of the diaphragm (the rear cavity is filled with foam), which we control with a simple series 6 dB network. Bass resonance is pronounced, but fairly high Q and controllable with a simple 12 dB crossover with one pole centered on the low frequency resonance and another pole about 1/3 octave below.

T-lines

T lines are not my favorite bass load because of the delay. In one memorable demo I heard, a pulse was introduced into a TL speaker; two distinct pulses were audible as its output.

Anechoic measurements

As a professional speaker designer I will comment briefly on this subject.

Very few designers use anechoic chambers for speaker measurements any more. Most such chambers are not large enough for accurate measurements below 200Hz (the so-called "boundary dependent" region). Many engineers utilize gated computer-processed measurement systems, such as the one I use, Sysid. Sysid generates phase, distortion, transient and amplitude response measurements simultaneously. Most people who refer to "anechoic frequency response" measurements are referring to amplitude alone, which is indeed an inadequate spec that hides more than it reveals. I make my measurements in the near field (about 1/4" away) with a 1" B&K mic and a John Curl custom mic preamp.

I find such measurements most useful in transducer design, rather than system design, since the problems you can measure in a driver can be fixed either by driver parameter adjustment or with crossover filters. System measurement is limited by where you place the microphone. If people listened to their speakers at 1m on axis, such a measurement might be helpful. They don't, and it isn't. Still, this is the most common "anechoic frequency response" measurement.

On the whole I would say it is impossible to design good speakers without accurate measurements. I would also say it is impossible to design good speakers with accurate measurements alone.

The mics in any SPL meter I have used (including B&K) are not flat enough, particularly in the bass, to give an accurate reading, and most signal generators aren't either. It takes very sophisticated equipment to make accurate measurements on speakers. If you really want an indication of how good the speaker is you wish to measure, listen to it full range on familiar music. Better drivers sound cleaner, clearer, faster. Train your ears and leave the test gear to the guys with big bucks and experience. And even they are pretty clueless, most of the time.

Wide range ribbons & full range speakers

Ribbons function in the same environments and follow the same rules as other transducers.
If you want a ribbon to extend into the bass range its suspension must be compliant enough and the moving mass high enough to permit accurate reaponse in that area. This necessarily limits its usefulness in the trebles. The Raven 2 tweeter, which we use crossed over at 6.5 kHz, is an excellent example. Its resonance is much lower but it is comfortable as a tweeter with flat response from 5 khz to about 30 kHz. R3 attempts to responddown to 500Hz but is still a monopole with a chamber behind the diaphragm which is not capacious enough to absorb an energetic 500 Hz backwave. So, the R3 is really a tweeter with a fairly low (500Hz) resonance. Lots of softdome tweeters, for example, have resonances in the 700Hz to 800Hz ranges. Like the R3, their power handling is poor in that range.
If you were to apply a 30W sine wave at 500Hz to the Raven R3 the diaphragm would vaporize. Power handling is a real problem with the R3.

So, there are many reasons why there is no such thing as a fullrange speaker. Our panels with 166Hz to 6.9 kHz come pretty close. In the bass, nothing beats dynamic woofers in a very stable columnar array. Which is what we use.

Time alignment

"Time Alignment" is a trademarked, patented method for allowing the various passbands of a multiway system to reach the ear at relatively the same time. It involves physically advancing the woofer in front of the midrange, and the midrange in front of the tweeter, by amounts appropriate to the time delay inherent in the mass, inertia and reactance of each driver, plus time delay through the filter legs of the crossover. Ed Long originated "Time Alignment" in the mid 1970's and the first commercial application in a high fidelity speaker was the "Paradox" brand which disappeared from the market back then rather quickly. Since the offsets between ranges are fairly large (perhaps 6 to 10 inches between woofer and mid) "Time Alignment" requires a stepped or "pregnant kangaroo" front baffle. This proved cosmetically unacceptable to many audiophiles. The result was the sloped baffle found in many highend multiway speakers. Ed Long told me he always laughs when he sees "time alignment" performed in this manner, since the driver offsets are nowhere near large enough. On an expensive ($14,000pr) famous multiway I examined recently, the 10" woofer was about 2 3/4" acoustically in front of the 5" mid. The correct offset would be about three times that distance.
A sloping or pregnant baffle is in itself no solution. Crossovers must be first order and drivers of the minimum phase configuration. Unfortunately most "first order" designs (like the aforementioned system with the sloping baffle) are burdened with notch filters and comp networks very destructive of phase integrity and dynamics. I find it strange that many such designs must utilize these added filters and networks because they choose lively, poorly damped woofer/midrange diaphragm materials such as aluminum or titantium which ring like a bell.
Finally, many designers place amplitude linearity above all else in their design goals. This is most easily achieved with drivers wired alternatively out of phase, since amplitude response in the crossover region is flattest with this technique. Unfortunately such speakers exhibit a characteristic coloration due to fundamentals being reproduced in one polarity and overtones in the opposite polarity. There are ways to maintain good amplitude linearity in the crossover region without wiring drivers out of phase relative to each other, but all electrically in-phase
filter/driver configurations will measure less flat than when driver polarities are alternatively reversed.

