"JC Morrison" 300B SE design flaws/issues - redesign.

[StepheK]

As I posted in another thread, I did a video series build based off this "classic 300B" schematic that has been floating around for decades, using a direct coupled 6SN7 2 stage front end. After I got finished, did some basic scope testing and then listened to it, I wasn't impressed. Thin bass, overly bright top end but mainly saw weird asymmetrical clipping out of the front end to the point the amp would go into heavy distortion at anything past 2W RMS output. It did have a very high input sensitivity but this clipping before the output tube grid crippled it to flea watt status.

Given I planned for this to be just a starting point for this amp, I gave this some thought knowing I wanted to try using these 6SN7 tubes. I thought about trying a cap coupled twin triode two stage, but then wired up the frontend very similar to the mods I did to the Nobsound 6P1 amp: as a Cascode with a voltage divider feeding the upper grid. I also increase the DC offset on the heaters of the 6SN7 tubes from 35V to 50V. This mod fixed the asymmetrical clipping and now can be driven to 6W RMS. It lost some of the input sensitivity but most of that was wasted on a normal line input, as the amp went into heavy distortion so easily. It now sounds 100% better and I feel like with a few minor tweaks (I plan to try a 0.33uf coupling cap etc) will be a great amp.

Where I am baffled is I have seen several articles about builds based on this design, using some high end transformers etc., praising how great it sounds. I was also told some of the current production ChiFi 300B amps use this design. The problems I saw were easily visible on the grid of the 300B, so I don't see how $1500+ OT would fix those fatal flaws. I suppose if I had never heard a good tube amp and had very sensitive speakers I might have been impressed, but seriously my little 6BM8 amp sounds much better, more solid bass response and has almost as much drive! Rewired as a cascode, this 300B amp now sounds more like you would expect.

Anyway, if anyone ever plans to use that schematic, I would suggest they not bother, and instead try to this cascode front end design I came up with. I'm also interested in other designs using a pair of 6SN7's to drive a 300B.

 

[zackthedog]

Well, yes, that's a very inadequate design. ;-) The driver 6SN7 is running at 2.5 mA with only 70 volts headroom at the plate. Glad you found a fix for the time being!

Six watts with a 5K load at that operating point is pretty good, probably about as much as you can squeeze out of the 300B under those conditions.

 

[PakProtector]

The 6SN7 is not ideal for a cascode where B+ is in short supply. How did you calculate( or better still, draw) the plate curves?
cheers,
Douglas

 

[StepheK]

The 6SN7 is not ideal for a cascode where B+ is in short supply. How did you calculate( or better still, draw) the plate curves?
cheers,
Douglas

It's much more ideal than that JC Morrison mess Just watch the video and the actual scope patterns to see what a mess that one is.

What I posted here sounds light years better than this "Classic" 300B design that has been floating around for decades, and I plan to keep experimenting with the design. Feel free to post your own "ideal" schematic, after doing real world testing what it sounds like. I've learned what people calculate and sim should sound good doesn't always pan out that way.

 

[Palustris]

that JC Morrison mess

I'm not certain why you continue to cast disapprobation on JC Morrison when you have been advised that the schematic was only one of a dozen or two that JC published in the 1990s in a pamphlet mostly of Japanese schematics from the 1980s that he curated as examples for audio experimenters.

This schematic you now denigrate even though you were warned on the first page of your post in DIYAudio that it was poor quality, was developed by Sun Audio for its 2A3 amplifier for which it is better suited. This schematic is not a "classic" in the sense of a Mullard 5-20 or Williamson, or even the DRD amp or "monkey." It is simply the schematic for an amplifier that many consider a "starter" amp for those investigating single ended for the first time. These Sun amps continue to hold their value at close to $2k on auction sites. So despite their weak power supply and poor topology, people still value them.

JC Morrison certainly doesn't need me to defend him against this kind of nonsense, but I don't understand why you persist in calling this the "JC Morrison mess" when his entre responsibility was assemble a group of Japanese schematics from the 1970s and 1980s and publish them in a small pamphlet.

 

[StepheK]

#1 google hit on "300B schematic" is this design, attributed to him. Anytime someone discusses 300B schematics, this one comes up and everyone I have seen calls it the "JC Morrison" design. And actually, if he assembled a group of schematics to sell people and never tested them, then yeah he does own this mess. Here JC Morrison is quoted as having said "This amp is the quintessential triode amp. It just does its job, period. It is honest and accurate and sweet and will drive a surprising number of commercial speakers to acceptable levels. If you use the best available parts and materials in constructing it (as well as stash away a few tubes for a rainy day), it will pass down to your children and continue to wow 'em in the next millenium." - J.C. Morrison (1993)

http://diyaudioprojects.com/Schematics/300B-SE-Tube-Amp-Schematic.htm

If he said that about this amp, again he owns it. You say he just published these as samples, another person is quoting him as saying this is "the quintessential triode amp". So which of you two should I believe? If this isn't true, maybe you should try to get that webpage taken down.

