Spatial Audio Raven Preamp

I should add that all folded bass horns are tricky beasts. They only exist because a true straight bass horn that is flat to, say, 35 Hz, would be the size of medium-sized car. In other words, the size of an adjacent room. Two needed for stereo, of course.

PWK compromised with the real world by folding the horn (which creates internal reflections) and using the room corners to expand the size of the horn mouth. The internal reflections create ripples in the response above 150 Hz, and the cutoff region has +/- 5 dB ripples in the response, which interact with the room modes.

This is why adding a subwoofer is kind of tricky. You have to integrate not two, but three things: the Khorn response in its cutoff region (which is definitely not flat), the built-in filter of the subwoofer amp, and the room modes. Having two (or more) subwoofers is very useful because the room modes for one subwoofer will be at different frequencies than the other subwoofer, which smoothes out both of them. It’s also why multiple small subwoofers, in widely spaced locations, is a much better choice than a single subwoofer.

I should add the Khorn horn cutoff might be a lot higher than Klipsch says it is. 60~70 Hz would not surprise me.

The Raven will have no trouble driving an RCA output at the same time as XLR.  This has been tested by Spatial.  You should have no problem except that the XLR output is higher level than the RCA.  That is as it should be.  If your subwoofer amps have gain controls you should be able to match levels.  Subs are always tricky.  The other option is speaker level outputs to drive stereo subs near the main speakers.  That is the way I have always done it, but thankfully, my current speakers easily play 30 Hz in room so I no longer use subs.

The degree of overlap between the horns and the subs (at least two) will be entirely subjective. Unlike closed and vented boxes, horns do not have a smooth, predictable cutoff region (which is 12 dB/octave for all closed boxes, and 24 dB/octave for vented boxes). Instead, they drop like a stone, and the octave just above cutoff can be pretty rough as well.

Combining the horn with the subwoofer will require judicious use of the "phase" control on the subwoofer plate amplifier, and messing around with the lowpass filter. The magic spot might be anywhere from 30 to 60 Hz.

The Raven has split windings on the output transformer so both RCA and XLR’s can be used at the same time (the RCA output uses one-half of the secondary winding). So feel free to use to use the RCA output to power the active subs.

However ... the added cable capacitance on one-half of the winding, might, in principle, unbalance the XLR output a little bit. Maybe. Most likely not. Use low capacitance cable for the sub output, if possible.

Mostly, try it and see. The Blackbird power amps have pretty good common-mode rejection, so unbalance gets washed out in the input stage.

I keep hearing good things about the Rhythmik servo subwoofers, so you might check out getting a pair (using stereo bass, with the lowpass filter set to 40 Hz).

Although horn speakers have gratifyingly low IM distortion, they drop off very fast below cutoff. It's basically a brickwall cutoff, so "pushing" them below cutoff has limited utility.

was just thinking sub(s) and your post said it better. I'm not sure how to attach a sub(s) to the Raven Pre? Must be an easy way? 

@wsrrsw It appears there is a single-ended output in parallel with the balanced output. You'll have to check to see how this works but it might be quite simple- the main amps on the balanced outputs and the subs connected to the RCA outputs. I would expect to see a switch that grounds pin 3 to pin 1 so the RCA connections will work properly but they may not be connected to the output transformer driving the XLRs, negating the need for a switch.

@donsachs Yahsureyoubetcha I'm using Hijiri XLR's from a LuminX1 streamer into the Raven PRE and Cardas Clear Beyond XLR's from the pre into the Blackbird Monoblocks. 

I boosted the base a bit more w/ the roon parametric equalizer at 20 and 40 hertz.

Rather than take the pre and monoblocks 200 feet or so to the other speakers up and over hill and dale, I used the Lumin AMP w/ X1 streamer and compared that to the Blackbird Monoblocks and the Raven Preamplifier. Apples to apples; heck no. The Lumin AMP is 160W per side into 8Ω. It's a class A amp. The tubes vs solid state sounded too close to call. Take that SS.

@atmasphereI was just thinking sub(s) and your post said it better. I'm not sure how to attach a sub(s) to the Raven Pre? Must be an easy way? 

T H A N K you both.

I would be really interested to hear any reactions/reviews of anyone using the Raven preamp with the praised Class D monoblock amps such as the Atma-Spheres or the AGDs. 

@wsrrsw You have an ideal situation for using subwoofers. You might look into something called a Distributed Bass Array which is the use of 4 subs placed asymmetrically in the room so as to eliminate bass nulls and peaks in the room. This is often particularly noticeable at the listening position if the room has regular dimensions.

