The Raven will drive any power amp with an input impedance between 10K to 220K, in either RCA single-ended mode (which uses half the transformer secondary) or XLR balanced mode (which uses the full secondary). The output transformer is optimized to deliver clean square waves over this impedance range.
|
Transformer coupled line amplifiers were the recording, broadcast, and television-audio standard from the Thirties through the Seventies. After that, discrete-transistor op-amps (discrete output transistors have enough power to remain in Class A with 600-ohm loads) became the industry standard.
[A bit of history: when 16 and 24 channels became the industry norm, tube line amps became impractical due to heat and size concerns. The first integrated op-amp specifically designed for professional use was the Signetics 5532/5534 released in 1979, while the discrete-transistor Jensen JE-990 became an industry favorite after 1980.]
The Raven is simply a modernized and fully balanced version of a studio line amplifier from preceding decades, with modern power supplies, computer optimized transformers, Khozmo fully balanced switched-resistor volume controls, and point-to-point wiring.
|
Don and I are not part of Spatial’s pricing decisions. We just design the things and provide technical support, and show up on the forums when people have questions. He’s in British Columbia, Canada, while I’m in Colorado, USA, while the preamps and power amps are built and sold in Salt Lake City, Utah, USA.
Like most other electronic products, the sources are global. It is impossible for any country in the world to make a 100% domestic product if it is electronic ... that’s a pipe dream. What was true in the 1950’s and 1960’s is definitely not true now.
Which means the on-again, off-again tariffs/taxes, or whatever they’re called this week, are disrupting supply chains. In the old days, manufacturers had big warehouses full of product, both for finished goods and raw parts, and that worked well with the high shipping costs back then (before containers and cheap airfreight). Nowadays, there are few warehouses, most electronic products are built on demand, and shipping is quick and simple.
But ... if supply is uncertain, and future pricing is up in the air, everyone in the industry starts to hold back, and adopts a wait-and-see attitude. Parts availability can get spotty. That’s why I strongly urged readers to buy back in December. Products that use international components are (very) unlikely to get any cheaper.
I have zero idea what pricing and availability ... of anything ... will look like six months from now. Nobody knows, from Warren Buffet on down. Supply chains could completely break down if tariff wars erupt. Or maybe people could easily adapt to a 25% premium with the resulting rise in retail prices. Or not. If anyone has a crystal ball, let me know.
|
Electronics in particular are globally sourced. Here’s a simple example: say there’s a circuit board that AtmaSphere, or Spatial, buy in small quantities, say in the low hundreds. That’s a lot of circuit boards for a high-end manufacturer.
The circuit board might be assembled in the USA, Canada, the UK, the EU, Japan, South Korea, Taiwan, China, Vietnam, Thailand, the Phillipines, or many other countries. But that’s the finished board. It contains semiconductors, which come from any of the countries mentioned above. And resistors. And capacitors. And the glass fiber for the board. And the copper in the circuit traces.
From the perspective of AtmaSphere or Spatial, we just see a circuit board that does what we want, and we hope (and pray) we get a timely delivery without getting snagged in customs. Delays are a huge deal to any manufacturer. Every part is a critical part, and parts shortages shut down production, which is very expensive, because it costs money to keep the lights on, pay off capital expenses, and pay for trained staff. Every day without production is a day closer to bankruptcy.
On-again, off-again political posturing about tariffs, quotas, additional customs screening, etc. etc. is calamitous for small manufacturers. The big boys like General Motors and Ford have the capital reserves to ride things out for several months, but the little guys (that’s us, folks) are working on very thin margins with a fickle customer base.
By "fickle" I mean a customer for a Ford F150 is going to buy a full-size pickup truck no matter what. They might switch to a Chevy Silverado, but they’re not going to buy a VW or BMW that cost the same, because those are totally different products. By contrast, a high-end customer can always wait another year or two, and can choose from scores of different products at a wide variety of price points.
I’m not crying in my beer, because what I just described is how the entire high-end business works, and you just have to accept uncertainty as part of the deal. But you wonder why artisan-assembled high-end costs more? That's why.
|
I started as an audio designer at Audionics in 1973, so I’ve seen a few ups and downs in the audio industry. Don’s been doing his thing in Canada since the early Eighties, so between the two of us, we’ve seen what’s been happening in the North American market for quite a while. Players come and go, fads come and go, as well as turnover in big-name reviewers and gradual shifts in consumer tastes.
Will the high-end itself survive, given the aging demographics? Good question. Headphones, and the specialized amps that serve them, are a growth area, since they have a compact footprint and can live on a computer desk as a peripheral. Late-night headphone listening is also well suited to young families living in urban apartments.