BTW what you just posted isn't the same schematic, several resistor values are different, the voltages are different, driving a different tube etc. And why would anyone suggest an amp design with this nasty asymmetrical clipping as a "starter" amp for those investigating single ended for the first time? That clipping would happen no matter what tube this was driving. That's a recipe to drive someone out of this hobby, their first build sounds bad...

And I also can't help it if people have such bad taste they value an amplifier built with this design, nor do I believe everything some random person on the internet tells me. In the recent thread on DIY about 300B designs there are several people also praising this design, as do multiple web pages based off this design, so which of those posts/pages should I believe? The ones warning me or the ones praising it? If anything, I've learned to not believe anything audio related published on the internet without testing it myself.

 

[maxhifi]

In the early 90s, there was no internet yet, and info on tube amps was thin on the ground..... tube theory hadn't been taught in college for a generation already, and the only way to learn was dusty old books, and whatever was being published in audiophile magazines. Anyone working in that vacuum of info can be forgiven for a few mistakes!

If you're going to call any 300B design from that era quintessential though, I put forward the Joe Roberts design:

The Single 300B Amplifier A model 91 for 1992 Article By Joe Roberts Sound Practices Magazine Online (enjoythemusic.com)

This is a very safe intro to triode amplifiers, which also works well and sounds good. It is also known as the Angela Model 91, published by Angela electronics. They had a really great website back then, what remains doesn't give any hints how good it used to be. I built a version of this, and used it for years. I used 2A3 tubes though, mostly because I thought the price of 300B tubes was pure insanity and I didn't have the money for something better!

That so called JC Morrison design looks like someone lopped off the first tube from a Williamson amp, and used it to drive a triode directly. It would work fine with a tube which requires a small driving voltage, but the 300B just asks too much. I think this was previously discussed here quite recently?

 

[zackthedog]

In the early 90s, there was no internet yet, and info on tube amps was thin on the ground..... tube theory hadn't been taught in college for a generation already, and the only way to learn was dusty old books, and whatever was being published in audiophile magazines. Anyone working in that vacuum of info can be forgiven for a few mistakes!

If you're going to call any 300B design from that era quintessential though, I put forward the Joe Roberts design:

The Single 300B Amplifier A model 91 for 1992 Article By Joe Roberts Sound Practices Magazine Online (enjoythemusic.com)

This is a very safe intro to triode amplifiers, which also works well and sounds good. It is also known as the Angela Model 91, published by Angela electronics. They had a really great website back then, what remains doesn't give any hints how good it used to be. I built a version of this, and used it for years. I used 2A3 tubes though, mostly because I thought the price of 300B tubes was pure insanity and I didn't have the money for something better!

That so called JC Morrison design looks like someone lopped off the first tube from a Williamson amp, and used it to drive a triode directly. It would work fine with a tube which requires a small driving voltage, but the 300B just asks too much. I think this was previously discussed here quite recently?

Yes, it's the first tube of a Williamson. It's also worth bearing in mind that in the early 90's Japanese 300B designs tended to run the tubes very lightly. The idea was to maximize "tone," not power or bandwidth. Japanese 300B designs of this period were intended to drive highly efficient horns or single-driver speakers, like vintage Lowthers or the WE 755.

 

[Selmerdave]

Yes, it's the first tube of a Williamson. It's also worth bearing in mind that in the early 90's Japanese 300B designs tended to run the tubes very lightly. The idea was to maximize "tone," not power or bandwidth. Japanese 300B designs of this period were intended to drive highly efficient horns or single-driver speakers, like vintage Lowthers or the WE 755.

I don't know, Japanese 300B designs of the time often used power tube drivers like EL34...think Sakuma.

 

[trainbuftony]

Its fundamentally a feedback free triode amp and therefore if someone told me it sounds wildly different depending on what speakers its attached to, I'd believe them.
It likely does perform wildly different into a speaker (especially one with a crossover) than into a dummy load. That's to be expected in a feedback free triode output stage, consequently the original builder could have fine tuned parts values to work best with whatever speakers he was using.
If the speaker is unknown or unknowable then you're going to have different results with your speakers, or dummy load as the case may be. That's nobody's fault, per say, but we have fewer of those types of issues when push pull stages, and global feedback networks get the output impedance down.
It's a limitation of the chosen topology.
Some people look at a challenge as an opportunity. I'm generally one of those people. I'm also a person with a limited budget and I find that my hifi dollar actually goes way further when I build a push pull amplifier.
What I mean by that is that I can get more watts with less distortion and have it sound better and cost less using 2 smaller tubes in the output stage in push pull. It seems illogical but my experience bears this out over and over and over.

Triode SE feedbackless amps require extensive tuning and voicing. Many look at that as an opportunity, I ain't one of em.
That is what they do.
Comes with the territory.

 

[StepheK]

That's to be expected in a feedback free triode output stage, consequently the original builder could have fine tuned parts values to work best with whatever speakers he was using.