My main speakers are flat to 20Hz yet I have no deep bass at the listening position in my room. So I use a pair of subs to break up the standing waves that cause this problem. One is to my left and the other slightly to the right and behind me. Both are as close to the wall as I can get them.

In your case since your speakers don't go much below about 45-50Hz, you are an ideal candidate for all 4 subs. The Swarm from Audiokinesis is an ideal candidate for this since they are meant to be placed directly against a wall so as to be out of the way as possible. The subs are designed to roll of at 3dB per octave starting at 100Hz going down. The room boundary effect provides a +3dB/octave boost, so they are flat to 20Hz. They are 1 foot square and 2 feet high. I have the 10" drivers actually facing the wall so as to maximize the room boundary effect.

The idea here is to run them no higher than about 45-50 Hz, at which point its very easy to get them to blend seamlessly with your Klipsch speakers since below 80Hz in most rooms the bass is already entirely reverberant before your ears can sort out the bass notes being played. As long as the are not allowed to go too high they won't attract attention to themselves- the main speakers will convince you the bass is coming from them via the harmonics of the bass instruments.

It will be easy to get this setup to integrate with the Spatial amps and preamp. Then you will hear how much bass is really there and not only that, you may find that the mids and highs get more relaxed and detailed due to how our ears alter the tonality we perceive when part of the spectrum is missing. So getting the bass right calms down the mids and highs.

@wsrrsw

Also, yes, in my experience the streamer has a huge effect on things.  So play around.  The Raven will happily use RCA inputs, but your best performance will most likely be with a DAC that has XLR outputs.  Then you have an entirely balanced system through the amps all the way to the speaker.

@wsrrsw

It will not be the amps unless something is wrong with them or you have a bad tube. The easy test is to move them to your other system. Your Sonus are 86 dB, 4 ohm. Not the best load, but the blackbirds will drive them and that will tell you if the amp has issues. Khorns have limited bass below about 45-50 Hz. It falls off like a stone. They are what they are..... They do other things quite well. The crossovers can be upgraded if you have not, but the physics of the speaker in that cabinet are what they are.. the spec is 33 Hz, but that is - 4dB. It is an incredibly easy load for the amps. If you have the patience, put the preamp and amps in front of your other speakers and see if you are satisfied with the bass. I suspect so. If not, then you may have a tube that didn’t ship well. My guess is it is a combo of the Khorn and the room they are in.

The first thing people say when they hear the Raven and Blackbirds is how amazing the bass is.... and your speakers are trivial to drive.  The amps hardly know they are connected.

@donsachs @lynn_olson The Blackbird Monoblocks and Raven Pre have over thirty hours on them and how sweet they are. Great job gents.

But darn it Don right you are about the Khorns and lower base. Running Roon I have even gone into the paramedic equalizer (a sin?) and boosted the lower hertz but still compared to my other system w/ a pair of Sonus Faber Guarneri Evolution’s speakers (faster sounding than Khorns) the base is flat verses the aliveness and vibration of a nice walking bass line. Low piano notes should vibrate too. “Grandmama's Blues” by Cyrus Chestnut on the "Earth Stories” album is a great test track https://open.qobuz.com/track/2003813  or even “Ham Hocks and Cabbage” by the Christian McBride Trio https://open.qobuz.com/track/10462550. Lower Tom drums also not quite there.

The low mids..heck all the rest (room still not treated and it’s boomy ) get ‘er done. Your gear is great and my contact over at Spacial has provided great service.

I have to consider if’s my streamer adding/subtracting to the bass quandary. I’ll swap out my other streamer and dac.

Good cores are essential for low distortion.

Capacitive coupling exists between windings regardless of the quality of the transformer. For this reason alone its important for the transformer to be loaded correctly so it expresses its winding ratio without inter-winding capacitance as a parallel element. Otherwise the frequency response won't be flat.

To further the transformer discussion.  We don't need to name companies, but many take a sort of swiss army knife approach with multiple windings that allow for many connection schemes.  Sometimes they work, sometimes they don't.  But they are not optimized for the task.  I auditioned several with exotic core materials and found them to be a bit tipped up in the treble to give hyper detail, but they didn't sound right.  Of course, they look fine on the scope.  Interstage transformers from well known companies would ring at 15KHz or so, or others would run out of steam at 50 Hz and below in our circuit.  We are not talking Hammonds and Edcors here, but very well respected higher end manufacturers.   Our preamp output and amplifier interstage transformers are custom wound and use interleaved laminations of nickel and steel in a proportion that gives very even frequency response and hits the sweet spot balancing detail with proper tonality.  They have full frequency response without any oscillation.  This is based on years of experience by Dave Geren, the designer at Cinemag.  Dave tells me there are only a few sources in the world who can make the cores to his specification.   So a lot of design experience has gone into the transformers used in the 300b project, and as Lynn noted, several iterations for each transformer to get it right.  