Big-rig systems that take up a whole room? That seems to be a declining market, unless the system is domestically acceptable within normal living room decor ... the traditional "stereo system" we saw in the Sixties and Seventies. Swap in a digital streamer for the classic FM Stereo Receiver of the Sixties and Seventies, and it looks pretty much the same.
One thing Don and I find gratifying are a new breed of medium-to-high-efficiency speakers that work well with 25-watt amplifiers. This is a sweet spot in the market that’s been ignored a long time ... moderate power amplifiers that are optimized for listening pleasure, not bombast, and speakers that have low distortion and effortless dynamic range. More, please.
|
That’s the logic of inflation. Buy Now, because it will cost more tomorrow. Once this gets baked into everyone’s consciousness, it affects pricing (and expectations) throughout the entire economy.
Deflation is the reverse, but is far more dangerous, since businesses hate to sell at a loss, so prices are "sticky" in the downward direction. Deflation of much more than 1% is associated with demand collapse and economic free-fall.
The ideal target for stability and maximum steady-state growth is about 1~3% inflation per annum. Shocks in either direction are very bad.
|
Thanks for the review, phono2024! Always good to hear compatibility reports as well, much appreciated.
|
Although Don and I optimized the Raven and Blackbird to deliver their best performance with each other, we went to considerable trouble to make sure they add sonic value to other systems, too. For the Raven, this means optimal performance when matched with classical push-pull pentode amps and modern transistor amps.
Both RCA and XLR outputs offer equal sonic quality, since the only difference are the ratio of the output taps on the output transformer. The circuitry is exactly the same for both outputs, using a direct straight-through path from input to output. There are no "add-ons" like phase inverters to create the minus phase of the XLR output, no cathode followers, no dynamic loads, no current mirrors, no op-amps, nor DC servo circuits. The custom-design transformers solve all these problems at a stroke.
What you hear are point-to-point wires, triode vacuum tubes in a balanced Class A circuit, and studio-quality transformers. Nothing else. There are no coupling caps, no feedback, and no secondary signal paths in the circuit, just straight through from input to output.
|
Don and I are pleased that Raven owners have an equal choice between RCA-input power amps and XLR-input power amps. The Raven treats both the same, so the sonic choice is on an equal footing.
Classic vacuum tube amps almost always have RCA inputs, with phase splitting handled internally. Similarly, low-powered SET amps have RCA inputs because they are single-ended throughout, from input to output.
Older transistor amps from the Sixties through the Nineties typically have RCA inputs, with phase splitting done in a differential first stage with either bipolar transistors or FETs. Contemporary transistor amps may offer both RCA and XLR inputs, but the highest performance option is typically the XLR input.
It’s all the same to the Raven preamp, since all signals pass through the output transformers, which have both RCA and XLR output taps, just like the 4, 8, and 16 ohm taps on a tube power amplifier.
P.S. I should mention in passing the Raven may be used as a signal conditioner, like other "tube interfaces" on the market. It removes both DC offsets and RFI interference from other components and environmental RFI. Transistor amps often have issues with RFI incursion, which might not be as obvious as hearing AM stations in the background, but as an increase in veiling and listening fatigue. What’s actually happening are the input transistors rectifying incoming signals in the MHz (AM, SW, FM) to GHz (WiFi, mobile phones, Bluetooth) range, which then splatters IM distortion across the audio band.
Get rid of the RFI interference, and the amp sounds cleaner and fatigue goes way down. That’s where the studio-quality input transformer of the Raven comes in; it is designed to reject RFI, and preamp tubes themselves are resistant to RFI. The result is that all that comes out of the Raven is clean audio, in the 15 Hz to 60 kHz frequency range, with nothing above or below that. The audio-only feed to the power amplifier reduces the burden on the input transistors (or opamps) of the transistor power amplifier.
|
Another small note: conventional tube preamps can be dangerous for transistor amps unless they have an automatic muting relay with a timer circuit. A conventional tube preamp with cathode follower output typically has the output node (the cathode) somewhere between 150 to 200 volts, with a 1~3 uF capacitor keeping that DC potential out of the output of the preamp (a blocking cap). Unfortunately, as the preamp powers up or is turned off, large transients that are a significant fraction of that 150 volts can appear on the RCA output, which is why a muting relay (with timer delay) is required for safety.
This was harmless during the days of power amps using nothing but tubes, in the Fifties and early Sixties. The power amp would be warming up and unable to transmit the preamp transient to the speaker, and in addition, vacuum tubes are not harmed by high voltages ... they can accept hundreds of volts without risk of arc-over.
