Cap coupled is not an issue when properly designed. Make sure to have enough headroom to avoid blocking distortion, size the capacitor big enough to avoid phase and fr issues and use a sufficient driver (adding a Mosfet source follower doesn't hurt).
Looks like you are SOL unless someone stumbles on that English paper.
The DC Coupled Amp examples from EW Jan 2001 could easily be redesigned for lower power. And still no caps.
And fewer toobz. Back when toobz were still King a serious experimenter would solve that in an afternoon. 🙂 👍
Thank you for posting your design @KeesB ! Always good to see and try a different approach
If parallel PP EL84 was used .... What value would you suggest for common cathode resistor for B+ of ~300V ? 1k , 860ohm ?
My parallel EL84 uses 1k 20w. The one pair EL84 versions I made uses either 1k8 or 2k depending at what I happened to have at hand. It's not very critical as the 84's are to a high degree self-adjusting. The only thing I really paid attention was to make sure the anode voltage of the ECC 83 didn't exceed 90V. I also used different drivers such as srpp ECC83 (Rc 2k2). Regarding the parallel 84 version you might also use 1k8 or 2k for each push pull pair separately.
The example you cited (Vb 300V) applies to a cap coupled amp. The dc coupled amp uses 380- 400V because there is 100V across the common Rc. This Vb is also not critical (you can go lower than that) but of course the output power becomes a little less. Anyway good luck, I hope you like the amp.
The example you cited (Vb 300V) applies to a cap coupled amp. The dc coupled amp uses 380- 400V because there is 100V across the common Rc. This Vb is also not critical (you can go lower than that) but of course the output power becomes a little less. Anyway good luck, I hope you like the amp.
Thank you for the feedback Kees , appreciated . I will reconfigure for higher voltage , ATM the B+ is ~280V . Sounds really good though . So cool ... SRPP , direct-coupled , PPP in triode mode with no feedback . 😛
Completely incongruous but it sounds so good !!
Completely incongruous but it sounds so good !!
By what mystical property of Toobz does that work?
Just curious, maybe I'm missing something here. 🙂
Self adjusting EL 84's? Take a common powerstage: ECC 83 driving an EL84. When the coupling cap is leaking and there is f.i. +2V on g1 of the 84 the tube automatically shifts its cathode voltage making it 9V instead of 7V. I've repaired numerous Bocama's and Lafayette's with leaking oil coupling caps showing this phenomenon. With every dc coupled cathodyne phase splitter using high values of equal anode and cathode resistors the tube creates its own cathode voltage. If eg the anode voltage of preceding triode is say 100V then the following triode creates a cathode voltage of 102V.
This quality is not the same for every tube. ECC 83's and ECC 82's do this easily, but not the ECC 85. All cathodyne phase splitters using this tube are capacitor coupled. If you do try direct coupling with this tube the sound becomes horrible.
This quality is not the same for every tube. ECC 83's and ECC 82's do this easily, but not the ECC 85. All cathodyne phase splitters using this tube are capacitor coupled. If you do try direct coupling with this tube the sound becomes horrible.
What does that mean? The tubes rises it's cathode current to 9/7 times the correct value, presumably running into gross overload. Do you really call this self adjusting?
Best regards!
Best regards!
No Kay , you misunderstand , Kees is not talking about his design ,he is talking about another design. The self-inverting common cathode circuit does work , when done right . It is an old design from the 40'ies.
See section called "Cathode-driven inverters"
https://dalmura.com.au/static/Phase Inversion Circuits Radio Craft 1948.pdf
https://dalmura.com.au/static/Phase Inversion Circuits Radio Craft 1948.pdf
That brings back memories! I tried to build the version in figure 4 when I was starting with tubes. Never got it to work. Power transformer voltages were not exactly what they were supposed to be (I used a Bocama Lafayette transformer), the current source ECC88 red plated, and I hardly knew what I was doing back then (some 15 years ago). That was the end of it.
I understood him very well. His assumption that a valve can adjust itself by just raising the cathode resistor voltage is incomplete. It may work with high cathode voltages, hence absurdly high wastes of energy (see above).
Best regards!
Best regards!