I used to think that rules-of-thumb applied to transformers. Well, no. Maybe in the Fifties when they were designed with slide rules, but now you work closely with the designer and their simulation software, followed by a sample build. Model, build, put on the test bench, send the complete set to Don, he re-measures and auditions, and repeat as often as necessary.

The Raven is now on the fifth version of the custom transformer set. As noted by Don, they are optimized for the 6SN7 in balanced mode and the most common range of loads presented by solid-state and vacuum tube power amplifiers.

The secret of transformer-coupled audio design is a close working relationship with the transformer designer. In effect, the circuit is co-designed with the transformer designer.

One of the great things about working with Don is he has good working relationships with key vendors, so we can get custom designs on a timely basis, and we know exactly how they were designed. This lets us further optimize our circuits around the custom parts, rather than work around an off-the-shelf part.

@donsachs Is this meant to answer my question?

The transformers in the Raven preamp and Blackbird amps are all custom designed to work perfectly with the tube sets chosen.  They don't require loading resistors, and there is no need for a grid resistor on any of the tubes.  The number of parts in the direct signal path is very low and all are of very high quality.  You cannot get an off the shelf transformer to achieve these specs or operate that way.  You have to work with a very talented transformer designer to get what you need.

It might sound odd, but both the Geshelli and the Raven have a similar parts cost to retail price structure. Solid-state isn’t really that expensive if you buy the parts in hundreds to thousands quantities. And the expensive, heavy case with the 1/4" front panel adds nothing to the sonics ... that’s 100% marketing.

Tube circuits, though, get expensive very fast. Top of the line transformers are NOT cheap. Precision regulated 300 to 450 volt supplies aren’t cheap either. Tubes, by themselves, are kind of mid-price, but again, the top-of-the-line models aren’t cheap either. Vacuum tubes have always been handmade, even in their heyday in the late Fifties. Same for transformers.

None of this should be surprising. The majority of the solid-state market use parts that are made in quantities of hundreds of thousands to millions. This drives down costs. By contrast, the tube sector uses parts that are made in quantities of hundreds to a few thousand, several orders of magnitude smaller than the solid-state sector.

I own a Geshelli DAC and their new balanced headphone amplifier. Although value-priced, both products use the Sparkos (made in Colorado!) discrete op-amps, with a very large Class A operating region. If you’re going solid-state, this is an attractive approach.

The Raven is kind of the opposite: Zero feedback, all-triode balanced, using some of the lowest distortion vacuum tubes ever made (6SN7). What makes it possible are the custom-designed input and output transformers ... before that, Don and I were limited to off-the-shelf transformers, which in turn limited the selection of vacuum tubes that would be compatible with the circuit. The original Raven, designed way back in 1998, used tubes in the 5687/7044 family. These are very good, but are not as linear as 6SN7 family, and the subjective difference is pretty noticeable.

The zero feedback approach brutally exposes the sonics of every single part in the signal path, right down to the sonics of the volume control. Our custom transformers have outstanding sonics (subjectively), and a flat response to 30 kHz in-circuit. The Raven has a signal path of: a discrete-resistor volume control, wire, a balanced 6SN7 dual triode, and a custom output transformer. The unbalanced RCA inputs have their own custom input transformer for phase-splitting and ground isolation.

Zero feedback vacuum tube circuits are not for everyone. If tubes make you nervous, I can recommend the Geshelli with Sparkos discrete op-amps (if the low price doesn’t offend you). If you are very price sensitive, there are lots of Shenzen-made products for $200 or less, with ESS converters and pretty decent op-amps.

lynn_olson

Unfortunately, opamps are limited in not being able to dissipate much heat due to the small package size.

That's why so many good balanced components use discrete circuits for the operational amplifiers.

Unfortunately, opamps are limited in not being able to dissipate much heat due to the small package size. Very few opamps are designed to be used with heat sinks. So the only way to keep heat emission low is efficient Class AB output stages, relying on feedback to linearize the crossover region (opamps typically have very high feedback). Higher powered transistor and tube amps also use Class AB to keep heat emission to acceptable levels, at the expense of higher distortion in the Class AB transition region.