So yes, many designers care about phase response and "time alignment", but only if these do not impede flat amplitude response, particularly through the crossover region. Once this happens phase and time integrity are neglected or destroyed.
My designs are first order, with all drivers electrically in phase, and without comp networks or notch filters. That's they way the sound the best, even though they might not measure as flat as some other speakers of more conventional construction.

You can consider those sloping baffles cosmetic triumphs with little real utility.

Famous ribbons

Magneplanar does NOT make fullrange ribbon speakers. They make single-ended planar dynamic speakers mated to ribbon mid/tweeters. Big difference. Ribbons are not suited to bass frequencies; their low mass works against them. The Apogees were another example of a single-ended planar dynamic mated to treble ribbons. Single-ended speakers are not like single-ended electronics: the distortion is very high, the waveform fidelity low. People like them because they are quick and clean sounding, not noticing that half the waveform is only approximated.

VMPS electronic crossover

There is an electronic crossover. I'm making a power supply for one now. Take sheet metal, punch 15/16" hole for XLR connector. Punch 3/4" hole for AC jack. Sorry, 3/4" hole too small, spend 15 minutes nibbling with Dremel tool to size. Drill 9/16" hole for fuseholder. Sorry, 1/32" too small, Dremel tool time. Drill holes to mount transformer, rectifier, million mic capacitors. Wire, solder, wire, solder. Install surge protection. Bring up slowly on Variac. Allow to stabilize at 15.8V DC output. Screw chassis together. Don't touch anything, that's a fully charged cap in there, lethal. Repeat process for actual crossover. Ugh. Kills a whole day.

I end up with a 24 dB characteristic centered at 108Hz, though I can vary continuously from 50Hz to 190Hz. Dual 10 turn pots allow 0.01dB level changes. You need this kind of precision in order to mate big cone woofers to long ribbons at a low frequency without huge losses. Expensive, time consuming. Don't dare charge too much, hard enough to get the customers to invest in a separate bass amp. Sell xover for price of parts, donate my labor. Everybody happy.

Dispersion

A symphony orchestra radiates about 2 acoustical Watts playing fortissimo. There are speakers that generate that much energy. However, it is HOW the orchestra disperses that energy that makes its sound so different from speaker sound. The orchestra produces a spherical radiation pattern bass to treble that impacts the body head to toe. Speakers at best produce a mismash of patterns that change radically with frequency. Strange that omnidirectional speakers are no better at producing the "pulsating sphere" or "ball of sound" than conventional designs. Oh well, we'll keep working at it. Transduction errors and other losses in the recording chain will forever prevent us from reproducing live sound in a live fashion, but I think we can come pretty close.

Acoustical watts

A technical question such as "what are acoustical watts" only gets you a technical answer.

Acoustic power and intensity of sound are related. The intensity of sound, independent of its frequency, is proportional to the average of the square of the pressure taken over a complete pressure cycle. Intensity is defined as the power in watts that is transmitted across one square centimeter of wavefront, perpendicular to the direction in which the sound is traveling. The power of even the most intense sound express in watts is small: the unit of acoustical intensity is 10 to the minus 16 watts per square centimeter, slightly less intensity than that of the softest 1 kHz tone audible to the human ear. A painfully intense sound has the intensity of about .001 W per square centimeter. The square root of the average square (rms) of the sound pressure that corresponds to 10 to the minus 16 W is 0.0002 dyne per square centimeter. A dyne is the force equal to 1/980th of the weight of a gram. Sound intensity is generally express in 10 to the minus 16 W units, and sound pressure in 0.0002 dyne per square centimeter.

Having cleared that up, I would like to point out that acoustical power and sound pressure have an inverse square relationship, SPL's depending on how far you are from the sound source regardless of how much acoustic power it is generating. In other words, sound pressure falls off rapidly as you move back from the source. Which is why I sit as close to the orchestra as possible, preferably in it.

More ribbons

It is expensive to cross over properly at 166 Hz. We use the Axon polyprops (250V) but bypass them with smaller polyprops and trim them to exact value (that's four decimal places on our B&K cap meter).

