300b lovers

The even-order distortion of the driver section is cancelled in the primary of the interstage transformer, reducing driver distortion even further.

@lynn_olson Actually if the circuit is fully balanced/differential from input to output, even orders are cancelled at every stage along the way. In this manner distortion is compounded less from stage to stage.

The result is the 3rd harmonic is the dominant distortion product rather than the 2nd. Many people do not realize that the ear treats the 3rd in much the same way as the 2nd; its the only odd ordered harmonic that is musical to the ear. The 3rd is very good as masking higher ordered harmonics.

BTW, any properly functioning analog tape machine will have a 3rd harmonic as its dominant distortion product as the tape approaches saturation.

Mathematically this type of distortion can be described as a 'cubic non-linearity' as opposed to the 'quadratic non-linearity' of an SET. As Daniel Cheever pointed out in his paper from 1989, its important that the harmonics fall off on an exponential curve. Both an SET and a fully balanced amplifier can do this (its part of the reason people regard SETs as musical despite their many failings). The advantage of a balanced circuit is harmonics fall off on a higher exponential rate so higher ordered harmonics are at a lower level than seen in an SET; its inherently lower distortion.

This allows the distortion signature to be innocuous.

The advantage is greater power output with lower distortion. So at any power level an SET can make, in a circuit using the same tubes the PP amp can have vastly lower distortion and so be smoother with greater detail, since distortion obscures detail. 

By the way, cascoding the input section is how you get both voltage resistance and a hundredfold reduction in Miller capacitance. Since Miller capacitance in transistors is grossly nonlinear, this is a very good idea. Tubes exhibit Miller capacitance too, but it is an order of magnitude lower, and it is stable and predictable instead of being nonlinear. There are cascode tube circuits as well, but they offer no improvement in linearity (unlike transistors), and are mostly seen in phono preamps and FM tuner input sections. In the tube universe, pentodes behave similarly to a pair of cascoded triodes, and are more commonly used when a cascode is called for.

We've been using a differential cascode circuit for decades. It has several advantages over pentode or cascade operation; one obvious one being a reduction in the need for a coupling capacitor. Differential circuits benefit from the devices (tubes in this case) having a lot of gain. That increases the differential effect so distortion cancellation is improved and noise is reduced. A differential cascode circuit can have a lot of gain.  

This circuit can have a very high CMRR even in a tube embodiment. Its linear enough open loop that you can run it without feedback (something you can't do with pentodes), but if you want to do it, its possible to operate it in ultra-linear mode, where the plate Voltage of the top tube is applied through a divider network to the grid of the same tube. You can do that with a pentode too, but you can't run a pentode zero feedback and the amount of feedback available in UL mode is limited. 

Since a cascode circuit is lower distortion, another advantage is it can be used in a circuit with feedback and result in less higher ordered harmonic generation than if a pentode is used.

Certainly a great benefit in a phono preamp where you have to overcome the wall of noise in the very first stage, and yes, cascodes are more linear than pentodes. The EF86 mike-preamp pentode so popular in the Fifties and Sixties is now very expensive and hard to get, so cascodes make more sense today.

Don’t need all that gain in a power amp in the absence of feedback, but if we ever needed feedback, yes, that’s a good way to get it.

I completely agree about the benign nature of low-order distortion. Most of all, it reduces nasty high-order IM distortion which is objectionable and obviously electronic-sounding. I also agree about the heavy 2nd-order distortion of SETs, which limits their effective dynamic range. Clever design techniques can mask the 2nd-order distortion (like coupling caps with complementary sonics) but the IM distortion remains, and is audible with symphonic, choral, and heavy rock music. (Music with a sparse spectra, like jazz quartets or chamber music, doesn't expose IM distortion, but music with a dense spectra turns IM distortion into a wall of noise that goes up and down with the music.)

Since transistors are notorious for high-order distortion, cascoded differential sections are about the only way to tame the things, while also getting rid of nonlinear Miller capacitance. The classic John Curl topology.

