Hi Folks,
I already have a pair of this output transformers:
Primary: 2,5k or 3,4k
18W
I would like to build a 300b stereo amplifier or two mono amplifiers for my HiFI.
For the moment I have a SE PCL 86 amplifier with 4watts, which is actually enough for my speakers, but I hope to achieve 2x8watts with the 300b's.
There are a lot of schematics:
I would like to know which one fits my output transformer best - or if there are other schematics - feel free to post them here
What would also be nice: If the amp wouldn't need a preamp - I thought about an aikido preamp to drive the 300b but the schematics from tubecad require different primary impedances.
Have a nice day!
Felix
I already have a pair of this output transformers:
Primary: 2,5k or 3,4k
18W
I would like to build a 300b stereo amplifier or two mono amplifiers for my HiFI.
For the moment I have a SE PCL 86 amplifier with 4watts, which is actually enough for my speakers, but I hope to achieve 2x8watts with the 300b's.
There are a lot of schematics:
I would like to know which one fits my output transformer best - or if there are other schematics - feel free to post them here
What would also be nice: If the amp wouldn't need a preamp - I thought about an aikido preamp to drive the 300b but the schematics from tubecad require different primary impedances.
Have a nice day!
Felix
The "Ceasar II", with its pentode μ follower small signal topology, is (IMO) worthy of close study. I've seen that sort of circuitry using a 6BM8/ECL82. The triode in the 6BM8 has a μ of 70 compared with only 15 for the 6H30Π (6n30p). The increased gain might be sufficient to eliminate line stage gain, when fed from a "standard" 2 VRMS digital source.
Last edited:
Personally, if I were going to build a 300B from a proven design, I'd build the Tubelab TSE II:
After a 14 year run, the TSE must DIE!
This has active loading on the input stage as well as a direct coupled driver for the 300B to drive it into A2 (a little bit higher power).
There's a board available from George, but he also publishes the full schematic.
After a 14 year run, the TSE must DIE!
This has active loading on the input stage as well as a direct coupled driver for the 300B to drive it into A2 (a little bit higher power).
There's a board available from George, but he also publishes the full schematic.
First of all, as mentioned, any driver is going to work with your output transformer, that's not important. Second, are you going to use this with a preamp or direct-input? If you already have a preamp, what kind is it and how much gain is it capable of? That matters because among all the SE 300B amplifier variations, there are very different gain structures. For instance, the first schematic you posted, a cascaded 6SN7 with the first stage bypassed will have a *lot* of gain, more than enough for direct input but probably too much for a preamp. The Time Bandit, however, is going to need a preamp with a healthy amount of gain to reach full power. The Caesar II is a problem because D3A tubes are almost impossible to get now. There are substitutes but I'm not sure what those are at the moment.
Also, be prepared to try different topolgies as you become more acquainted with an SE 300B amp.
For starters, the JE Labs design is easy to build and pleasant enough. It has good gain but not so much that you couldn't use a preamp. And it's infinitely "tuneable" with different 6SN7s. There's very little difference between that and the first schematic,
Also, be prepared to try different topolgies as you become more acquainted with an SE 300B amp.
For starters, the JE Labs design is easy to build and pleasant enough. It has good gain but not so much that you couldn't use a preamp. And it's infinitely "tuneable" with different 6SN7s. There's very little difference between that and the first schematic,
In support of Grover's post and adding;
You could build the JE labs to start and then following the suggestion of Palustris in the third post of this thread expand it to the Komuro amp. Also, instead of using the two separate driver tubes you could switch to a dissimilar dual like the 6DN7 . There are a good number of them that would work well. ( and they can sound great!)
You could build the JE labs to start and then following the suggestion of Palustris in the third post of this thread expand it to the Komuro amp. Also, instead of using the two separate driver tubes you could switch to a dissimilar dual like the 6DN7 . There are a good number of them that would work well. ( and they can sound great!)
The OP is looking for an "integrated" amp. "Ceasar II" is interesting for its topology, not the specific implementation. Switch to the 6BM8/ECL82, for both greater gain and ease of acquisition, and make R6 a potentiometer. IIRC, a Svetlana Technical Bulletin was published that describes the implementation in question.