AB amplifiers have no problem with the crossover region and do not rely on feedback to sort this problem out. Its a matter of the output devices being properly biased rather than anything to do with feedback. Of course, feedback is helpful (if applied properly) to improve the overall linearity of an A or AB amplifier. As Norman Crowhurst indirectly pointed out, most of the time feedback is incorrectly applied.

@lynn_olson Since you are using an output transformer, there is an ideal load for which the transformer is optimized. As you know, transformers transform impedance and  this goes both ways, so if the load at the output is higher, the load on the output tubes is higher too. But transformers can ring if the load is too high and conversely, roll off if the load is too low. Traditionally, balanced line transformers are designed for low impedance operation (in the old days 600 Ohms, hence the dBm rating). Ampex provided a switch with a 600 Ohm resistor on the back of their tube tape electronics so if the electronics driving something with a high impedance input, the load could be provided to prevent the transformer ringing.

So I'm very curious how you handled this issue- did you design the transformer for a low impedance and simply installed a loading resistor?

I saw a comment from you about no servos in the context of noise which I assume was a typo since as you know, servos do not introduce noise. You might consider one, since your circuit is direct coupled from input to output before the transformer. Offsets (possibly at the input) can result in an sizable imbalance which introduces distortion, even if you run matched tubes. A servo can be easily used to correct this issue, allowing the output transformer do have its lowest distortion and they are inexpensive to implement.

@jc4659  I know Spatial runs the tubes at least 20 hours before they send things out, just to weed out any failures.   A 6SN7 is pretty well run in by 20 hours.  Yes, there are subtle changes in them out to 50-100 hours, but most of the sound is there in 20 or so.  What you will hear is the cathode bypass caps running in.   They are very large film caps and they are idling along at literally about 2% of their voltage rating.  So they take a while to settle down.  When they do you get that last 10-20%.   Where the real magic lives if the rest of your system can present it.

@donsachs Soundstage depth, image depth and pinpoint imaging is already better than the Ayre which itself was an improvement over the K-5xeMP in these qualities. 

@jc4659  Glad it arrived safely and fired right up.  Thanks for your initial impressions!  It will become more 3D, with more sound stage depth and pinpoint imaging as it runs in for another 50-100 hours.  Tone colours will become more vivid as well.  I am sure the Ayre is a very nice preamp, but it should sound a little flat and two dimensional compared to the Raven, once it has run in.  That eery sense of space will develop on the Raven over the next week or two.  We call it the "trippy" sound:)

My Raven preamplifier from Spatial Audio Labs arrived today and I thought I would post initial impressions.  More folks should be receiving theirs any day now so additional comments from them will be helpful.  The packaging was first rate and the box itself looked like it had been hand carried from the next town over.  Everything was pristine inside the double box and the finish on the preamp was perfect.  The only change in my system today was swapping my Ayre K-1xe preamp for the new Raven.  Everything else is the same.  The Raven has a cherry base and I upgraded the standard 6SN7 tubes to the Linlai elite version.  I let the preamp idle for an hour before listening to female vocals, then instrumental, then male vocals, female, etc... After only a few minutes listening the word that struck me first was REFINEMENT. There was no grain as one might expect with only one hour burn in.  Initially, I was pleasantly surprised how similar the Raven was to the Ayre; clearly better but not by a wide margin.  However, the gap began to widen over the next several hours.  Music was fluid, clear (like a sense of zero distortion), natural, well balanced top to bottom, and became more beautiful and emotionally engaging with every song I played.  I am astounded how good this thing is after only 3 hours.  I will provide more detail next week after spending more time listening but there is no question in my mind that this is an amazing preamp and I will be keeping mine.  I look forward to getting 20+ hours on it and reading the impressions of other new owners.

RFI = Radio Frequency Interference

EMI = Electromagnetic Interference (includes magnetic fields)

15.75 kHz (or close to it) is the horizontal scanning rate of 525/60 NTSC (color or monochrome) analog television. The 625/50 PAL or SECAM rates are similar. Analog television environments were notorious for high interference levels, as well as electrical noise from early SCR light dimmers for on-set illumination.

Modern television is digital from camera, to signal processing, to transmission or storage, to decoding and display. No more sync noise, just computer hash at MHz frequencies.

The Raven accomplishes several things at once:

1) Moderate voltage amplification (from the 6SN7).

