My latest breadboard, a 3 stage 300B SET with interstage transformers between each stage.
The tube/transformer lineup is: CV1135 - LL1660 mA - 6S4S - LL1671 50mA - 300B - LL1679 70mA.
Each stage has its own tube rectified HT supply, 6AX5GT for the input and drivers and dual 6AU4GTA for the output tubes. Coleman regs for the direct heated drivers and output tubes.
Plenty of hum to get rid of before I can say much about the sound...
The tube/transformer lineup is: CV1135 - LL1660 mA - 6S4S - LL1671 50mA - 300B - LL1679 70mA.
Each stage has its own tube rectified HT supply, 6AX5GT for the input and drivers and dual 6AU4GTA for the output tubes. Coleman regs for the direct heated drivers and output tubes.
Plenty of hum to get rid of before I can say much about the sound...
That looks very interesting and has me wondering what the silicone and heat sinks are doing.
A schematic would be interesting to review as there is lots here to look at in your pictures.
I would like to nominate your build for the "most iron per amplifier award".
Haha thanks! The heatsinks are for the filament regulators, a bit too small though but easy to replace.
I'll post a schematic when I'm at a computer. This is only one of several topologies I intend to try before settling for a final design.
The output stage is pretty much determined by the OPTs and the HT supply. I wanted to try this 3 stage idea as I already had both pairs of interstage transformers but I'm not entirely convinced that it is the best sounding solution.
I'll post a schematic when I'm at a computer. This is only one of several topologies I intend to try before settling for a final design.
The output stage is pretty much determined by the OPTs and the HT supply. I wanted to try this 3 stage idea as I already had both pairs of interstage transformers but I'm not entirely convinced that it is the best sounding solution.
I have not tried a 300B build yet to date sticking with "modern" tubes so seeing your work is of interest to me.
Perhaps some day I will stray into 300B land however judging from your build it looks to be fraught with a danger of drifting into a pleasant obsession.
Perhaps some day I will stray into 300B land however judging from your build it looks to be fraught with a danger of drifting into a pleasant obsession.
To be honest, I prefer transmitter tubes with TT filaments but I got a few pairs of 300Bs as a payment for a job I did a year or two ago.
The idea was to use them together with a pair of LL1688 OPTs that I've had on the shelf for way over a decade but I decided to save those for a large transmitter tube project and bought a pair of smaller OPTs for this amp.
We'll see what becomes of this, a finished version of this three stage design with DHT drivers would require three separate chassis (amp, HT supplies and filament supplies).
Such a project would have a lot of "audiophile swag" for sure but a simpler design without the DHT drivers and their affiliated filament supplies would be a lot more manageable.
The idea was to use them together with a pair of LL1688 OPTs that I've had on the shelf for way over a decade but I decided to save those for a large transmitter tube project and bought a pair of smaller OPTs for this amp.
We'll see what becomes of this, a finished version of this three stage design with DHT drivers would require three separate chassis (amp, HT supplies and filament supplies).
Such a project would have a lot of "audiophile swag" for sure but a simpler design without the DHT drivers and their affiliated filament supplies would be a lot more manageable.
Nice tubes but a bit too large for my purposes.
I already have three SETs with TT triodes (808, T20 and TZ20) and most of the parts for a fourth one that will be designed to allow a wide range of different output tubes as long as they have UX-4 sockets and and a filament current draw below 5A or so.
A conceptual schematic of the amp section. Well, the only parts missing are the gate stopper resistors.
As I mentioned, this amp had a serious hum problem at the first startup. I figured it would quite an effort to get rid of it so I put the whole thing on the kitchen table for some debugging:
Turned out to be due to inadequate filtering of the negative bias supplies. A few extra caps at the right places removed 95% of the hum, what remains can probably be fixed with better HT filtering.
A quick run with the tone generator and oscilloscope shows around 8W at the tap before the output waveform starts to look a bit funny, presumably because of ringings in the interstage transformers when the output tubes hits class A2. Real clipping occurs somewhere around 11-12W.
The bandwith is perfectly fine considering the design, nothing unexpected happens between 30Hz and 30k.
As a funny side note, those ape-periodic speakers sounded much better than I remembered. No bass at all of course, but the music was completely "released" from the speakers and sort of hovered above the power supplies in a way that seldom happens with more complex, multi-way speakers.
Turned out to be due to inadequate filtering of the negative bias supplies. A few extra caps at the right places removed 95% of the hum, what remains can probably be fixed with better HT filtering.
A quick run with the tone generator and oscilloscope shows around 8W at the tap before the output waveform starts to look a bit funny, presumably because of ringings in the interstage transformers when the output tubes hits class A2. Real clipping occurs somewhere around 11-12W.
The bandwith is perfectly fine considering the design, nothing unexpected happens between 30Hz and 30k.
As a funny side note, those ape-periodic speakers sounded much better than I remembered. No bass at all of course, but the music was completely "released" from the speakers and sort of hovered above the power supplies in a way that seldom happens with more complex, multi-way speakers.
Got a possible contender. Slight disadvantage due to it being just a two-stage amp:
Power TX, Main power autoformer, 10A/6.3V for || 6AU4, 10B Variac feeding 2x filament Iron, Main B+ chokes( L-L-C), final filter choke( last L-C stage ), bias TX, OPT, and grid choke for the finals.
On a per Watt, is under 4lb/Watt. There are more than 50 ov them tho... 🙂
Douglas