Platenspeler.Com |
 |
Platenspeler.Com |
|
HeadHunter
Page 2: December 22, 2006
After taking some decisions about this project on the previous page, such as 6AS7 tube to use, it is time to determine a good circuit for the first/prototype version of the headphone amp.
The simple solution to building an amplifier with a 6AS7 is as follows: Use half a triode as a driver and half of the 6AS7 triode as the power tube. Building this circuit still requires some serious thinking about the powersupply and the enclosure (heat problems).
Simulations tell me that the circuit above will give a gain of 14.96dB at 5Hz and at 20Hz 15.30 dB which is constant until 30kHz. Now we all no that simulations are not reliable but at least this is a good starting point. The plate current is 50mA for the 6AS7.
Building the more difficult version along the lines of the Aikido board (which I have at home in the 9-pin/noval version) will require even more thinking about heat dissipation, power supply dimensioning and heater supply (AC or DC).
My motto is: Travel Lite. I do not like all the very dificult power supplies and other circuits if these are not necessary and do not add in some way to the sound. On the other hand, the circuit above is a little bit too simple for my taste ...
I would like to use DC coupling for this design, which is easier when using the "SRPP" variant used in Akido, since the output taken on the plate of the lower tube is exactly at half B+ and if we a similar approach in the power stage with the cathod follower on top and the other halve of the tube as cathode resitive load. Although strictly spoken this is not a SRPP design, I used this setup a while in Jerry and went back to the SRPP in order to get a lower output impedance.
The circuit above still has a relatively high output impedance. We could do with lower values for the cathode resistors but then I have to take care of the maximum current of the amp. Looking at the datasheet, the grid needs to be at -35V for a plate current of 80mA.
In order to have the grid at -35V (at 125mA plate current), the cathode resistor needs to be at last 35/0.125 = 280 Ohms. For the moment we'll use 330 Ohms as this will prevent our amp from frying itself. Plate voltage is about 135V in this case. In order to get a lower output impedance, C4 would be necessary, and maybe we would get rid of R10. However, for the moment I will leave C4 out and R10 in. We'll see later how the amp behaves if we make a modification.
The powerrails are at 240V. R9 and R10 will bias the tube at about 80 mA. This means a voltage drop of 26.5 V per resistor (about 52 Volts in total). For both halves of the tube, about 90V plate-cathode remains which is OK for the 6AS7 tube. One point of concern is to get the PSU to deliver enough power, still do some necessary filtering and not make the total circuit too complex and heavy. Since the used circuit has excellent PSRR (Power Supply Rejection Ratio), I will not worry too much about this in the prototype. We'll see whether the setup holds for the final version of the amp.
Since the 6AS7 requires 2.5A for the heaters, I started with using AC for the heaters. This means careful layout of the amp in order to prevent Humm.
The DC powered alternative is a high-power voltage regulator, which becomes very, very hot as well. I've ordered the LT1083 voltage regulator which has a power rating of 7.5A. Using this voltage regulator I'll test the amp with DC heating as well.
Given the relatively high PSRR of the circuit, the power supply can remain relatively simple. It is powwible of-course to build the amp using one transformer, which saves space but may be more expensive than two separate transformers for heaters and plate supply.
NOTE: Please not that the 12V output for the heaters means that the 6AS7 driver tubes and the 5670 or 6N1P tubes in the first stage are connected in series.
Well, the proposed PSU might in the end be a little too simple, but on the other hand it's never good to add too many components.
Temperature IS an issue when building an amp with the 6AS7 tube. The tube will dissipate 6*2.5 = 15W from the heater supply and another 20W from plate current. So if we have two in one box, the total amp might dissapate in the order of 100W. That's a lot! And what's more: In the prototype I use DC to power the heaters and therefore you need an LT1083 voltage regulator (7.5A) to deal with the current requirement of about 4 A. Should we therefore switch
Should the AC powered heater solution not work I expect that the aluminum box I selected does not provide sufficient cooling for all components in that cramped space. On the other hand, I hate the idea of using one of those enormous wooden boxes.
I will probably build the amp in two boxes: Amp and Powersupply. This will leave me the option to use DC for heaters and better regulation of the B+ rails.
| << Page 1; HeadHunter Intro |
^^ Back to DIY Page ^^
|