Let’s Try Again…
This is my second attempt at building my first operational headphone amp. It actually turned out much better then my first attempt. If you have read “The First Amp” you already know that I had some initial problems. But making mistakes is alright so long as we learn from them. And I did. For this amp I used the lessons I had gleaned from my first attempt. I spread things out more on the chassis and I inserted some more carefully thought out shielding. I also paid much more attention to my grounding scheme. The results were much better.
For this try I still wanted the layout to be simple. I also wanted to add some reasonable spacing between the tubes so that interactions were simpler to avoid. I used the same approach I used for the first amp. A plate of aluminum for mounting the components on top of a walnut chassis. In fact, the amp looks remarkably similar to the first one only more spread out. Here is the layout for the top plate.
Again the amp is eight inches deep however it is twice as wide as the first amp; ten inches instead of five. This allows for a simple linear layout of the signal tubes and provides plenty of internal space for component wiring. Again, limiting unwanted interactions is my primary goal in this layout. The interior layout for this amp shows how much more open and accessible things are.
This time, there is a physical separation between the noisy AC power sections and the signal sections of the amp. The divider down the center is a sheet of aluminum with one penetration for the filament wiring and a second for the high voltage plate supply lines. The output capacitors are mounted on the signal side of the shield and the power supply inductor (increased to 2H for this design) is mounted on the power supply side. What is not shown on this diagram is the addition of a hum shield across the front inside of the chassis. With the top plate, the bottom of the amp, the divider, and the hum shield, the amp electronics are shielded on four of six sides with grounded aluminum plates. This will also help to minimize interaction between the power supply and the signal electronics.
If anyone were to ask what is the greatest difference between this amp and the first one, I would have to say the amount of metal around the electronics. As can be seen in the photo below, this design required a considerably larger amount of aluminum than the first one.
On the top left is the top plate drilled for the major components and on the top right is the chassis bottom cover with some vent holes to help keep the power section cool. Below left is the divider plate drilled to take the output capacitor clips and the power supply filter choke. Below right is the rear plate which takes the power cord receptacle, the fuse holder, and the two switches. The hum shield is not shown above but is almost identical in size to the divider plate. Over all,a lot more metal went in to this amp than the first one.
As can be seen in this photo taken while I was wiring up the amp. There is much more space to keep things separated.
In this photo, the primary side of the power transformer is completely wired, as is the high voltage secondary for the rectifier tube and the first filter cap. This was taken as I was wiring up the filaments for the three signal tubes. There is actually another major benefit to making the chassis larger. There is much more space to reach into it to work. And anything that makes amp building easier is going to make it more enjoyable as well.
After I analyzed the failure with my first amp attempt, I was convinced it had nothing to do with the schematic design and everything to do with the implementation of that schematic. Because of that, this amplifier was built using the same schematics as before, with one small exception. The power supply filter choke for this design was raised to 2H. I did this to more evenly balance ripple currents between the two filter capacitors. This yielded a two fold improvement in overall ripple performance for the power supply; a much larger improvement then would have been expected from just the increased inductance alone. For completeness, I’ll include the schematic diagrams here as well.
Again, there is nothing new about these schematics. The amplifier is Chu Moy’s schematic from the Headwize website and the power supply is the same as before except for the choke value being raised to 2H.
Let me begin by saying that this amp is an unqualified success. Fed by my iPod or my portable CD player it sounds incredible. Rich lows, clean pure highs, full bodied tone like I have never heard through a straight transistor signal path. I set up a test where I could bypass the amp while being driven by my iPod in less then a second. This allowed me to rapidly switch back and fourth to compare musical sound quality. This amp is a definite improvement over either the iPod or CD player alone. Even in a blind test with my son doing the switching, I could clearly tell the difference.
For my 18 year old son this was an eye opening experience. He had never heard the sound of a well constructed tube amplifier having lived his life entirely in the era of transistors. Watching his eyes as he switched from song to song was really something to see. With my first successful amp I made a real convert.
Now this amplifier sits on top of a short book shelf just a few feet from my leather recliner along with a pair of Sony V6 headphones. This way I can sit back, close my eyes, and rediscover music that I haven’t truly heard in years.