I would think that the value of the resistor below is too small to develop the correct voltage to balance the 2 tubes , rule of thumb for balanced operation is 5x the value of the resistor were it shared in a "normal" push-pull configuration , ie both grids driven .
@KeesB never claimed the circuit to be the most efficient , but only that he likes "the sound" of the amp , even more so direct coupled and even more so without grid-stoppers .
I have built and listened to a parallel push-pull , triode-mode , with no NFB , but with grid stoppers . It is quite pleasant to listen to driving highly inefficient 80's B&W DM12's (rated @ 86db/w) , compared to my standard Leak Stereo 20 .
@KeesB never claimed the circuit to be the most efficient , but only that he likes "the sound" of the amp , even more so direct coupled and even more so without grid-stoppers .
I have built and listened to a parallel push-pull , triode-mode , with no NFB , but with grid stoppers . It is quite pleasant to listen to driving highly inefficient 80's B&W DM12's (rated @ 86db/w) , compared to my standard Leak Stereo 20 .
Last edited:
It's been a while since I have posted the schematic from Kees "EL84 PP direct coupled". In the meanwhile I have also build the actual amplifier and have lot's of fun building and playing it. This weekend I also have finished a PL519 parallel single ended with also a ECC85 (beta-follower) direct coupled to the PL519 in parallel. Nothing sounds so sweet, clear and real (as if you are there). I am not a mathematic specialist or far in theory to judge weather this schematics are good or inefficient. But my ears are telling me this is the best sounding amp I have listened to lately. I have always focused on Hifi tubes like 45, 2A3, 300B, EL84 and 807. But this TV tube Philips broad to the market in the 1960 is probably one of the best hifi tubes. Playing over G2 as the anode. Thanks Kees for sharing. Like you said with just a couple of components you can make an awsome sounding hi-end tube amplifier. I'll guess our ears are the best instrument for judging. It's not all about the theory. If it were we probably have less inventions. Anyway I will go back to listening and enjoy what 's it all about, DIY builds and music. CHEERS to you all.
More than 20 years ago I also built a stereophonic SE amplifier, powered by a pair of screen driven PL519's which still were relatively cheap then. This amp was heavily influenced by a Bob Danielak (then Svetlana) design that I needed to modify to get rid of some severe flaws. Anyway, I didn't use it a lot, mainly due to it's apalling ineffectiveness. A mere 10 watts of output power doesn't stand in any reasonable relationship with 12 watts of heater power plus 36 watts of plate dissipation - notably for the power tube alone. A screen driven pentode or BPT is nothing else than a triode in terms of effectiveness, and sweep BPT's in general are poor audio tubes, unless they're operated in PP. Anyway, through this built I learned to define my standpoint wrt tube amplification.
Best regards!
Best regards!
BTW , for those who don't know , the American tube that can sub for PL519 or PL509 is called a 40kg6a .
These can sometimes be found at a way more reasonable rate than PL519 .
There was a Austrian man named Gunter Steinhart that lived in South Africa for much of his life who built many iterations of Push Pull amps using PL519 . From 50wpc stereo amps to 400w monoblocks . Fine sounding amplifiers , usually with a Williamson style frontend .
These can sometimes be found at a way more reasonable rate than PL519 .
There was a Austrian man named Gunter Steinhart that lived in South Africa for much of his life who built many iterations of Push Pull amps using PL519 . From 50wpc stereo amps to 400w monoblocks . Fine sounding amplifiers , usually with a Williamson style frontend .
Coupling directly you eliminated the source of dynamic distortions, - coupling cap driving non-linear grid current, shifting bias by the signal envelope such a way. Our hearing is dynamic, and output tubes work at the edge of their maximum.
One more idea: if you connect the cathode resistor to the secondary of the output transformer, you get negative feedback for the upper tube decreasing it's internal resistance, and positive feedback to the bottom tube, increasing it's internal resistance. Such a way the sound will be more single ended than push-pull - like.
As the US grid voltage was and is about 117 V, I'm somewhat astonished about that. I'd expect a 600 mA series heater string in US color televisions, not a 300 mA one? Hence a 20KG6A?
Best regards!