Transformerless push-pull tube circuit??

[quote author="PRR"]

And when people throw "600Ω" around, they usually have in mind levels like +4dB VU, which is like +20dBm or 11V peak. Which means 11V/600Ω= 18mA peak. This is HARD WORK for a small vacuum tube. [/quote]

Arhhhh..... I see! :thumb:

I totally neglected that in my plan, thinking that 2Vrms(driving 10K load) was more than enough. Since I've done some listening test on the previous tube circuit I built, by running headphone output on my CD player into the tube amp then to my power amp. I could never turn the output of the headphone to pass 3 on the dial, without getting totally blasted in front of the monitor.

Like your plan with the 20K plate resistor: even if I did not know what a 12AU7 was, or had some very good tube to use, the absolute maximum current into the load is 250V/20K= 12.5mA, well short of our 18mA goal (if we ever find that historic 600Ω load). Knowing a little about 12AU7, I doubt you have even 8mA flowing in the plate resistor. If we turn the 12AU7 hard-off, most of that 8mA diverts into a 600Ω load, but that's only 8mA*600Ω= 4.8V pk, 3.4V sine, not even +13dBm. Taking 16dB headroom, we can't crank the VU meter past -3dBm, or the -7VU mark on a meter calibrated for the customary +4dBm at 0 VU.

Click to expand...

Thank you so much PRR for the explaination!!!!!! :grin: I so needed that!!! :grin: :grin: :thumb:

Now with a clearer picture of where I should be in terms of the spec, I guess what I need to do is to beef up the output a lot more.

I've lower the bias resistor and add an extra tube to spread the current load in the revised schem below, I think I will attempt to build this layout with a 2 x 12A6 Beam tubes, a 4 x 6BA6 tubes circuit and also a 4 x 6SK7 circuit any thoughts/advise on this ?? Too much heat?? :green:





I've also got some 6SN7, 6SL7 and 6AS6(CV4011) and 6AK5W tubes, I think I will use these to build differential and 2 stage RC coupling as pre amp stages. Any thoughts on these tubes as pre amps?

There ARE good autio interfaces that don't cost a pound of iron. TASCAM -10dB 10K unbal works very well in small studios. Hell, it is how 99% of HiFi playback systems use.

Click to expand...

Where can I get these? Are they any different to the Edcor transformers?

:guinness:

>> There ARE good autio interfaces that don't cost a pound of iron. TASCAM -10dB 10K unbal works very well in small studios. Hell, it is how 99% of HiFi playback systems use.

> Where can I get these? Are they any different to the Edcor transformers?

Huh? Transformers? Where?

This IS the HiFi interface: all those RCA plugs on the back of your home HiFi CD player, tuner, receiver, cassette deck, etc. Maximum signal level is 2V RMS, 2.8V peak, nominal (VU dynamics) level is -10dBV. Aside from HiFi rigs, this was key to the first artist-price studio gear: the Tascam 4-track tape machine and the Tascam Model 8 console. The backs were peppered with RCA jacks, and signal levels were just-right to interface with home HiFi EQs and reverbs. An early Model 8 has mike input transformers but no output transformer anywhere.

12A6 sure is affordable. The load/power curve shows 2.5W at 4K load, 1.25W at 2K load, apparently smoothly declining with load. Going down: 0.6W at 1K, 0.3W or +25dBm at 500 ohms. That's good power, but the low-Z distortion plateaus around 15% THD, a rather "colored" (some would say "nasty") output. You'd probably be below 15dBm before THD fell to 1%. You better like "phat toob soundz".

And that's transformer-coupled. You could choke-couple, but that's no cheaper than a transformer (and much less versatile). You could resistor-couple... let's see... since the plate voltage swing is nearly zero compared to 250V, you could let the resistor drop say 60 volts. Plate current is 30mA. That means a 2K plate resistor. This is in parallel with your 600 ohm load, so counting on my thumb the tube sees 2000||600 is around 400 ohms. Power is down to +23dBm at 15% THD. It will "work", but seems nasty.

Also the output impedance is the pentode's 70K in parallel with the 2K plate resistor, or a wee bit under 2K. That means if it is set to deliver 1V unloaded, it makes 10K/(2K+10K) 0.83V in 10K load, 600/(2K+600) 0.23V in 600 ohms. It is arguably correct (though no longer common) for a 600 ohm load to drag a "600 ohm" output down to half, but not to 1/4. And the high output impedance negates some of the point of hot 600 ohm line drivers (it won't suck-out interferance as well as a 600 ohm or 60 ohm source impedance).

It works a little better as a triode. Still, Rp is 2.5K. Even with a 2K plate resistor, that's a 1.1K source. Your peak current, even with 250V supply, is rather small, even if you abuse the 7.5W plate rating.

Oh, heck: no wonder. This is only a 2 Watt heater. Weeny cathode.

Stick a single 2A3 in your simulator. Use 200V supply, 1K plate resistor, adjust cathode resistor for 75mA cathode current (try 250 ohms). 2A3 plate resistance is around 800 ohms, so output impedance is a little under 600 ohms (maybe 450 ohms). Gain will be small, like 2 or less. THD won't be horrible at +20dBm. An actual 2A3 is costly and awkward, but 6EM7 has a big triode a lot like 2A3 plus a small triode a lot like 12AT7. It is a mini audio power amp in a bottle.