Right you are, Eli, I failed to notice that. In that case the JE Labs would probably be perfect, or he can add a bypass cap to the cathode of the first stage for more gain, like the first schematic. Your suggestion is very good.
An Aikido, in my experience, has *way* to much gain for a 6SN7 cascade, but would be perfect for the "Time Bandit" version.
Last edited:
Geez, for a beginner? (I assume the OP is a first time SE builder, but maybe I'm wrong.) I'm pretty good at reading schematics but even I am struggling with that one.
thanks for the address, the prices are correct and if we can make work a European manufacturer, it's even better.
Don't over think it. Input + driver stage/buffer run from -ve voltage to facilitate output stage filament at 0V WRT common. Lots of the good 'audio note' stuff ran like this. Much to like but who needs two cap coupling - especially if you have already split supplies like that?.. Use JCM inspired 'upside down' and DC the three stages. What you save in premo parts just went a long way to paying for the entire build - and it might be better.
I've had issues with input transformer for this type of build - something to do with reference to 0V on the primary and then (anything but that) on the secondary screws with BW, not my forte.. Anyway, something if not nothing. And to the OP, build any of the schems and you'll be fine = 'start now'.
HK
Last edited:
What kind of distortion profile do you want?
Iow do you want euforic "tube sound" or not.
Some of these designs have more tube sound than others.
I would go for a lot of tube sound, because other technologies can do the less tube sound a lot cheaper. And circuits with more tube sound are mostly easyer to build.
Iow do you want euforic "tube sound" or not.
Some of these designs have more tube sound than others.
I would go for a lot of tube sound, because other technologies can do the less tube sound a lot cheaper. And circuits with more tube sound are mostly easyer to build.
Hi,
I like the idea of having a lot of tube sound. Right in the moment this amplifier drives my speakers pretty good, but at louder volume sound gets muddy:
http://www.walter-elektronik.de/infos/PCL86stereo.pdf
my speakers:
8 Ohm
(2,83V, 1m): 95db
(-3db): 50HZ-25kHz (-8db 40-24000Hz)
tweeter: Celestion CDX-1430
Woofer: Monacor SPA-115PA
So my idea would be something like the JE Labs Circuit but with a DC heated 300b - I found the idea on the German webside frihu.com and attached the schematic.
I also contaced tubelab for a TSE-II board - but I would prefer a point to point layout and the sharp voltage regulator is also a problem.
Cheers,
Felix
I like the idea of having a lot of tube sound. Right in the moment this amplifier drives my speakers pretty good, but at louder volume sound gets muddy:
http://www.walter-elektronik.de/infos/PCL86stereo.pdf
my speakers:
8 Ohm
(2,83V, 1m): 95db
(-3db): 50HZ-25kHz (-8db 40-24000Hz)
tweeter: Celestion CDX-1430
Woofer: Monacor SPA-115PA
So my idea would be something like the JE Labs Circuit but with a DC heated 300b - I found the idea on the German webside frihu.com and attached the schematic.
I also contaced tubelab for a TSE-II board - but I would prefer a point to point layout and the sharp voltage regulator is also a problem.
Cheers,
Felix
Last edited:
First, I heard both the original "Flesh and Blood" amp and Komoru's all-DC-coupled SE 300B amp at the Fi store in NYC, back in the day. I heard the F&B at a friend's place many times after that. I thought Komoru's amp sounded much better. I found the F&B kind of 'thin' sounding. I absolutely loved Komo's amp. That's a great one for a simple design, but it had stacked DC supplies, with >700V DC to ground. Could be dangerous.
If you took Herb Reichert's "Flesh and Blood" circuit (which is as simple and straightforward as it can possibly be) and improve on that, you should get a pretty nice intro to tube amp building, with a high chance of success.
What would I change?
