Fisher Cabinet Amplifier:
Background:
I need an amplifier that will go inside the entertainment center in my living room. It must have a high WAF and fit in with her existing room decor, which is mostly dark wood and antique furniture. I have built a pair of speakers from a matching pair of 1941 vintage Zenith console radios with modern 15 inch Silver Iris drivers installed in them. They have a 96 db efficiency rating and would be happy with a few watts of power, but are capable of handling much more. They can be seen on My Systems page. This amplifier must fit on a shelf in the cabinet between the speakers (behind the closed doors). It is sharing the space with the DVD player and some other equipment. It must be reliable enough to run for extended periods since it will be used for TV audio as well as music.
The last time that Sherri was out of town, I dragged several amplifiers into the living room and performed listening tests with these speakers. I tried the Tubelab SE with 45's and 300B's, a Simple SE with several different tubes, the 300Beast, the 845SE and a Scott EL34 P-P amp. I really liked the sound of these speakers when fed with the 300Beast, but it did not fit the in the available space or satisfy the WAF requirements, neither did the 845SE. The Tubelab SE with 45's was the clear winner on a lot of music, but didn't have enough power to rock. An amplifier built with 300B's would be too tall to fit in the available space. My next choice was the Simple SE with either 6L6's (cranked to a dim glow) or 6550's. The 6L6's would probably melt if pushed hard for extended periods of time, so it was decided, I would build a Simple SE with enough power supply capacity to crank out about 15 WPC using 6550's, but conservatively bias the amp for longevity. It would need to fit into the entertainment center with the TV, VCR and DVD player. That puts limits on the outside dimensions. I took a trip to my warehouse full of old junk in search of an existing cabinet that might work. This would save some time.
Planning:
I have an old Fisher receiver that came with a wood cabinet. I decided to build an amplifier that fit will inside that wood cabinet. I did this because I wanted to build an amplifier for use in my living room (and I am impatient), and I am not good with wood. I plan to build a permanent outer cabinet for this amplifier at a later date and return the cabinet to the Fisher receiver (which works, but I don't use). The use of an existing outer cabinet makes the job easier, and defines a narrow range of dimensions for the amplifier chassis.
The first step in building an amplifier this way is to determine the maximum dimensions for your amplifier chassis, width, depth, and height. Measure the height of all of your tall components, and make sure that they will fit. The tall items will be the transformers, the output tubes (including the socket and board height, and the supplemental power supply capacitor (if used). Plug one of the output tubes into the socket on the PC board and measure the total height. You must add the thickness of the chassis material and the height of the spacers used to mount the PC board and the other components.
You need to choose a material for the chassis itself. It does not have to be metal. The material needs to be rigid enough so that it can support all of the heavy components without damage. The chassis can be a flat piece of material. It does not need to have the usual box shape. A flat piece of aluminum in the usual .050 inch thickness would be too flexible for use in this amplifier, since I am planning to use 30 pounds of transformers. I could use a piece of .100 inch aluminum (I have used it in the past), but I didn't have any large enough for this amp. I looked around my work shop and spotted a piece of 1/2 inch particle board that was covered on both sides with white Formica. It was a leftover shelf from my equipment cabinet. I could have used aluminum, stainless steel, brass, or even thick plastic (Plexiglas or Lexan) for this chassis. I am using the wood to demonstrate a simple, easy to build amplifier. The construction technique would be the same with any of these materials.
My tallest component is the Hammond 1628SEA output transformers. A quick measurement shows that the transformers can be mounted 1/4 inch above the board (spacers are needed to avoid pinching the wires against the transformer body) with some room to spare. I have enough room for the tallest tube that I may want to use (the 6550's).
I measured the inside dimensions of my cabinet, and decided to use a 17 inch by 14 inch chassis.
I cut the shelf to fit inside the cabinet, and set it inside for a trial fit. I then measured the room left over for transformers, and they will indeed fit. I planed to use a piece of 2 inch angle aluminum to mount a 5 3/4 inch by 17 inch aluminum front panel to the wood chassis. I planned to use a piece of the angle aluminum alone for the rear panel.
Another important step is to get accurate dimensions for your components. Many of the component dimensions can be obtained from the web. Even if you find dimensions for the components on the manufacturers own web site, check them. I have found some of them in error. The dimensions for the Hammond transformers are WRONG, off by almost an inch!
The next step is to plan the layout of the components on the chassis. There are several ways to do this. Many people will use some type of CAD software. It is also possible to use a drawing program like Visio, or even PowerPoint.
Here is a simple drawing that I did for a different amplifier using Microsoft Visio. This amplifier will be featured on these pages as it is completed. I may settle on a different layout before I build this design. I did four different layouts to determine what size metal to order for the chassis. In keeping with the simple, low tech approach to this amplifier, I will demonstrate another way. This is how I planned chassis before the home computer was invented (I have been doing this that long).
First, I draw the outline of my 17 inch by 14 inch chassis on a piece of grid paper.
Then, I make paper cutouts of each of the large components using the same scale as the chassis drawing above. The supplementary power supply capacitor was too tall to stand vertically, so I chose to lay it down horizontally. This takes up more chassis space, but I have plenty of room in this amp.
