Most of the work that I have done in the past with vacuum tube and solid state electronics has been repair. So, I have ventured into the realm building. Building from scratch is nothing like repairing. It takes time, a lot of thought and reasoning go into a build no matter how simple it may look. I have been tinkering most of my life with this hobby and I think I forgot much! My advice to anyone thinking of building an amp from scratch is read first, ask questions later. Even if you have been around this stuff most of your life (I'm 60) you can never know enough. I recommend three good books. They are pricy but hey it is easier to learn from the other guy's mistakes not your own! The first book is Building Valve amplifiers by Morgan Jones, and also The Tab Guide to Vacuum Tube Audio by Jerry Whitaker. The third is not for a beginner and contains a lot of theory, but if you have a basis in electronics and can do the math it will fill in the blanks. Fundamental Amplifier Techniques with Electron Tubes, by Rudolf Moers. In the past, I have found that using off the shelf parts can give great results no matter what you are building. My idea here was to make a low power amp keeping it a simple as possible with nice sounding output, and I was pleased at the result.
Specs of my build are;
- Edcor XPWR193-120 Power Transformer 240V, 200mA and 9V, 4A
- Edcor GXPP15-8-8K Output Transformers used in ultra-linear configuration.
- Glassware Audio PS-14 power supply kit which supplies the high voltage and also regulated 6.3 VDC for the heaters.
- I used a familiar circuit, the classic Dynaco 6V6 output circuit with a floating paraphase inverter . It is a simple and effective circuit. I say let your ears decide and that is all that counts. I am pleased with it. I've listened to some of the other tubes and I like the sound from the 6v6's.
- You will notice the controls. That is a Glassware Audio A3-Mini Stepped Attenuator which fits the bill nicely.
- Hammond 156R choke. 1.5H 200mA 56 ohms. Hammond 1444-14103 enclosure.
PC Board. There are a couple of versions of this board out there. The second revision has gold plated pads but as to the quality of the gold, that is questionable. This board is laid out simple enough with provisions for different physical sizes of capacitors. If you wanted to modify the Dynaco circuit you could make changes in your pcb design program and get the boards made from Dirt Cheap Dirty Boards for a very reasonable price if you do not mind the wait from China. There has been a lot of discussion in the forums on the floating paraphase inverter. Many have built this circuit and are happy with it. But many will also point out it's deficiencies with detailed technical explanation. But I have yet to see actual bench test results from those who would criticize the circuit, or better yet, blind test results.
The schematic that has been floating around the internet is wrong. If using a 16 ohm output transformer the feedback resister should be 10K. Since I am using 8 ohm output transformers the math works out to about 6.8k for the feedback resistor and the capacitor close to 1500pf. Since that is a hard to find item I used a 390pf in parallel with the 1000pf.
So this is the chassis ready to be painted. The back of the chassis has two vents along with vents in the bottom cover to facilitate a little cooling inside.
6V6 boards installed. If you are careful with the layout the sockets line up nicely with the boards read to solder in.
Glassware power supply board installed. Normally the heat sink and regulator transistor mount directly to the board. I mounted the heat sink to the top of the chassis and extended the leads on the regulator to reach the PC board. since the heat sink is now grounded to the chassis an insulator is required between the regulator and the heat sink when mounting.
Transformers, sockets and regulator heat sink mounted.
Finished wiring. The large cap on the right side is a dual 50/50 mfd which is tied to each of the 47mfd caps in the glassware power supply. B+ & B- runs from the power supply to the center posts which is the central point for power. The smaller cap and resistor feed B+ to the 6sL7 tubes. Chassis grounding is accomplished at a single point near the input jacks and runs back to the common ground post in the center. This amp is dead quiet due to the DC on the heaters and proper grounding. The glassware power supply has a grounding connection that is fed through a system of diodes and capacitors which I ran directly to the AC line chassis ground located at the power cord input. The power supply is giving me 290vdc B+ with 280vdc on the plates of the 6Sl7's. 6V6 shared cathode resistor current is 74mA with an 18v drop . Calculating with that works out to 9.9 watts per channel with a plate dissipation of 71% each channel.
There is something to say for the sound of the 6V6 tubes. Those Edcor output transformers have the screen taps at 40% which is most common. From what I have been reading the 6V6 is unique in that due to it's construction it performs best in ultra linear configuration with screen taps at 29%. It would be interesting to have Edcor wind a transformer in that way to see the results. Back in the 50's and early 60's Magnavox who was known for their audio equipment (way before Phillips acquired them) used the 6v6 in most of their high end console's. If you examine the schematics you will find that they used practically the same 6V6 circuits in all the high end console's. In the mono units there were 6V6 amps, 6V6 bi-amps and in the top of the line quad 6V6 bi-amps. The same applied to the high end stereo console's. I have one of the amps which is an AMP-182BA chassis. It has a compliment of six 6V6's and two 12AT7's and one 12AX7. There are four channels to this amp, Left/right push pull bass amps and left/right channel single ended treble amps with the crossover networks in the input stages instead of before the speakers. Then in the most expensive console they used two 6V6 quad mono block bi-amps which had four 6V6 tubes in each channel of the bass amp sections.
They could have just as easily used 6L6's which would have cut the output tube compliment in half to attain the same power output, but for their own reasoning they stuck with the 6V6's. Some of the amps I have owned in the past have had EL84's, 6L6's, 7868's etc. But the 6V6's just sound different. After hearing my 6V6 amp, all I can say is that for a low power amplifier the 6V6 just sounds beefier to me, more meat on the bone than some of the other low power amplifiers I have heard But hey, everyone's ears are different so make your own decision to what you like.
I don't have top of the line test equipment, but what I do have is sufficient to get me results for repair and testing.
Got a new toy today, a Velleman HPS50 Personal Handheld scope. Perfect for a hobbyist like me with enough detail and function to troubleshoot the electronics I mess around with. Below is a capture at 80hz. You can see a nice waveform. Oh boy, now I'll need a distortion analyzer !! My signal source is a BK 3001 audio Generator which is fine for my use. The Velleman HPS50 can be set to read Volts RMS and PP, and also Watts RMS and PP. I wonder how some of the low power solid state amplifiers of the same wattage would stand up to a low watt tube amp. I have a low watt solid state amp that will hold it's own against the 6v6. That is my Optimus STA20 AM/FM Stereo Receiver which is 5W RMS per channel and coupled to a pair of Optimus Pro7 AV's the Optimus sounds nice for a low watt stereo.