Well, first, I would load the second stage plate with a plate choke, so that it can swing beyond its B+ (the choke stores energy). However, the 6SN7 has too high a plate resistance (rp) for almost any plate choke, so you'll want to change that tube to something with much lower rp, perhaps a 6N6P or 5687, or a triode wired 6V6 or EL84 type of tube. Or go really low rp and use a triode-wired EL86 (or Russian 6P43P). Komoru's 300B amp used a 6V6-triode with plate choke. The second stage is the one with the big voltage swings, so will be the one requiring most careful design.
It would be nice to DC-couple the 2nd stage to the 300B, which would again get you closer to Komoru's design. But that's more advanced stuff. You'd need to 'stack' the power supplies, which gets you into truly high voltage territory (>700V DC). One oops and it could be all over. Probably best to RC-couple the stages, keep the max DC voltages to <500V.
The input tube can be almost anything you like. It won't need to give you much gain. You could use the other half of the 6N6P or 5687, or you could use something like an EF86, 6SJ7 or 6AU6 triode wired. Or you could use a 6J5 (again, like Komoru did). You could even try a nice vintage low-mu DHT like 26 or 01A. But then you'd have to worry about getting rid of the filamentary hum. Maybe triode-wired 2P27L?
As you can see, even if limiting yourself to a 3-stage RC-coupled all-triode topology, there are lots of choices of tube types. Perhaps...
6J5GT (Russian 6S2S) RC-coupled to 6V6GT-triode (Russian 6P6S) with plate choke RC-coupled to the 300B.
Easier to lay out would be...
6N6P RC-coupled to 6N6P with plate choke RC-coupled to the 300B. Same idea, different tubes, only two required per channel.
All of the above tubes have Russian equivalents, so you could breadboard with cheap tubes and once you're happy with your design, search out fancy vintage tubes (if you want to).
Anyway, that's an idea.
--
If you took Herb Reichert's "Flesh and Blood" circuit (which is as simple and straightforward as it can possibly be) and improve on that, you should get a pretty nice intro to tube amp building, with a high chance of success.
What would I change?
Well, first, I would load the second stage plate with a plate choke, so that it can swing beyond its B+ (the choke stores energy). However, the 6SN7 has too high a plate resistance (rp) for almost any plate choke, so you'll want to change that tube to something with much lower rp, perhaps a 6N6P or 5687, or a triode wired 6V6 or EL84 type of tube. Or go really low rp and use a triode-wired EL86 (or Russian 6P43P). Komoru's 300B amp used a 6V6-triode with plate choke. The second stage is the one with the big voltage swings, so will be the one requiring most careful design.
It would be nice to DC-couple the 2nd stage to the 300B, which would again get you closer to Komoru's design. But that's more advanced stuff. You'd need to 'stack' the power supplies, which gets you into truly high voltage territory (>700V DC). One oops and it could be all over. Probably best to RC-couple the stages, keep the max DC voltages to <500V.
The input tube can be almost anything you like. It won't need to give you much gain. You could use the other half of the 6N6P or 5687, or you could use something like an EF86, 6SJ7 or 6AU6 triode wired. Or you could use a 6J5 (again, like Komoru did). You could even try a nice vintage low-mu DHT like 26 or 01A. But then you'd have to worry about getting rid of the filamentary hum. Maybe triode-wired 2P27L?
As you can see, even if limiting yourself to a 3-stage RC-coupled all-triode topology, there are lots of choices of tube types. Perhaps...
6J5GT (Russian 6S2S) RC-coupled to 6V6GT-triode (Russian 6P6S) with plate choke RC-coupled to the 300B.
Easier to lay out would be...
6N6P RC-coupled to 6N6P with plate choke RC-coupled to the 300B. Same idea, different tubes, only two required per channel.
All of the above tubes have Russian equivalents, so you could breadboard with cheap tubes and once you're happy with your design, search out fancy vintage tubes (if you want to).
Anyway, that's an idea.
--
Attachments
Last edited:
D3a is accessible.
https://www.ebay.com/sch/i.html?_fr...e&_sacat=0&LH_TitleDesc=0&_osacat=0&_odkw=d3a
I use high gain trioded pentodes (D3a and -almost- substitute ones: E280F, E810F, E180F, or C3M, C3g) as first stage VAS for decades.