We do a lot of things that don't make sense in the price range the RM2 inhabit. The very idea of a $2000/pair speaker with a midrange of that quality (and price: $70 ea wholesale) is absurd. We do it because I wanted a fullrange ribbon people could afford. Maybe later when the model catches on we can double the price to reflect actual parts cost....

AC Quality

One of my first jobs (1968)as an electronics engineer was operating the Quality Control Electronics Lab as a civilian employee of the US Army in Munich Germany. One of my primary activities was monitoring the AC line (220V 50Hz); I even had a 110V 60Hz generator for comparison purposes. Many items of electronic gear purchased from various vendors had a hard time with the AC line obtaining in Europe; I was supposed to weed out the incompetent products before they were purchased in quantity.

European AC is considered high quality but the waveform was always poor. Visually, there appeared to be chunks missing from the sine wave, RFI rode along, symmetry was often impaired.

A regenerator such as the PS300 produces a visually perfect sine wave. There is no comparison. I consistently measured 3% to 10% THD in the European AC line, not counting the nonlinear and nonharmonically related distortions and noise. Paul has published his distortion specs; the output of the PS300 is free of all typical AC line source anomalies. I own one and only wish he had a 5 kW version.

Any device supplying a relatively low amount of AC power will act as a current limiter on the electronics hooked up to it. There is a tradeoff between purity and dynamics, so try out any AC conditioner or regenerator first in your own application and decide if it is worth the price.

The only VMPS electronic product is an electronic crossover which draws less than 10W. It sounds better hooked up to the PS300 than plugged into the wall or my other AC line conditioners (Tice, MIT, Line Rover, Bybee). This inspite of a 6 element pi filter on the AC input of the crossover power supply.

Values

The production budget for a system retailing at $2,000 in the US will run about $200 (US) for the pair. This explains the small woofers and the $4 tweeter. There is no excuse for the electrolytic tweeter cap; a polypropylene of the same value might have cost $3 instead of 80 cents.

There are many ways to design and all have salient points of advantage. I will say that complex crossovers are not necessarily a good idea, since each extra leg or element introduces losses and phase shift. Using better drivers that require fewer corrections is what I prefer. For example,all woofers have a high frequency resonance. With a first-order filter you can place a pole at the resonant frequency and, if its amplitude is 6 dB, knock it down completely. Then of course you have to bring in the driver above, which can also be accomplished at 6 dB (parallel network) or 12 dB (series network, same number of components, still first order, higher slope).

Crossovers

I use Monster Y connectors on the RCA outputs of the processor and use the digital remote volume control. One set of outputs goes to the electronic crossover which controls the bass only. the other set drives the mid/treble amp. The reason there is an electronic (loss-free) crossover to the bass is that it is accomplished at 24 dB per octave to keep the woofers out of the ribbons' operating range. The highpass to the ribbon section is at 99Hz and is passive for two reasons: 1. When done at the high parts level, it sounds better than an electronic crossover; 2. the passive crossover can be trimmed to 1/2000th of 1% tolerance, necessary for best sound but not achievable electronically.

The electronic crossover to the bass amp has plus or minus 9 dB of gain/cut available for precise level matching, and increments are very small, 1/200th of 1 dB per step.

The electronic crossover is single-ended, input and output. It normally is equipped with RCA jacks, but can be ordered with XLR's if you want them. Since your power amps (Plinius 250) have RCA inputs, and your source has RCA outputs, the crossover is correctly configured with RCA jacks.

Motivation

My experience is that the High End exists primarily as expression of company owners' vanity. These outfits never make a profit, go through huge sums of Other Peoples' Money, fail and reform only to repeat the process. In other words a large, costly ego trip readily supported by the audio press, some of which is performing the exact journalistic equivalent for the very same selfish reasons. Or perhaps you think "The Audio Critic" (one example) exists to sell ad space?

Mid fi and consumer audio corporations are out to make a buck; I never met anyone from those quarters who cared much about sound quality or even music. That area is dominated by offshore mass marketeers with all the heart of a large granite slab.

So just what am I doing here? Hopefully advancing the art, taking chances, indulging a taste for high quality and low prices and, I might add, making a living. And there is always the music. I make sure my rig gets the music to me intact; it's my job and my reason for existing. My first setup was a Garrard Lab 80 and a Grundig table radio, which did just fine for the first year or two in this hobby. I built my first speakers at 15 and have done little else ever since. I've gotten better at it. If you find reproduced music lacks soul, maybe the lack is in the listener, not the equipment. Hear some live music and see what effect it has on you, if any. You might study music like I did, or pursue musical genres with which you have no familiarity. The unadventureous never develop the depth to receive the true message of op 131 or early Elvis or Fisher-Dieskau or Louis Armstrong.