2) Substantial current multiplication (from the internal step-down transformer).

3) Signal conversion from either RCA or XLR to RCA, XLR, and headphone outputs.

4) Volume control via stepped resistor array, with L/R balance control on the remote control, as well as volume and input selection.

5) Signal conditioning, with removal of DC offsets *and* RFI interference, and breaking of ground loops between components (via transformer coupling).

So it’s not just a preamp or passive volume control. These benefits extend to all types of power amplifiers ... Class A or Class AB transistor, Class D Mosfet or GanFET, or triode or pentode tube amplifiers.

RFI break-in is the bane of modern hifi gear, since most homes are bathed in microwave signals from WiFi, Bluetooth, and RFI noise from multiple switching supplies in TVs, computers, various gizmos that use ARM processors, etc. etc. Just scraping off all this RFI cruft before it gets to an analog circuit can make quite a difference in low-level sonics ... no more barely-audible buzz or hash getting into the power amplifier.

The classic tube preamps of the Fifties and Sixties were designed at a time when nearly all homes were RF silent. No computers, WiFi, Bluetooth, or switching supplies. No wall-warts. None of that. The only RF-noisy places were TV studios (15.75 kHz TV sync buzz is everywhere), AM and FM transmitters, microwave relay towers, or military installations ... where isolation transformers were routinely used to isolate and suppress RFI incursion into audio signal paths. We are applying the same isolation technology used back then, with custom transformers that are designed with modern computer modeling software.

@brbrock It is a cumulative thing.  Especially if your source is a DAC with balanced output.  The Raven will happily communicate with all things RCA on both input and output.  The gain is slightly less because XLR is a 5V swing and RCA is at most 2.5V.  But the Raven will work fine with anything.  That said, it can be in the middle of a fully balanced system, which has advantages.   I am pretty sure Spatial has a 45 day return policy with little or no restocking fee.  So you would just be out the shipping to try one.

I would say if your SS amp can be driven to clipping by 1.5 V or less, you will be just fine.

@donsachs @lynn_olson I have been looking at maybe purchasing the  preamp for a couple months and have a few questions. I have a nice SS state amp but it is not balanced or have zero feedback.  What will be the benefit of the Raven Preamp if  the amplifier is not balanced or has zero feedback.  Does this negate all the benefits of the Raven or is it a cumulative thing?  

The Raven requires a 6sn7 tube on each channel that has matched sections.  Really, a difference of 1 mA is no big deal, but tubes with wildly mismatched sections will not have good bass response.  Any modern 6sn7 you buy will have sections within 1 mA of each other.  The two channels don't have to be matched to each other, just the tube on each channel needs proper section matching within that tube.   The Blackbird amps require similar matching between pairs. 1 or 2 mA difference is fine.  Any more than that and bass response will suffer.  Of course both units are supplied with matched tubes and it is not hard to buy them.  

If you want to run NOS tubes or ANOS, you should ensure the matching is as above.  Smoke won't come out of the preamp or amps, they will work with mismatched tubes, but they won't sound as good.

Both the preamp and power amp are so symmetric we have to take extra care during assembly to make sure the phase is correct at the output. Multi-color wire comes in handy here.

For that matter, the circuit is so robust it still amplifies with one side non-functional. Distortion is higher, of course, so it’s another thing to be checked during assembly and test. Both phases need to be present and accounted for, polarity correct in every stage, and both channels pair-matched.

Thank you, Don. 

@marantz2270 

No.  The Revelation preamp and the Blackbird amps both preserve phase.  So just hook them up straight.

Is it recommended to reverse polarity at the speaker terminals with the Revelation?

What Lynn said is very audible.  Both the Raven preamp and Blackbird amps use this approach to power supply and balanced circuits.   Once you start listening to circuits built and powered this way it is very hard to go back to conventional approaches because they sound just a bit cloudy or muddy.  It is like a veil being lifted.  The constant merry go round of trying different coupling caps and other things to color the sound in a way you prefer comes to an end.  Instead, once you understand what is going on, you spend a year or two eliminating every bottleneck you can so that the circuit can perform at its best.   What becomes evident is that the circuit and the approach are incredibly transparent.  If you change anything that supports it, you hear it instantly.  Cloud, the main tech at Spatial made the same comment.  You can instantly hear any subtle change you make.  The type of wire becomes very noticeable.  Tube choices are very audible.  Of course you hear these things with other more conventional gear, but not to the extent you do with this circuit and power supply architecture.

