Hiend Audio, DIY, Hi-Fi, Stereo, Electronics site, for lovers of high-fidelity music reproduction. High-End Vacuum Tubes. Silicon. Do-It-Yourself Audio Systems. Schematics: amplifiers, speakers, horns, CD, DAC. Circuits, Topologies, Acoustics, Cables, Speakers, Voltage Regulators, Equipment Upgrades, Modifications, Problems, Solutions, Tips, Tricks. ------ Dla pasjonatów Hiend Audio, najwyższej wierności odtwarzania muzyki na sprzęcie budowanym, modyfikowanym samemu. Wiedza: elektronika, układy lampowe i krzemowe, wzmacniacze, głośniki, kolumny, tuby, DAC, kable, DIY, porady, sztuczki, problemy, rozwiązania, tor audio.

«

»

Aug 23 2014

Print this Post

Technics Futterman Tube OTL Hybrid

A Super-Triode ….  That would be a Triode with some kick-ass current booster associated with it.

Such as a Hybrid Tube.  A tube that is connected to a gear-box.  A current gear-box.

For each and every 10 miliamperes of tube current – You define the gearbox ratio – you get, say, a 100:1 current gear ratio, i.e. 1 Ampere of kick-ass current on each and every of the Mosfets of the current booster.

Or maybe 100:1 is too much?  Would we not “loose” too much of the “tube” sound then?
Indeed, this is a tube output stage amplifier.

Hey !   Are you JOKING MAN ??? …

Technics Futterman OTL Hybrid v1.0b

… and what about all those mosfets hanging around there ???

Well, strictly speaking, they are “not mosfets”.  :)
They are “SLAVES”.  Slaves that do what they are told to do.

So … if they are slaves, then who is the MASTER ??

Well, the master is a tube. It is the tube that gives orders and determines the ultimate sound of such a setup.

How come it’s the tube that is giving the orders?
Easy. It is a 100% type of feedback setup. Something like a “Darlington” …. but the driver transistor is a tube. One catch here though.

There is no such thing as a “PNP” or “complementary” vacuum tube. That is why I suggest using those helper transistors at the upper anode and lower cathode. They allow to “convert” one of the tubes into a “PNP” type, of sorts. Very much similar to the “Quasi-Complementary” transistor pairs of an audio era_long_gone and of the past, where there was no chance of purchasing PNP power transistors, suitably “matched” with their NPN complements … So what people did in those ancient times was to try to “emulate” a PNP device by using a special kind of setup with the use of a low power PNP driver transistor. That was the “quasi-complementary” output stage.

So, what do you think of the Technics Futterman Tube OTL Hybrid?

Ah, by the way.  Although you could “swear” that that inverter stage does not have a ghost of a chance of working properly, that it inherently “seems” to be providing uneven signal drive, with one tube “amplifying” an the other “cathode-following”  … believe me, this particular topology: yes, it actually DOES work. It is the so called “Inverted Futterman”, or “Technics_Futterman” inverter stage. The thing is, that by “grounding” the inverter to the speaker output, and NOT to ground, as well as by criss-crossing the capacitors feeding the individual driver output tubes, you get a symmetrical gain in both of the tubes. YES. It is a setup, where BOTH the tubes, the bottom one as well as the top one, operate as “cathode followers” of sorts.  Although it may seem that one of the tubes is a common cathode, whilst the other a common anode … well, not all is what it seems.

For shorts: believe me. It WORKS.   I use the Inverted Futterman in my all-tubes OTL. Works quite well.  please be advised that some call this topology a “totem pole”, and If you want to dig deeper into the subject, then this is the right place to dig into:

http://www.tubecad.com/articles_2001/totem-pole_output_stage_psrr/Totem-Pole_Output_Stage.pdf

The first version of the schematic that I depicted above works off a single pair of +Vcc and -Vcc rails, but as you may well expect, the amount of wasted heat and stress imposed on those mosfets, due to the excessively high rail voltage, as would be necessary for the tube to operate, the thermal / power stress and heat waste would be immense and not very neat. So, a more practical version actually requires a dual rails type of setup, allowing to cut down on the heat dissemination on those mosfets…

So, here are some more hybrid OTL Tube-Mosfet musings …. but this time, with a double pair of power supply rails. This way, we can significantly “optimize” and minimize the heat that is lost on those mosfets:

Technics Futterman OTL Hybrid v2.0b

Oh, and by the way … Where the hell is that “GEAR-BOX” ?

Look closely at the cathode resistor R2.  Calculate the voltage on it.
It is  V_R2  =  R2 * I_Tube_Cathode_Current.

Now, think about the Voltage on R3, R7, R9, R11, R13, of the last drawing, each of those being a mosfet source degeneration resistor. The voltage on top of those is of a slightly lower value:
V’ = V_R2  - Vgs

This is a reflection of the fact that there is a Vgs gate-source voltage drop between the gate and the source of a mosfet.

What is the current flowing through each of the Mosfets and associated mosfet degeneration resistors?  It is such, that  I_Mosfet  *  R_mosfet_degeneration_resistor = V’.

So, basically, the relation between R2, essentially being a tube’s cathode resistor, and each of the mosfet degeneration resistors, establishes a certain “proportion” of current distribution.

R2 * I_Tube_Cathode_Current – Vgs = I_Mosfet * R_mosfet_degen_resistor.