There are 3 or 4 sets of the final production Blackbirds in the world.  One is probably still in the boxes (go figure) as the person was building a new house or some such.  There is one guy who loves his and posted on the spatial audio circle I believe.   There are a few sets of the prototype ones in the smaller chassis with 6V6 drivers floating around and people like them.  But no one is really active on chat boards.  C'est la vie.   A review pair will make its way into someone's hands this year most likely.  The Raven preamp has had more people posting because it costs less and there are around 30 of the production ones out in the real world.

I have one of or the original stereo prototypes. I discovered it because of this thread though didn’t expect to be able to own one. I did see these amps (Blackbirds though they may not have referenced the name) brought up on another forum when I was looking up 6V6’s and it was brought up as a modern example/option using 6V6’s as drivers. I guess they didn’t know production version changed to KT88’s. I have some NOS matched pairs on order and am curious to see what differences they make. I still don’t know squat about tube design, but when I first got it I didn’t really know the difference between them and say similar looking 6SN7’s, but they seem like a great tube in their own right. 
 

There are a couple owners of the mono prototypes in another circle. I don’t think those owners will ever willingly let them go. The guy I got my stereo amp from definitely wouldn’t have sold it if he hadn’t gotten the monoblock prototypes. 
 

I still see Don Sachs preamps referenced and also notice Raven brought up as a top notch preamp option, even by non-owners.  There is definitely lots of talk about amps, but I see that people are more likely to roll through various preamps and DACs while holding onto their amp and speaker pairing once they’ve found a good one.  Other than not being many Blackbird 300b owners out there, there are many who don’t know about Don Sachs products, and many that do don’t associate with amps or especially 300b amps. Also experienced audiophiles I’m talking with in person who aren’t familiar with Spatial Labs. Some audiophiles don’t even visit forums at all to see such information if it is there. So only a few owners out there and I think there are a lot of folks out there that don’t know about the Blackbirds to chat about them. 
 

I hope Southwest Audiofest went well and generates some buzz. Has to be taken with a grain of salt due to all the variables, but always interesting to catch impressions from those who visit the feats. Too far for me, but I’m glad Gary and Lou got it going. I would definitely stop by and visit if ever at Axpona. 
 

$20k for amps is pretty unattainable for many consumers, but I’m glad there are those who can afford them so they can potentially be available on the used market some day. Are there any thoughts to a more entry level amp that could also pair well with the Raven as well as Spatial Labs speakers or similar, say 95 dB or better as previously referenced as a good sensitivity for ~20 WPC?

@yoder 

Thanks for the kind words.  The original stereo one was a proof of concept of the circuit.  I hope yours is running fine and making you happy.  The final production mono block version is considerably better in pretty much every way, but your stereo one is a lovely amp.  It runs a bit hot for my taste, and I could not fit all of the power supply and signal path refinements into a smaller, stereo chassis.  That said, I was very impressed with the amp, which, of course, led me to keep refining things into the final version.  For those who are interested, yes, there will be a more "entry level" product coming.  I am about to start prototyping.  This is a 300b lover's thread, so not really appropriate to discuss in detail, but it will have kt88 outputs, push pull in triode, and DHT drivers.  Similar circuit ideas to the Blackbird and your stereo amp, and obviously scaled back a bit to fit in a stereo chassis.  Fully balanced of course.  Probably by year end.  I have no idea of price, but it will be in the reasonable range and about 30-40 watts/ch.