Obviously, there are lots of very nice preamps and amps in the world that sound very good.  I used to make some of them myself.  But they don't sound like this.  When you eliminate a lot of the "grunge" that you didn't even know was there, you get a very spacious and airy sound, with incredible detail that you have never quite experienced before.  That is what I hear, and most others who have heard it have made similar comments.  It is not so much about what the circuit sounds like, but rather what the music sounds like when you eliminate a lot of the coloration and distortion that you were never really aware of.  For example, we touched on the idea that there is very subtle spatial information in the signal that is partially obscured by other circuits.  These things are hard to measure, but they can certainly be heard.

I understand that two more preamps have just shipped, so we should get some more reports from owners here fairly soon.  I know it is hard for people because you cannot just go to a dealer and hear the gear, and there are only a few of these in the world, and most are prototype versions.   The production versions are entering the market now, so when people post, the rest of you will get a better impression of how the preamp sounds in a variety of systems to a variety of ears.

I should go into regulators and their sonics a little. Yes, regulators have "a sound". Regulators are amplifiers that feed amplifiers, with the difference the "amplifier" amplifies incoming audio, while a regulator amplifies a DC reference voltage. But it’s an amplifier nonetheless.

Most "linear" type regulators use an internal servo feedback loop to maintain a steady output voltage ... a regulator basically simulates a perfect battery, using feedback to get as close as possible to the ideal. But ... that is an approximation, not the real thing. There are very slight delays responding to a change in current demand, and that is where coloration enter into the sound.

Some audio amplifying circuits have a steady current demand on the supply, and others bounce up and down, following the audio signal. A single-ended audio amplifier, whether tube or transistor, will have a current demand that mirrors the audio. You could put a current sense probe on the supply rails and hear perfectly good music (along with some buzz).

A Class AB amplifier, by contrast, will have quite distorted music on the power supply rails, because it is switching between (B) the upper device, (A) both devices at once, and (B) the lower device. This changes the efficiency of the output stage as the different operating modes change with the music. The switchover between modes can either be hard or soft, depending how the amplifier is biased and how the devices enter the AB cutoff region.

When the load is a Class AB device (like an output stage or an opamp), great demands are placed on the regulator. If it is not a perfect regulator (instantaneous and distortionless), coloration enters the picture. This is why regulators sound different, because a nonlinear load (such as Class AB) then exposes nonlinearities in the regulator.

A balanced Class A amplifier has the great advantage that the load looks pretty much like a resistor at all times, short of heavy clipping. By contrast, the load of a single-ended stage looks like the music it is playing, always varying, while Class AB is quite distorted thanks to a pair of devices switching on and off as the music goes through it. Only well-balanced Class A has a steady draw that doesn’t vary with the music, whether loud or soft, all the way down to zero.

Unfortunately, opamps are limited in not being able to dissipate much heat due to the small package size. Very few opamps are designed to be used with heat sinks. So the only way to keep heat emission low is efficient Class AB output stages, relying on feedback to linearize the crossover region (opamps typically have very high feedback). Higher powered transistor and tube amps also use Class AB to keep heat emission to acceptable levels, at the expense of higher distortion in the Class AB transition region.

The nonlinear load challenges the regulator design, and regulators for the output stage of transistor and tube power amps can be as large and heavy as the output stage they are powering. In effect, one amplifier driving another. This is why it is very rare for medium or high power transistor or tube amps to have regulated output stages. Usually they have a simple lowpass filter with no regulation, saving a great deal of cost and weight compared to the regulated alternative. With no regulation, the sound will always change, depending on the incoming voltage fluctuations, the AC waveshape, and the noise riding on top of the AC power.

The rigorous solution is fully balanced Class A operation for every stage of amplification, not just one or two, and low-noise precision regulators for each of those stages. This keeps the workload of each regulator to a minimum, and the current draw on each regulator is constant regardless of audio signal. It also maximizes isolation between the AC power line and the incoming audio signal.

The Raven also uses an isolation and phase splitting transformer for unbalanced RCA inputs, while balanced signals go straight to the 6SN7 tube grids. Regardless of the incoming signal, whether balanced or unbalanced, the stepped-resistor volume control and internal electronics are always in Class A balanced mode.

I’ve been using damper diodes (from old TVs) since 1997. I’m frankly surprised why people are still using the audiophile favorites. Damper diodes have (much) quieter switching, have substantially higher peak current, and sound noticeably better. The only downside is they consume a lot of heater current and require 6.3V heaters. The majority of damper diodes also use unusual sockets, so they are not pin-interchangeable with the standard types.

