and the power supply schematic here

click here for Dave Ewing's
50 watt Guitar Amp as of 8/8/2004

In the picture above you can see the big pile  of Amplifier parts that formed under my desk

Here is the paper design, kind of dirty after its trip to the fabricator

This is the punched aluminum before painting

Here is the cabinet carcass after staining

Very Very Very expensive tubes

fitting the front panel, these pieces of wood were mock ups, no dove tail joints

Time for parts inventory

This picture was taken right before starting to wire everything together

A front view of the completed amp

A rear view of the completed amp

And finally a side view of the amp

I first started looking at using 811's as the output tubes, but finally decided to use 300B's because of the reviews posted just about everywhere raving about their audio quality.  Now all I had to do was find a circuit that did all of the things  I wanted.

I thought the most important feature was a well filtered power supply since hum is a known problem in a vacuum tube design, next, I wanted an adjustable bias on the 300B's  to extend their life, they're way too expensive to replace every couple of months.  I also wanted an Integrated design.  I didn't want to have a separate pre-amplifier.  Then  lastly, I wanted a full stereo design on one chassis, not a pair of mono-blocks.

It turned out to be a tall order, I researched a lot of 300B designs on the internet, and I couldn't find a full circuit diagram and chassis layout  for just the amp I wanted.  So I finally decided to use parts of different designs, and piece them together into a design of my own.  It took a little longer, but it was well worth the extra time and energy.

After I decided to design my own amp, I downloaded one of the more popular circuit design programs, and started tinkering.  I came up with a design using a 6SL7 pre-amplifier, and a 6SN7 and 6AG7 mu-follower to drive the 300B.  For the power supply I found a design that uses 5R4 rectifier tubes, and a big 1000uf 450 volt computer grade capacitor to filter the high voltage.

Once the circuit design was done, I had to think about what kind of chassis I wanted to build to house what was looking like a 40 pound giant sized amp.  Because I am not a machinist or cabinet maker, the only chassis design that I thought I could build was one which uses a 1/4 in thick piece of aluminum on a wooden box.  This didn't require a lot of metal bending, and the wood working was easy enough for even me to put together, just four dovetail joints, and some routered edges.

On the front panel, I put a couple of personal touches, a headphone jack, a guitar jack, and an a switch for 3 back panel line inputs.  This way I could feed the amp with my CD player, my television, or  my turntable pre-amp, or I could just plug in my Stratocaster, and jam.  What a deal.

I also found a couple of round Simpson meters for the front panel to monitor the bias full time, and a blue LED power indicator to give it at least a pseudo high tech look.

Once I got the designs all down on paper, it was finally time to order parts.  The more parts I looked at, the more expensive the project got.  I started by buying the most expensive parts first, thinking that I would get them out of the way, and then I could just order the rest with the money I had left.  It didn't exactly turn out that way.  I live close enough to Cary Audio that I can drive there in just a few minutes, so I bought my output tubes, and transformers from them.

The 300B tubes that I bought are the K&R 300BXLS tubes.  These are good for a little higher output than the normal 300B's, and I've read that they have a better high end response than some of the other 300B's as well.  To go with the K&R tubes, I bought matching output and power transformers also from Cary Audio.

I then ordered the passive components from a few different sources.  I ordered a lot of the high end components like the Roderstein resistors, and the Black Gate and Solen capacitors, and the high end Alps volume pot from Michael Percy Audio, the regular everyday components including the main 1000uf filter cap,  from Allied Electronics, then most of the other tubes, the tube sockets and chassis components from Antique Eletronic Supply, and finally, I ordered the foil in oil caps, and filter chokes from Angela Instruments.

After accumulating all of the parts from the various sources, I sat down, and fitted the parts on a couple of big pieces of graph paper to get an idea of how they were going to fit on the chassis, and on the front and rear panels of the cabinet.  Once I saw that I was going to have to punch over 100 holes in that 1/4 inch sheet of aluminum, I decided to get a company here in town called American Precision to fabricate the 5 aluminum plates that go into the chassis, and I would do the wood working myself.