@donsachs 

The stereo prototype is a very fine looking and sounding amp, so the final versions must be amazing!  I went all in with this amp being my first 300b after reading up much on 300b's elsewhere prior to seeing this thread.  I did not comment much on the sound as I don't have previous 300b experience and am not great with such descriptions.  Also maybe not highly relevant to current models, but probably as close as I will get!  It does run a bit hot.  Not so much heat up the room to me, but the transformers are hot to the touch and it does need space around it to radiate.  Definitely not to be tucked into a small cabinet, but probably no 300b is.  To hide all these great looking tubes wouldn't seem right either.  The sound to me mostly matches up with how you previously described the prototype monoblocks (if I recall correctly something like SET-like tone and vocals, but the punch of push/pull).  I have yet to own a 300b SET, but for the tube tone I would say it does compare to one of my favorite SET examples so far - it compares favorably with my 45 tube monoblocks for tone such as guitar tone.  It does not get too soft or weak in bass response as I've experienced with some non-300b tube amps.  Bass is deep, powerful, and I think has very good pace.  Power wise with no measurements just judging by volume adjustment level, output compares favorably with a couple other well regarded 30 WPC tube amps that I have or had.  Right now I have 97.5 dB 6 ohm speakers in a medium sized room so no power challenge there.  I will have ~next year some 93 dB 7 ohm speakers in a large room and am curious to see how it will match up.  I only asked about more entry level options as that is usually my gateway to learning/hearing about the top level gear.  

Yes, Don and I are working on a medium-price (by high-end standards) amplifier, most likely a stereo integrated amp. Not much to say about it now, since it's mostly conceptual at this point. If people are looking for value, I think Spatial may still be making Don's previous Kootenai amp, which is a superb amp.

HI

No Spatial is currently only making the Raven and Blackbirds... 

@yoder  I have played the mono 300b amps with 86 dB speakers and it drove them with no issue.  I am sure your 93 dB ones will be just fine.  It sounds like a 100 watt tube amp really.....  Yours runs a bit hot as it was the first build for proof of concept.  The final versions.. well they can play all day long at the show and you can put your hand on any transformer or the top panel and it is barely more than warm.  Lots of improvement since the first prototype build.  If you ever have trouble with that prototype send me a PM and I will of course try and help you.

Thanks for the update, Don.

Yoder, there was a lot of progress sonically after the first prototypes. What we have now is actually simpler in some ways. We went for a pair of spacious monoblock chassis mostly to simplify assembly, and to improve cooling a bit.

I would suggest that this thread has planted a seed that has now become a seedling.

Between my last post and very recently I have become the owner of a SE 300b Power Amp.

I am yet to fire it Up as it was purchased without Tubes, but am very close to having the first experience.

A dual concentric Speaker is also a new addition at 96dB efficiency, these are superb driven by the PP 845's and know they are going to work with the 5W 300b just fine.

Within my Local HiFi Group there are Six Brands of 300b able to be tried inclusive of my recent purchases of Two Brands of 300b. What is very interesting is that there is a good selection of Driver Valves up to approx' 70 years vintage to be used in conjunction with 300b's. The anticipation being, but also founded from prior experiences had, something will connect as a more than perfect synergy and be a Super Combo.

A lot to look forward to, the mould is broken, the new outlook is quite different and removed from my long term maintained Audio Ideals, where I have been very very committed to maintaining my choices made.  

Don and I wanted to avoid unobtanium vacuum tubes, while leaving the option of "tube rolling" up to the owner. All we ask is the owner uses matched pairs, whether vintage or modern. This retains DC balance while keeping distortion low ... with typical matched pairs, the dynamic gain-match is usually within 3% of each other, which gives a 30 dB distortion reduction compared to an equivalent SET circuit. All without using either local or global feedback.

The high-power Class A balanced drive for the 300B grids is a big part of the Blackbird’s sound. This is a sharp departure from "Golden Age" practice of the 1950’s and 1960’s, which used low-current "concertina" drivers for the output tubes (Dynaco circuit, also widely used in vintage receivers), or a differential long-tail pair (Mullard circuit). These are (barely) acceptable for pentodes and beam-tetrode tubes, but not very good for triode output tubes.