The parameter I care most about is smoothness of switching. This is hard to get right, with most vacuum diodes having sort of a Class AB switching region. This can be examined by using a scope probe with a 100X internal attenuator and a safety rating of 1kV or better, connected to the secondary of the power transformer. Voltages are very high, so great caution should be exercised while making the measurements.

The worst diodes have a rough transition between 0 and 50 volts, with holes chewed out of the waveform (generic solid-state bridge). The OK quality ones are fairly smooth but the zero crossing region (measured at the power transformer secondary with special probes) is quite obvious, with small variations between the usual audiophile favorites (which is where the famous tone color comes from). The best diodes almost look like Class A triode, with very smooth transitions that are complementary. Only damper diodes do that. They also have peak current capability that is 2X to 5X higher than any standard 5V heater diode.

With skill, snubber circuits tuned to the transformer, HEXFREDs and high-voltage Schottky’s can approach damper diodes in quietness, which makes them useful for power amps that have to handle a lot of power.

Why the obsession with switching noise? It’s much easier to reduce noise at the source then attempt to filter it later. The harmonics from the 100/120 Hz switch noise sneaks past regulators, magnetically induces noise in nearby circuits, and radiates back out the power cable. Better to minimize it right at the source, which is a function of the transfer curve as the diode is switched on and off.

At Spatial, we use a belt-and-suspenders approach to power supply design. We select the quietest diodes for the application, use CLC filtering as a pre-filter, then apply that to a precision regulator with 130 dB of noise rejection. The regulated output is then applied to a balanced audio circuit with another 35 dB of noise rejection (due to inherent balance). The servo circuit in the regulator has very little to do since the current draw from the audio circuit is very nearly constant, thanks again to the inherent balance of the audio circuit.

@charliee

The damper diodes are extremely tough, they are very quiet, and we are running them VERY conservatively. They have tons of headroom and will last a long time. That, and you can buy lots of them for very little money. So a customer can buy a couple of sets for the price of one decent quality conventional rectifier, and FAR less than an NOS Mullard GZ34, or even new production high quality 5U4 or 274b types. Truth is the damper diodes sound better anyway. Also, in this preamp you can use the 6W4 and there tons of those out there as well for under $10 per tube.

My previous preamp sounded best with the 6BY5, a dual damper diode. So that was my first experience listening to them a lot in a preamp circuit.

@donsachs 

I'm looking to purchase the pre within the near future, it seems to be the top option in my price range. Just looking at the Raven on the Spatial site I saw that it uses two 6AX4 rectifiers. If you don't mind me asking, why that rectifier and not one of the units that are popular with tube rollers, such as the 5U4G, 274B, or the GZ34?Would they not work with the design, or too expensive, etc?  Thanks.

@jc4659   Well, it's ok to post an initial impression:)  But then please follow up with the 20+ hour report.  You get most of it within 20-30 hours, but there are subtle improvements out to 100 or so as the big cathode bypass caps run in.  Most of it is there by 20-30 or so though...   Enjoy.  Hope it works well in your system.

Heard from Spatial that my Raven should ship tomorrow.  I am very excited to hear this preamp in my system.  I will try to refrain from posting initial impressions until adequate burn-in (20+ hours).

I have word from Spatial that the next batch of 5 Ravens is in production, so those folks awaiting builds should have them pretty soon.  Then some of you can post your impressions here!  It will be good to know how people like the preamps in a variety of systems.

Indeed the signal path on both the preamp and amp is very simple.  That said, the power supplies supporting it are rather complex and they are kept well away from it.  Also, the preamp output transformer and amplifier interstage transformers took over a year of prototyping to get right.  So the circuits are "deceptively" simple, but there is a lot supporting them, and as Lynn stated, these circuits are kept well away from the signal path, both electrically and physically.

Also, if you spend time reading this thread and the very long 300b lovers thread, you will see that the "deceptively simple" circuits are also cleverly designed to eliminate distortion at its source as well.

Part of the reason for the sonics (aside from the circuit) is the physical simplicity of the preamp. No circuit board is needed because there really isn’t that much to the audio signal path. Input selector -> balanced switched-resistor volume control -> balanced 6SN7 vacuum tube -> output transformer. That’s it.

There are no coupling caps, no multi-transistor current sources with Zener-diode references, no muting relays with time-delay logic, and no DC-balance servo circuits ... so there’s no need for circuit boards to contain all these secondary functions. Just very short point-to-point wiring.