It took about two months to get the aluminum parts designed, and punched.  It took several iterations on the graph paper to get the design as perfect as I could get it.  I didn't want to have to pay for the parts twice.  Then once I got the design submitted to the fabricator things really got going.

While the aluminum pieces were being punched, I bought several pieces of Oak planking, and started working on the wooden part of the cabinet.  I had to buy a Dove tail jig, and borrow my neighbor's router to cut everything out.  I wasn't sure how all the parts would fit until the pieces came back from the fabricator.  To my surprise, everything fit. In fact, every hole in the chassis was the right size, and in the right place with the exception of two of the four holes for the main power transformer.  Since these holes don't show, I just fixed them up myself, and started building.

I bought brass mounting hardware to put the whole thing together.  I thought the brass screws would look better with the wood.  I painted the aluminum parts black using a satin gloss automotive spray paint, and I finished the Oak pieces with a Walnut stain and a satin gloss polyurethane.

The next step was the most fun.  I began to assemble everything, and wire it all together.  I never built a breadboard version of the circuit, I designed it on the computer, and verified the design with others that were posted elsewhere on the net.  I thought that since my design wasn't too different from the others that it should just fire right up if I assembled properly, and took the proper steps to make sure there were no ground loops, or under rated power resistors.

I used teflon jacketed, silver plated hookup wire to hook everything up, and for the line level hookups, I found a bunch of teflon jacketed rg-174 coaxial cable.  This was necessary since the line level audio is routed all the way from the phono plugs in the rear of the cabinet to the rotary switch on the front panel.

The only other snags that I ran into during assembly were the pilot light on the front panel, and the fuse holder on the rear panel, both of which were too long. It is hard to think in 3D when designing on paper, and sometimes things like big filter chokes get in the way. I ended up getting a shorter  fuse holder that had the same hole diameter, and that fixed the problem on the back panel, but the pilot light was quite a different story.  I had ordered a blue LED that came in a nice panel mount assembly, and ran off the 110v AC.  I thought this would make it easy to wire into the circuit.  I ended up taking it apart, and using just the front portion of it with the lens, and mounting hardware, then buying another blue LED which I wired into the 6v filament circuit, this ended up being a pretty good fix.

Now for the smoke test.  Every thing was assembled, and I was ready to try it out.  True to form, I plugged it in, and nothing happened.  After just a few minutes, the power transformer started getting hot.  I checked and double checked all of my wiring, and couldn't find any problems.  Slowly, one at a time, I started unhooking all of the different transformer windings from the circuit.  Finally, The transformer was completely disconnected, and still got hot even though there was nothing connected to it.  I went to Cary Audio, and explained the situation about the transformer, and they exchanged it for a new one.  Once I got the new one wired back into the circuit, it was time for smoke test number 2.  I turned the amp on, waited for it to warm up, and voile, it works great.  I set the bias pots at about 95mA, and listened on my test speakers driving the amp with audio from my Dell laptop.

Since I opted to run about 420 volts on the 300BXLS tubes, I am getting about 8 watts out of each channel of the amp.  It sounds great, clean crisp mid range and tight bass.  Once I hooked it up to the real CD player, and the real speakers, it really shined.  I don't have the optimum speaker setup for the amp, I am running a pair of Klipsch satellite speakers, and a Klipsch KSW-12 sub woofer.  I guess my next project is a set of efficient speakers.  Even using the Klipsch setup though, it sounds great, and it fills the room with sound, who needs 100 watts.  The first watt is the most important.  Its just nice that I have about 7 more clean watts per channel more to go with that.

Even though it took about  six months, it still cost less than it would have if I had purchased one of the high end stereo integrated 300B amps on the market.  Of course the best thing about it is the feeling I get every time I turn it on, knowing that I designed it and built it myself.

Questions or comments can be mailed to my email address