The designers of the "Golden Age" amplifiers were struggling for every single watt, which is why Class AB was universal for these amps, to maximize power output and efficiency. Watts were very expensive back then, and power transistors were notoriously unreliable. (Partly due to lack of understanding of Secondary Breakdown and stability issues with full-power oscillation, and partly the devices themselves.)

This is why the transistor conversion, which began around 1964, took about five to ten years. Transistor amps didn’t really become reliable until the mid-Seventies, while they passed tubes in the watts-per-dollar competition a few years earlier. It wasn’t until the late Eighties and early Nineties when tubes started to ascend in the North American high-end market. By then, the tube factories in North America, the UK, and Western Europe had all closed ... but Russian, Czech, and Chinese tubes appeared on the market to replace them.

Hi @lynn_olson ,

I’m working on my DIY 300B SET amplifier, which has three stages coupled by transformers. The first stage uses 1/2 6sn7, and I’ve been experimenting with different driver tubes: 6f6, 6v6, and 5881. I’ve noticed that the more powerful the tube, the better the performance. My findings so far are that the 6f6 is the least impressive, sounding somewhat forward and harsh. The 6v6 offers a lush and warm tone, while the 5881 provides the largest soundstage and the best instrument separation in complex musical passages.

The interstage transformer I’m using between the driver and the 300B is a Hashimoto A-305, which I understand is based on the renowned Tango NC-20. The challenge with this transformer is its 40mA maximum idle current. Currently, I’m running the 5881 at 36mA, and it seems to be working well within this limitation.

My question is regarding potentially moving to even more powerful driver tubes such as the EL37, KT66, or KT88. What would you estimate to be the minimum idle current required for these tubes to perform well in single-ended (SE) mode?

I’m also aware that if I switch to a more powerful driver tube, I’ll need to upgrade the filament transformer from a 25VA Antek to a 50VA Antek, which isn’t a major concern. The main constraint seems to be the idle current limitation of the interstage SE transformer.

Any insights you could offer on the minimum idle current for the EL37, KT66, or KT88 in this context would be greatly appreciated.

Best,

Alex

@alexberger 

If I were you I'd be concerned about the current for a driver tube that must pass through the core of the interstage transformer. That current is DC and causes the core to be magnetized. That in turn leads to saturation which increases distortion.

To reduce distortion, a saw cut is made in the core of the transformer. This dramatically reduces inductance at low frequencies and such frequencies will result in a poor load for the tube because the load line becomes elliptical. 

If i were you I'd be looking for less current rather than more. All you really need is the ability to saturate the power tube. If you can do that with a 6SN7 you'll be in good shape. 

BTW this is one reason I really don't think interstage transformers are a good idea for SETs- its why I prefer to direct couple. 

BTW this elliptical load line thing is a problem for most SET's output transformers (excepting parafeed circuits). For this reason, if you're using an SET you really want to limit bass frequencies getting into it as they cause the amp to make a lot of distortion, and are also hard on the (sometimes expensive) power tube as the load impedance literally vanishes at lower frequencies due to the lack of inductance in the output transformer. 

If you want the most out of almost any SET, preventing bass signals getting into it is one of the major ways to do so. 

Hi Alex! Unfortunately, I think you are limited by the maximum DC current the interstage will accept. As Ralph mentions above, SET transformers, whether interstage or output, have air-gaps in the core so they can tolerate the DC offsets. This has the side effect of requiring much larger cores, since air-gapping the core reduces inductance several times. OK, so the IT grows in size. Unfortunately, there is another downside as the core gets larger ... stray capacitance between windings also gets larger, which is a fancy way of saying less bandwidth.

So the effective cost, assuming clean-sheet transformer design, is decreased bandwidth on both ends of the spectrum. But transformer designers are clever and now have access to computer modeling that wasn’t available in previous decades, so clever interleaving schemes, and analysis of flux density in different parts of the core, lets them "do the impossible" compared to 20 or 30 years ago. So I would suggest looking at Cinemag or Monolith if you want an exotic custom design, but I warn you that requiring 60 mA as well as wide bandwidth will push the transformer to the limits of what can be done. And it will not be cheap ... custom work never is.