Transistor power amps are completely different. The input differential pair, either bipolar or FET, can be destroyed or degraded by pulses larger than 10~20 volts, as well as passing on speaker-breaking pulses since transistor amps turn on instantly. The damage to the input pair can happen even when the power amp is turned off, unless the power amp has input muting relays (the relays are usually on the output side).
The classic tube preamps of the Fifties and Sixties never had muting relays because they weren’t needed ... the tube power amps didn’t mind, and took a while for them to warm up and amplify. Muting relays appeared on high-end gear in the Seventies and later, because users were losing amps and speakers because of these transients. The catch with muting relays is they can introduce a new source of distortion from slow development of corrosion on the relay contacts.
Transformer coupling has the charm that it can never pass DC, under any circumstances, and unlike a capacitor, does not store energy. In addition, the circuit itself is balanced, and is only slightly out of balance as it warms up. Although we do not recommend it, you might only hear a minor "click" when the preamp is turned on with a "live" power amplifier. Doing this is dangerous for your system and we do not recommend it; always turn on preamps before power amps, and always turn off power amps first (and wait a little while) before turning off source components.
Being careless with a conventional tube preamp is quite different. The survival of your system depends on the reliable operation of the muting relay ... for example, if you have a power outage, or a power interruption, does the muting relay instantly activate? It must, if your transistor power amp is going to survive. A tube power amp would give a helluva bang (tough on your speakers) but it would survive just fine.
The output transformer sidesteps all that. No energy storage, and does not pass DC, ever. (There is no metallic connection between primary and secondary, only magnetic.) No requirement for a muting relay. No timing circuits that require a separate 5VDC supply. Worst case scenario, you might hear an annoying click, but no amps are harmed, or speakers destroyed.
|
The most dangerous of all are DC-coupled preamps and power amps. If the DC servo system in a preamp fails, the failure can propagate all the way to the loudspeaker. I saw this happen when I was visiting the home of a reviewer. It took out one channel of the preamp, one power amplifier, and one woofer in a $50,000 speaker ... all in less than a second. I saw a brief flash where the woofer was, the left channel went silent, then smoke wafted out of the power amp and woofer. That’s what DC coupling can do. No thanks.
|
The anecdote is true, not fictional. I was there and saw it happen.
Not surprisingly, the preamp and power amp were all solid-state and were stupendously expensive, in the $100,000 range. As we old-timers know, price does not guarantee quality in our business, unfortunately.
DC servo failure is not rare in the high-end solid-state world. It should be, but isn’t. When it happens, the results are very expensive, not to mention the cost of down time.
Looking back, my guess is poor solder technique somewhere in the preamp, or maybe a regulator in the plus or minus supply gave up, yanking the preamp output to 15V or more. Or a DC servo went nuts. Results were the same, a destroyed power amp and loudspeaker. The only unusual thing was a designer (me) seeing the failure in real time, in a domestic setting. You usually see this kind of drama on a test bench, not in the context of an ultra-deluxe system in a home.
Should it have been better designed? Of course! But how is the consumer to know? They assume [higher price = better product].
That’s the real difference between pro gear and audiophile equipment. The pro stuff is reliable, because it has to be.
|
Oh it blew a fuse all right. After the output section was completely destroyed, and a massive DC pulse killed the woofer. This was European oligarch-class audio gear from a well-known manufacturer.
The manufacturer is still around, twenty years later. The buyers are expected to suck it up and not complain, I guess.
The joke amongst designers of transistor gear is transistors do a great job protecting the fuses (because transistors fail in milliseconds, while fuses can take a half second or longer). Transistors fail much faster than any human response time, whether on a bench or across the listening room. And a good, solid DC pulse will destroy any woofer ... audiophile-grade woofers can only take 1 watt (or less) of DC on the voice coil. Professional 15" woofers with 4" voice coils can take a little more DC, but certainly not 200 watts.
|
Happens all the time in the small-production-run solid-state world. Oversights in manufacturing like no heat-conducting paste between transistors and heat sinks, for example. Or a wave solder machine not properly set up for EU-required leadfree solder, so the solder joints on a circuit board start going bad after a while.
There’s some pretty sketchy tube stuff too ... I remember seeing bulky, heavy electrolytic capacitors "attached" to the chassis with RTV silicone goo ... and no bolts, clamps, or other hardware! From a highly reviewed European manufacturer that is still in business twenty years later. Or a well-known American manufacturer that omitted, or forgot, the required 10~100 uF decoupling/stabilization capacitors for the multiple 3-pin voltage regulators, so they sang happily in the 10 MHz range. Easily visible on an RF spectrum analyzer, but not easily seen on a scope.
|