Hence:
I_Mosfet =
( R2 * I_Tube_Cathode_Current – Vgs ) / R_mosfet_degen_resistor

I_Mosfet =
I_Tube_Cathode_Current * {R2/R_mosfet_degen_resistor}  - (Vgs/R_mosfet_degen_resistor).

So, indeed, a gearbox of sorts it is. The ratio of the gearbox is defined by the relation of cathode resistor R2 to the R_mosfet_degen_resistor.

A similar scenario applies to the elements on the lower half of the drawing, with the complementary mosfets, the “complementary” tube and the -Vcc supply rail.

Well, You could argue that it is not a truly  a “clean” gearbox. If it were a “true” gearbox,  the ratio of the currents would *SIMPLY* be a result of a ratio of the resistors, with no “additional term” hanging around at the end of the formula.

So, If indeed, that is your preference, then check out this “clean” version, which is based on BJT power transistors. BJT transistors tend to have a “voltage drop” of Vbe approximately equal to 0.7 Volts(*). But this value just happens to coincide with the voltage drop upon a forward biased diode:

Technics Futterman OTL Hybrid v3.0b

(*) { OK, OK, that was a really brute simplification, as we all know that the Vbe drop on power transistors sometimes significantly higher than 0.7V – especially, if they are fairly low gain devices, and exposed to a very high collector current. This simplification is just for the sake of explaining the general idea}.

Here, you have a “clean” situation. The Vbe voltage drop of the BJT’s is offset by the voltage drop on the diode, which has been added into the tube’s cathode current path.  So, in such a case, the “current gearbox”, or the current “spread”, essentially boils down to the proportion of R_cathode / R_BJT_degeneration resistor.  This is because that now the voltage on top of R_cathode and R_BJT_degeneration resistor is essentially the same voltage. In both cases(*) it is 0.7 Volts less than the voltage on the cathode.

So, for example, if for every 50 mA of cathode current you want that to translate into 1 Ampere’s (1000mA) worth of BJT booster current, that means that you need a gearbox with a ratio of 20:1. So, if the cathode resistor is say, 10 ohms, then each respective BJT emitter degeneration resistor would need to be 10/20 = 0,5 ohms.

Or, if you turn it the other way around:  If you insist on BJT degeneration resistors to have an optimum value of say, 0,33 ohms, then 20 times that value would result in a cathode resistor of 20 x 0,22 = 6.6 Ohms.

Neat.

A Tube amplifier capable of multiple amperes worth of kick-ass bass ….

{{ 2014-08-25 Update:

Ok, after some additional musings, I came up with a piece of front-end, so as to visualize, how the whole concept could look like.  Mind you, there is yet one minor “issue” that still needs to be resolved, that being the joining of two conflicting design goals, pertaining to the output driver tubes:

a). These tubes should be appropriately biased. It would be “nice” if they could be “auto-biased”, without the need for some fancy and FLOATING external fixed bias dedicated power supplies.

b). The cathode “auto-bias” resistor should remain within some fixed “gear-box” relation to the emitter degeneration resistors, so as to control the proportion, the split of the currents.

Presumably, some fiddling with diodes, with LED’s or yet some other innovative concept could render the result of BOTH an appropriate auto-bias voltage, and a convenient current split. Play around with the concept and see where you get from there.

Please also note that the paraphase technics / inverted Futterman, together with the rest of the input stages, is actually supplied from a yet higher B+ voltage.  It actually NEEDS to be supplied from quite a higher voltage. This time, in the whereabouts of 380V DC. Why such excessive B+ voltage for the paraphase ?

This is so as to provide an ample voltage swing on the output capacitors of the inverter.   Take note that the “ground” of the inverter sits atop the output signal, atop the “red” output speaker terminal, so some additional headroom is required to service the swing in both directions.  Especially in the case of the maximum positive output swing, where the “net” operating voltage on the paraphase splitter is actually “decreased” by the value of the peak positive swing of the output voltage, as routed to the speaker. The phase splitter is NOT grounded to the ground, but it is “grounded” to the positive speaker terminal. Remember ?

Ah, and one other thing. The PS that supplies both pairs of the complementary +Vcc and -Vcc rails is somewhat “floating” – in the sense that the capacitor filter bank’s midpoint is NOT connected to any transformer mid-tap or anything. There *is* no center tap. Just a plain rectifier, pumping into the +Vcc and draining from the -Vcc.
The ground centerpoint, within the filter itself, is somewhat virtual and “derived”, as can be seen on the schematic below. Therefore, a need for the “symmetry” measurement tabs arises.
Take note that the Technics / Inverted Futterman paraphase inverter actually “injects” its own current and hence “loads” the lower of the driver tubes with the burden, of draining this additional current too; ie. with the burden of a total current which is slightly greater than the current “sourced” by the upper driver tube. The lower tube conducts more current than the upper one. Therefore, a certain level of “asymmetry” may be observed on the +Vcc and -Vcc rails. This level of asymmetry should be counter-acted by means of slightly varying values of the bleeder resistors within the upper and lower capacitor filter bank, respectively.

}}

Technics Futterman OTL Hybrid v4.0

 

 

Permanent link to this article: http://hiend-audio.com/2014/08/23/technics-futterman-tube-otl-hybrid/