• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

A VCCS tube amp, how?


2007-11-16 5:41 pm
I'm interested in designing an tube amp for ESL. After reading some theory, I've found out that in order to get flat frequency response, you have to use current drive. So I began to search for a way to design a VCCS tube amp.

I'm a relative newbie in electronics and can't yet think in detail efficiently, only on level of several standard circuits, maybe somewhat better. So the problem is, I haven't found a good, or at least, well explained way to make suck an amp.

What i've found, is that VCCS comes in two ways - an inherently VCCS design, or a feedback - controlled one. I'm more intrested in the first approach, as from what i've heard, current-controlling feedback is hard to harness and tune, but i'd be glad if someone would have some idea based on it.

One of the problems is that transistors seem to be out of the question entirely, due to really high voltages involved. The first (and only, as I've discovered the need for VCCS amp only recently) idea i've got was to simlpy load a tube's anode with a current source, and derive signal from anode, but it didn't seem to work as I thought.

Thanks for help in advance.


2007-11-16 5:41 pm
I know it's dangerous, I don't have intent to do it asap, I dust want to design it, and then, when i'll have enough experience, i'll try to make it.

Besides, a VCCS amp for fullrange panel isn't going to be very powerful, high impedance at low frequencies must limit is maximum current swing.
This is definitely a dangerous project to undertake.

I would suggest you try making up something in transistors to drive a normal speaker to at least realize the theory behind transconductance amplifiers.

One of the easiest ways to implement this is to use an amplifier with a differential input and implement the feedback loop to measure current in the load (speaker) instead of output voltage as would normally be done.

A transconductance amp for ESL speakers will work in the same fashion for the most part, but for the best results must have a balanced output which complicates things a fair bit. I don't think this would be easy unless you are very confident in your design skills from the standpoints of stability, safety, and proper circuit design.
If you plan to drive the panels directly from the plates of the output tubes, then you will need many kV of swing, so I agree with others that this is probably not a first time project. OTOH, if you use a step up transformer then the project goes from very dangerous to merely dangerous.

Nevod said:
The first [...] idea i've got was to simlpy load a tube's anode with a current source, and derive signal from anode, but it didn't seem to work as I thought.

That should be a good start if you use tubes with very high plate resistance (i.e. pentodes.) Bias them with unbypassed cathode resistors and you'll increase the output impedance even more. Put that through a 1:5 or 1:10 step up transformer and you should be well into the hundreds of thousands of ohms output impedance.

Really, you don't need the CCS on the plate; just use a step up output transformer instead of the more usual step down. If you apply any loop negative feedback make sure it's current derived instead of voltage derived.
Ak! It's difficult for me to imagine a VCCS that's not a feedback control system. Below are some links to some good VCCS theory. They mostly show opamps. But the feedback control theory aspects and overall topologies should still hold true.





That last one has a 'historical' link to some Philbrick stuff, that probably includes some vacuum-tube VCCS circuits, or links to other material including them (I haven't checked).


2007-11-16 5:41 pm
Thanks! Will read that. :)

Yes, the Acoustat X has an OTL amplifier, but it doesn't seem to be VCCS - on the circuit's i've got, it is stated to use tube op-amps in output. Though, it may be VCCS, as it's output current is low, and at once, it is said to be fullrange.

Dave Cigna
Well, i'm not going to implement it right after designing. I'll have to get some experience first, and only then, i'll try it.

Well, an obvious idea came to me - use two current sources in series, one CCS, and another controlled by input voltage. First tests have proven that it does work, but current isn't constant - i have selected tubes with way too low mu and resistance. Have to do some more simulation.