• 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.

Found a new interstage coupling method (?)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Guys,

In my personal search for removing the interstage coupling cap I have found a different way to couple between valve amplifier stages, a way that I have not seen anywhere before. Because I am not sure if this is new or not I would like to ask you if you have seen anything similar to what I will describe in this topic.

Because I am careful not to disclose my design yet I will only describe it as good as I can. Unfortunately the topology is so extremely simple that I cannot even say that much about it, but let's give it a try.

I have noticed that it is possible to shift the DC voltage between two stages without using a cap or transformer, while retaining the original AC signal. The topology that I have found uses medium-mu valves to couple between two amplifier stages.

I have found the topology by using SPICE and I have build a Single Ended stereo amplifier to try out the topology in real-life. In my test amplifier I am using a simple grounded grid voltage amplifier with a ECC83, the coupling valve is a 6CG7 and the output valve is a 6550. The DC input voltage for the coupling stage within this amplifier is +160 volts, the output DC voltage is -43 volts.

The coupling stage is capable to drive the grid of the 6550 positive. In my simulations I have noticed that it will go positive for as long as the grid current of the output tube is not too much. For real subclass 2 operation an extra buffer stage should be implemented to allow for true subclass 2 operation.

The coupling stage does not show changes in resistance with varying frequencies and also shows a zero phase shift.

Because the coupling valve is also responsible for setting the idle current of the output valve I have been looking for a method to use a DC servo. I am now in the middle of developing a push pull version in which I will make use of a DC servo, connected to the coupling stage, to set the idle current of the output valves.

The topology described here is not the same as the topology as described by Darius-Karim Mottaghian-Milani (url: Coupling Triodes). I have sim'd his topology, didn't like the results and dropped his idea.

Please forgive me if I have overlooked certain properties. I am not technically educated, so it could be possible that I am not aware of certain key properties that require explanation.

I would be very interested to know if any of you forum members recognize what I am doing, and perhaps could provide me with examples of similar topologies. Also any other feedback is more than welcome.

Best regards,
Anthonie

PS.
I have attached a few pictures of my test amplifier. I do realize that there is a cap on the circuit boards that might look like a interstage coupling cap. This is a 10u cap that is used for de-coupling the high voltage for the first two amplifier stages.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 
I was going to guess the inverted triode scheme (long known but obscure, ie the last tube in the Darius et al scheme using a low Mu tube), but since you say yours is different ... I can only guess that you have reversed the plate and grid in the Darius' scheme and put negative V on the plate to reduce the grid current. Steve Bench played around with some flipped tube configurations, but I think they were oriented toward still getting some power gain from the stage. Might want to check out his stuff.

Or, there is a scheme around that just uses the tube as a diode to drop voltage, used for cathode bias sometimes, usually using damper diodes.

A current source through a resistor also works fine for dropping voltage. Needs a SS CCS to get to useful low enough voltages for grid coupling though, unless you were to run a pentode or high Mu triode CCS from a negative cathode level.

Probably could also just put a resistor voltage divider across a triode: plate to cathode divider with the middle tap to the grid. Sort of a shunt V regulator. A top zener in the divider would stiffen it up more.
 
Last edited:
Hi Guys,

In my personal search for removing the interstage coupling cap I have found a different way to couple between valve amplifier stages, a way that I have not seen anywhere before. ...

I think the term is "direct coupled amp" or sometimes they call it a DC coupled amp. I think you are about 50 years to late to invent a new valve amp topology.

There are many variations of it. but in every case the trick is to bias the second stage so the output of the preceding stage can be directly connected to the grid of the second stage.

Williamson might have beed first to use this? Actually I doubt that but Williamson is the fist really well known amp to use it

The Williamson Amp, part one | Preservation Sound

Another common example of this is seen in the Dynaco ST35 there the gain stage triode is directly coupled to the grid of the phase splitter. But the second tube could be a gain stage too.

Dynaco ST35 Tube Amplifier Schematic and Manual


Using only two triodes there are only a few ways we can connect wires to six terminals. I think it comes to something like only 50 possible topologies. Many of them already have names.
 
The second tube (6CG7) is acting as a DC Level Shifter.

A floating shunt V regulator would be a good fit. (resistive V divider across the tube, plate to grid to cathode)

On the other hand, I see a bunch of heatsinks on the board which could be CCS's. So just a triode configured as a resistor would work to drop a fixed voltage using a CCS source.

Have to agree on the Tektronix scope designer or radar designers comment. If it is practical and works well, DC coupling would have been their forte.
 
Last edited:
If I read the op post correctly, he has eliminated the coupling cap to the output tube(s) as well.

So it is a fully direct coupled amp.

The second tube (6CG7) is acting as a DC Level Shifter.

The usual problem with direct coupling is headroom. You use up a lot of it to get the biasing done.

I think you just might want a coupling cap. How else to roll off the low end and keep DC out of the transformer? If the input line level has any bias on it or infra-sonics from turntable rumble. Some place you need a few caps.
 
If the tube is used as some kind of diode it's dynamic resistance is not zero, and it is not linear, so a current source from it's cathode to the negative supply would help. But in such case resistor instead of the tube can be used that does not have filament and capacitance between it and output tube's control grid. But in both cases added non-linearities may be higher than non-linearities of removed capacitor. However, dynamic distortions caused by coupling capacitor loaded on non-linear grid input are absent.

If you want servo you can use pentode for current supply, with it's filament in cathodes of output tubes, but they must work in class A otherwise dynamic distortions will be introduced back. Actually, the servo that senses average current of output stage will introduce dynamic distortions back, so there is no point in eliminating of coupling caps in such case.
 
Last edited:
:drink:
 

Attachments

  • coupler.png
    coupler.png
    6.4 KB · Views: 457
Last edited:
TheGimp said:
Well you know, even if it was developed 60-80 years ago and he wasn't aware of the design, then he is still due credit for developing it independently.
Yes. That is a good test for whether you have real understanding. Can you independently invent something/discover something which you then find was already invented/discovered and is regarded as a good/valid idea? If so, you are thinking at the level of the state of the art some years ago which means that you are not far off the frontier of knowledge today.

On the contrary, if you keep 'inventing' ideas which have been thought of but discarded many years ago it means you still have some stuff to learn, so you can get better at spotting the flaws in a scheme.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.