Don and I are using both Cinemag and Monolith, and they are custom designed for us. They are not off-the-shelf parts, and we collaborated with the transformer designers over several iterations before we got to where we are now.

You are limited in scope as long as you are buying off-the-shelf transformers, whether IT or output types.

However, it isn't bad as it sounds. Even a massive tube like the KT88 works just fine at 36 mA. Conveniently enough, all of these family of beam-tetrode tubes have the same octal socket and the same pin wiring. So just twiddle the cathode resistor to run the tube at 36 mA, and off you go. And if it's a little guy like a 6V6, just run it around 25~30 mA, and compare your favorite types. Frankly, the choice of cathode bypass cap may be more audible than the tube you choose.

If my vague memory serves, a neighbor of mine, Thom Mackris of Galibier Design, has been building an amp similar to yours. The lineup is: 1/2 6SN7 RC-coupled to a triode-connected 6V6 driver, with SE IT coupling to the 300B grid. Isolated B+ supplies for the input+driver and output section.

I haven’t spoken to him in a couple of years, so it might be different now, but before the pandemic hit, it sounded pretty good, a lot better than most commercial SET amps I’d heard at shows. But that was a while ago, and it could be quite different now. But the overall concept was solid ... basically an updated Herb Reichert Silver 300B amplifier.

If an 8-watt SET is your thing, what you’re already doing is as good as any. The rest comes down to cap tuning and whichever wire you build it with.

Hi @lynn_olson ,

I think the Hashimoto a-305 is an extraordinary transformer. The measured frequency response of this transformer (with 6f6) in my amplifier is 6Hz to 90KHz -3dB. The difference in sound between RC coupling and this IT was huge, it is a completely different amplifier.

http://www.tube-amps.net/images/Hashimoto_Specs/A-305.jpg

I also can recommend 5881 over 6f6. With 5881 the sound is bigger, more air around instruments and better separation. 6f6 gives more chamber and lush sound.

Hi @atmasphere ,

I understand your point. I will not implement it in my current amplifier, but I can try this idea in the next project.

I have another question, how do you balance a number of tubes in parallel on your OTL amplifier without fixed bias. Is it necessary to select all these tubes by parameter or the amplifier circuit balances all parallel tubes automatically? 

The measured frequency response of this transformer (with 6f6) in my amplifier is 6Hz to 90KHz -3dB.

@alexberger while that looks excellent, I suspect these measurements at the link are not at full power. If you want the amp to perform, full power bandwidth is important. Did you measure your amp at full power?

how do you balance a number of tubes in parallel on your OTL amplifier without fixed bias. Is it necessary to select all these tubes by parameter or the amplifier circuit balances all parallel tubes automatically? 

Actually our OTLs are fixed bias. A manual DC Offset control is provided to balance the two output tube banks using the front panel meter.

We've never practiced power tube selection.

You can get a bit lower distortion by doing so; the distortion created by mismatch tends to be lower ordered harmonics. 

Most of the distortion in our design depends on how well matched the internal sections are in the input tube, since the input Voltage amplifier comprises all the gain of the amplifier. Usually that defines the measured distortion; apparently the output section is very low distortion on its own. 

This approach allows for much lower distortion than is possible with an SET. For example our MA-2 (which is zero feedback) can make 220 Watts (230 at clipping) with distortion at full power around 0.5% (with excellent tube matching); up to 1.8 to 2% if the tubes are not matched at all. Most SETs cannot make that sort of power; if the MA-2 is running at any level an SET can make the distortion is several orders of magnitude lower and its mostly 2nd and 3rd harmonic. Lower distortion is directly associated with greater transparency and detail and this is easily heard. 

Unlike an SET or push-pull tube amp, the distortion of our OTLs is unaffected by frequency since the amp has a direct-coupled output. Its full power down to 1 Hz.