OK, that drives 600, but we wanted "balanced" output. (Actually, floating would be better, but that is impossible to do right without a transformer, and often not essential.) You need two such high-current drivers, two 2A3 or 6V6 or 6EM7. And even they will strain bad, so you may want to take some feedback around them to prop-up the THD numbers and reduce loading droop.

NONE of this is new-news. Driving long line hard is tough to do when you have to drag electrons through empty space. Transformers give you a LOT of leverage to turn the tubes' high-voltage low-current ~5K resistance into something low enough to drive line impedances. The classics all used the iron. Even when they did use "advanced"(?) topologies like the push-pull WCF in the LA-2, they used iron to transform 600 ohm loads up to ~2K at the tubes. More often they transformed up to 5K, 10K, even 20K to the tubes. That makes the tubes' ~5K dead-loss resistance "small" compared to the (transformed) load impedance.

[quote author="PRR"]
This IS the HiFi interface: all those RCA plugs on the back of your home HiFi CD player, tuner, receiver, cassette deck, etc. Maximum signal level is 2V RMS, 2.8V peak, nominal (VU dynamics) level is -10dBV. Aside from HiFi rigs, this was key to the first artist-price studio gear: the Tascam 4-track tape machine and the Tascam Model 8 console. The backs were peppered with RCA jacks, and signal levels were just-right to interface with home HiFi EQs and reverbs. An early Model 8 has mike input transformers but no output transformer anywhere. [/quote]

Right, I wonder where I can find a audio hi fi junk yard to harvest these parts....... I should try to get some tips off CJ in how to scope out a hifi junk collector to buddy up with! :shock: yo got anotherone to spare CJ!??? :green:

12A6 sure is affordable. The load/power curve shows 2.5W at 4K load, 1.25W at 2K load, apparently smoothly declining with load. Going down: 0.6W at 1K, 0.3W or +25dBm at 500 ohms. That's good power, but the low-Z distortion plateaus around 15% THD, a rather "colored" (some would say "nasty") output. You'd probably be below 15dBm before THD fell to 1%. You better like "phat toob soundz".

Click to expand...

mmm......I donno what that sounds like but I am keen to find out! :grin:

And that's transformer-coupled. You could choke-couple, but that's no cheaper than a transformer (and much less versatile). You could resistor-couple... let's see... since the plate voltage swing is nearly zero compared to 250V, you could let the resistor drop say 60 volts. Plate current is 30mA. That means a 2K plate resistor. This is in parallel with your 600 ohm load, so counting on my thumb the tube sees 2000||600 is around 400 ohms. Power is down to +23dBm at 15% THD. It will "work", but seems nasty.

Click to expand...

The lowest distortion point on the data sheet is 7% with a 7K ohm load disspating 3.35W, does that mean this is the absolute minimum distortion you can get out of this tube???? Can feedback help to lower the distortion???

Stick a single 2A3 in your simulator. Use 200V supply, 1K plate resistor, adjust cathode resistor for 75mA cathode current (try 250 ohms). 2A3 plate resistance is around 800 ohms, so output impedance is a little under 600 ohms (maybe 450 ohms). Gain will be small, like 2 or less. THD won't be horrible at +20dBm. An actual 2A3 is costly and awkward, but 6EM7 has a big triode a lot like 2A3 plus a small triode a lot like 12AT7. It is a mini audio power amp in a bottle.

Click to expand...

Will try that after DL the sim from duncamamp.com.

Back again for some more guidance.....

I am trying to figure out how to bias the 12A6 tube by plotting the load line on the data sheet, does this looks right???? :?

> plotting the load line on the data sheet, does this looks right????

Looks odd. What are you plotting?

A "low"-distortion line, transformer-coupled, 250B, 30mA, 7K, would be about like the blue line:



Distortion drops with level. Not in any simple way, but for a gross approximation assume that a naked tube's THD drops about as fast as power. 7% at 2 Watts might be under 1% at 0.2 watts. Often better when looking at mass-market pentodes: they were often rated for a power level will into saturation or bottoming, level just a little lower might drop the THD number from 10%-15% down to 5%.

[quote author="PRR"]
Looks odd. What are you plotting? [/quote]

:green:

A "low"-distortion line, transformer-coupled, 250B, 30mA, 7K, would be about like the blue line:

Click to expand...

If I remember correctly from my transistor biasing class, to plot the load line by connecting the point of my power supply voltage to the current output at the collector set by the load.

I am looking to eliminate the use of negative voltage to bias the grid, therefore the red line to make use of the 0V grid voltage. Looking at it now, I can see that there is at least 50mA of current being wasted at all time generating unecessary heat........

Looking at the graph, each increment of 5V at the grid does not seem to have a linear output of current. Wouldn't this be a "better" point of biasing at -5V according to the graph in respect to linear current output? except that it is wasting at least 30mA at all time turning into heat........ bah!

Would this be the cause of the fat toob sound?? Due to not linear output of current vs the input grid voltage???