Page 2: Circuit Description
January 4, 2007
Given the fact that the 832 asks for push-pull type of amplifiers, I had a look at the following circuit defined by TubeCAD. Of Course I would need to change the design in a way and make a mix between the Coldwar and some other designs but it was a start.
The idea started with this design which is not quite common as both halves of the double triode share the same cathode. The circuit below is basically a push-pull amp where the output transformers have been exchanged by resistors R13 and R14 in order to generate a certain load.
The circuit above is just a braindump which I'll have to simulate in order to test its feasibility, but of-course this was not serious. There are not at all many amps built with the 832A tube and none of them is a headphone amp. And as I will not build this amp with an output transformer, I wonder whether I need the extra swing provided by the push pull architecture. And apart from that, the architecture does not offer the befenits of cathode followers (as these are shared). So the alternative is to use both halves in parallel like I did with the Headline.
For the moment I'll make some calculations with a parallel architecture. Needless to say that using one halve per channel of one 832a tube is not possible since the cathode is shared between the two halves of the tube.
Since the grid current is 6mA, we need a good driver tube to provide the required amount of current. And also I want to recalculate the grid voltage of both the 832A tubes.
The typical plate current for the 832A is 40mA per channel (80 mA per tube). And since I would like to prolong the life of this tube we will use a lower plate current of 70 mA per tube (35 mA per channel). I used a standard Amplimo transformer with 2 * 115V secondaries to generate the B+. After rectification the voltage is 300V (loaded) after which I used two resistors of 680R as a basic CRC noise filter. The resulting B+ for the 832A tubes is 250-260V. Looking in the datasheets this may be a litlle on the low side, as the tube is rated at 450-600V for continuous operation. But as lines are straight already for 250V plate voltage this will do.
For the output tubes this means that we need a cathode load of about 700 Ohms in order to get the desired current.
The plate voltage for the driver stage and the screen of the output tubes needs to be lower than the 250V used for the output tubes. Therefore another resistor (R6) is used with a value of 1K to 2K2, here the screen voltage is taken and the plate resistor of the driver tube is also connected.
The screen therefore is fixed at about 230V (which may still be a little too high)
BTW, should we at a later stage decide to use a SRPP stage for the driver,
then the plate voltage of 230V will probably too high.
Also, when building the final version, I changed the value of R6 tot 2K2 Ohms
and R8 to 390 Ohms in order to lower the voltages of the driver tube and the
screen even further.
According to Remco's datasheet:
Class-A audio push-pull
Anode voltage 250 V
Anode current 40 mA
Grid voltage -17.5 V
Grid current 6 mA
Screen current 2 mA
Anode dissipation 15 W
Screen voltage 200 V
Power output (2 tubes) 14 W
I've decided that should I not find over the next few days a very good reason to use a push-pull architecture with this tube that I'm not even gonna bother about any other architecture than a cathode follower (which can be either resistor or FET load).
| << Page 1: Project Description |
^^ Back to DIY Page ^^
|