The same is true of the Blackbird power amp, as well as the Raven. The audio path is surprisingly simple.

Wonderful! Have my eye on the preamp! 

Let's just say that before building my own tube creations I restored probably 300-400 pieces of vintage tube gear, including about 75-80 citation II amps, over 40 citation I preamps, and many Scotts, Fishers, Sherwoods, Macs, Marantz 8b, etc... I saw what were clearly high quality build practices, and cheap ones too.  I always appreciated the gear that was easy to work on, and hated designs that were difficult to rebuild.   So everything we build is designed so that it is easy to work on should anything ever go wrong.  It is designed so that things are never run anywhere "near the ragged edge".   Tubes are in class A, but still conservatively run and should have long lives.  I want the customer to enjoy the gear for years, and should there ever be a problem, I want the tech to easily be able to swap out a part and have it running again. 

I remember working on Marshall guitar amps for musician friends.   They have a dozen little pcbs, tied together with jumpers cables that all have the same terminals, so that you have to mark each board and cable so you know how to put it together again.  The problem would always be on the bottom board!   So we avoid construction like that.  This gear is a bit complex, but very modular, the layout is neat and clean, and it is designed to be trouble-free.  For example, all AC is on one side of an internal shielding bracket, and all signal path on the other in both preamp and amps.  Star grounding, with strong attention paid to current loops.  No hum!

Impressive build and handling of the important details. Thank you for such a detailed answer. I’m very impressed with the quality and design of these products as well as the transparency and comprehensiveness of the responses. You and Larry are to be commended. Thank you.

@grannyring The preamp, and the matching amps are point to point wired, yes. The tube sockets are all mounted directly to the top panel in both pieces. The only pcbs are there to hold large film capacitors to facilitate mounting them securely. The regulated power supplies, both high voltage and tube filament, are built on small pcbs, which are point to point wired to the rest of the components. Again, all signal path wiring is point to point with very high quality wire. The power supply wiring is all teflon insulated mil spec copper. Signal path is all copper as well, but there is no teflon insulation in the signal path. There are NO electrolytic capacitors except in the regulated DC filament supplies. The entire high voltage B+ supplies in both preamp and amps are built wtih all film capactitors, all resistors are wire wound.

All parts are chosen so that they are run very conservatively. No part exceeds 60% or so of its voltage or current or temperature rating. For example, the DC filament supplies can deliver 3A of current and in the preamp they provide 1.2A to the pair of 6SN7 tubes. Resistors are run at no more than about half of their power rating anywhere in the circuits. The preamp and amps are designed to last a long time and to survive the odd tube failure without damage to the unit.   You can put your hand on the power transformer of the preamp or amplifiers after hours of use. Even though both circuits are class A, they run at reasonable temperatures and do not stress their transformers at all.

This question may have been covered, but I have not seen it.  Is the preamp part to part wired? Combo of circuit boards and PTP wiring? Are the tube sockets mounted to circuit boards? 

Most interesting product and thread.  
 

As Don mentioned, avoid the thick and crazy-expensive audiophile cables. You have to remember most audiophiles have noticeably colored transistor gear, and use aftermarket products like power conditioners and $5000+ cables to minimize glare, grain, and excessive HF output in their systems. The better approach is using low-coloration electronics and loudspeakers with a smooth response, especially above 2 kHz.

The key spec in any cable, much more important than any other, is capacitance per foot (or meter). Capacitance should be well under 100 pF/foot, preferably much, much less. Inductance *does not* matter unless you are running an AM transmitter (those are RF cables). Inductance does not load down the preamp, but capacitance does, occasionally causing transient instability in a preamp with high feedback. The Raven has zero feedback, but solid-state preamps typically have very high amounts of feedback (40 dB or more), resulting in load sensitivity to the preamp/amplifier interconnect.

The quality of the insulator (dielectric) in the cable also affects the sound, and Teflon is not necessarily the best. It’s the first choice for aerospace applications, and has exceptional DF and DF measurements, but in my experience, may not the best for audio. Various types of plastic all have their own colorations, and the process of fabricating the cable applies mechanical stress to the plastic, which changes its dielectric properties. The more complex the construction, the more complex the coloration, and the longer it takes to settle down (possibly never). Most of all, DO NOT TRUST the reviews you see in magazines or on the Internet.

Thank you for the advice Don and Lynn. The Paul anticables do sound enticing.