But the clipping problem is seen on the grid of the 300B, I don't see how the output load would effect that. Maybe I am totally ignorant about something here, first to admit I might be. I also can understand the problem I saw might not exist if the design was around running the 300B very lightly, but the schematic shows running more plate voltage/hotter that I am using. I couldn't get more than 2W out of the original design without seeing hard clipping on the bottom side of the sine wave.

 

[trainbuftony]

Bias shift under load?

 

[Palustris]

[QUOTE="StepheK, post: 14944766, member: 333381]"I couldn't get more than 2W out of the original design without seeing hard clipping on the bottom side of the sine wave.[/QUOTE]

You were warned about this when you started the project. The Sun Audio 2A3 or 300B schematic from the 1980s is fundamentally for driving very efficient speakers, classically the Western Electric 755A, Altec or Tannoy 15" in a small room. That was a typical audio system for the Japanese at that time. They generally didn't need more power than 2W. Sakuma was very fond of the 845, but he was not exactly typical of the average audio enthusiast. To see a broader range of typical Japanese audio amplifiers of the last 30 years or so visit Asano Amplifier.

I have a friend with one of the Sun Audio 300B amps that I repaired for him 15 years ago (replace those Chinese sockets with USA NOS!). He still uses the amps in a tiny room with some 4" tuned port speakers and likes them just fine.

 

[zackthedog]

I don't know, Japanese 300B designs of the time often used power tube drivers like EL34...think Sakuma.

Sakuma used a lot of power tubes to drive other power tubes, stacking them up with multiple interstage transformers, but he typically ran 845s and 211s at less than 500VDC, sometimes higher, rarely at anywhere near their maximums. He advocated for "tone," not power. His favoriter speaker, IIRC, was the WE 755. He was finally persuaded to attend a NY Noise gathering at some point, hauling one of his massive amplifiers along. Attendees were shocked at how limited the bandwidth of the amp was.

 

[maxhifi]

Sakuma used a lot of power tubes to drive other power tubes, stacking them up with multiple interstage transformers, but he typically ran 845s and 211s at less than 500VDC, sometimes higher, rarely at anywhere near their maximums. He advocated for "tone," not power. His favoriter speaker, IIRC, was the WE 755. He was finally persuaded to attend a NY Noise gathering at some point, hauling one of his massive amplifiers along. Attendees were shocked at how limited the bandwidth of the amp was.

Is he still alive? I would loved to have visited his restaurant and heard for myself what it was all about. I still kind of want to buy a DL-102 after reading his website years ago. And some 6B-G8 tubes!

 

[Selmerdave]

I don't know if he's still alive, but he would be quite old if he was. He seemed to be mid-60s around 2004 when I visited.

 

[maxhifi]

I don't know if he's still alive, but he would be quite old if he was. He seemed to be mid-60s around 2004 when I visited.

View attachment 2291243 View attachment 2291244

Wow! So how was it?

 

[Selmerdave]

Wow! So how was it?

Well, it was something to behold for sure, I haven't seen that much iron in one room before or since (the photos show maybe 1/4 of the amps there). It was definitely about the midrange, and I was really not used to listening to mono at the time so in some ways it was hard to appreciate. The immediacy of the Altec A5 being driven by the 6C33 amp above was definitely something, and the Lowther TP1 playing harpsichord music had a startling realism. My favorite of his setups was probably the 845SET above powering an Altec biflex speaker, that sounded fantastic. I wasn't as enamoured with his keg-cabinet Lowther PM6 playing Johnny Hartman as he seemed to be, but at the time single-speaker mono was very far out of my wheelhouse. He was a beautiful and generous man, to be sure.

 

[zackthedog]

Well, it was something to behold for sure, I haven't seen that much iron in one room before or since (the photos show maybe 1/4 of the amps there). It was definitely about the midrange, and I was really not used to listening to mono at the time so in some ways it was hard to appreciate. The immediacy of the Altec A5 being driven by the 6C33 amp above was definitely something, and the Lowther TP1 playing harpsichord music had a startling realism. My favorite of his setups was probably the 845SET above powering an Altec biflex speaker, that sounded fantastic. I wasn't as enamoured with his keg-cabinet Lowther PM6 playing Johnny Hartman as he seemed to be, but at the time single-speaker mono was very far out of my wheelhouse. He was a beautiful and generous man, to be sure.

Very nice, thanks for sharing that.

 

[maxhifi]

Well, it was something to behold for sure, I haven't seen that much iron in one room before or since (the photos show maybe 1/4 of the amps there). It was definitely about the midrange, and I was really not used to listening to mono at the time so in some ways it was hard to appreciate. The immediacy of the Altec A5 being driven by the 6C33 amp above was definitely something, and the Lowther TP1 playing harpsichord music had a startling realism. My favorite of his setups was probably the 845SET above powering an Altec biflex speaker, that sounded fantastic. I wasn't as enamoured with his keg-cabinet Lowther PM6 playing Johnny Hartman as he seemed to be, but at the time single-speaker mono was very far out of my wheelhouse. He was a beautiful and generous man, to be sure.

Fantastic story, great to hear you got to experience it! I always wondered how on earth that barrel speaker could possibly sound good!