Next, I try placing the paper cutouts onto the chassis. Try different arrangements until you get a layout that looks like it will work. You need a layout where everything will fit, with some space between the components. All of the components should be placed such that the transformer wires will reach the proper terminals on the PC board. The core (laminations) of the power transformer should be at a right angle to those of the output transformers. This is how I would layout this amplifier if it were open, or used with good ventilation. When you have a possible layout, you should make a copy of it.
Usually there are several possible layouts for an given amplifier design. This is another possible layout for the same amplifier. I then try to choose the best layout from the possible choices. I had 4 possible layouts for this amplifier. Try to figure out which would work the best and why. Usually each layout has strengths and weaknesses.
The top amplifier layout above (green tint) has the shortest path for the input wiring (there will be a volume control on the front panel). However the bottom design places the output tubes near the rear of the open backed cabinet for best ventilation. Since I tend to operate my amplifiers at (or above) the maximum tube dissipation ratings, and the entertainment center restricts air movement, I decided to use the bottom design.
Next, I place all of the components in their respective locations on the chassis, to verify fitment. I then tested to make sure that this chassis would fit into the cabinet.
Some would ask, why couldn't I just start with this step. In some cases (and in this one) an experienced builder could do this. In this case, I have an oversized chassis. The size of the chassis was dictated by the existing cabinet. I had all of the parts on hand before I started building. If any of these were not the case, the advanced planning could easily save the cost and time of starting over because something didn't fit right. The blue Visio drawing above was done to determine the smallest size metal top plate needed to build a similar amplifier using under chassis mounting. Then I ordered $120 USD worth of aluminum (for 3 amplifiers). I have 6 similar drawings, 2 for each amplifier. I will choose which ones to build later.
Mounting the Components:
The next step is to mark up the actual chassis. If you are working with metal or plastic, cover the surface with masking tape, and mark the tape. Mark the position of all of the holes. I used the actual transformers to mark their holes, and a blank unpopulated PC board, to mark the holes for the PC board.
Next, I drilled all of the holes, and inserted screws through the chassis from the bottom. The PC board and the transformers must be spaced above the chassis. These transformers must be spaced above the chassis to avoid pinching the wires between the end bells and the chassis. See the photo below for details. This is not needed on some transformers (ones that look like the choke used here). A nut is threaded onto the screws that are used to mount the PC board and the transformers. I then added a plastic spacer to these screws. Extra nuts can be used if spacers are not available. Test fit the components. Check the transformers for clearance around the wires. Check the PC board to make sure that none of the solder joints or components touch the chassis or even come close. There should be plenty of clearance between the board and the chassis. I use about 1/4 inch (6 mm). This is obviously important when using a metal chassis, but it is also important with wood or plastic. The PC board gets hot, air space helps cooling. Tighten the nuts if everything fits OK.
Everything fits OK, so I put the nuts on each screw to hold the components, and tightened them.
This is a side view of the output transformer and the PC board illustrating the mounting technique. The power transformer is mounted the same way.
Another view.
Wiring:
After all of the major components are mounted, I connected the transformer, choke, and capacitor wiring to the board. This is easily done with only a screwdriver (and maybe some wire strippers). Note that there are two wires in each of the secondary connectors for the OPT's. This is common. Other connectors may have two or more wires depending upon the choice of optional switches, choke and capacitor. Consult the wiring diagram in the PC board manual corresponding to the board that you are using.
Another view of the wiring. I installed some tubes for the photo.
A close up view of the transformer wiring.
The Rear Panel:
I used a piece of angle aluminum for the rear panel. This amplifier will be located in my living room entertainment center forced to live with the VCR and the DVD player. The rear panel will not be visible, so I didn't spend much time making it look nice. The rear panel is used to mount the power connector, the speaker connectors, and the input connectors. I used a power connector that has the fuse holder built in. This just means one less hole to cut or drill. I measured, and cut the rectangular opening for the power connector using a Dremel tool.
I inserted the power connector to verify fitment. I then removed it to avoid getting metallic debris in it during the drilling of the remaining holes.
Next I drilled the holes for the speaker connectors, and the input connectors. I drilled 3 holes in the rear panel for mounting it. I then installed the rear panel on the chassis.
The output transformers are mounted close enough to the rear of the amplifier that the rear panel must go under the rear transformer. Better planning could have avoided this. The two holes on the left (as viewed from the back) side of the panel are carefully measured and drilled so that the same screws that mount the output transformer hold the rear panel. I had to loosen the forward two transformer mounting screws, and remove the two rearward ones to do this.
A Small Detour:
During the course of building this amplifier, I had to go out of town (a lot). While traveling, I stopped by to see a friend who sells vacuum tubes. He told me that he had just received a selection of "previously auditioned" Russian vacuum tubes. I purchased several pairs of power pentodes, and had to test them as soon as I returned, for two reasons. He would replace (or refund) any defective tubes, and more were available. Since I have been evaluating tubes for use in the Simple SE, I had to try these out in a Simple SE. My workbench can only hold one amplifier at a time, and this was it, so it was time to fire this one up. I checked over the wiring, and made temporary (but safely covered) power, speaker, and input connections. I added ground wires for the rear panel and the transformer cases. I DON'T NORMALLY RECOMMEND FIRING UP AN UNFINISHED AMP UNTIL ALL SAFETY CHECKS HAVE BEEN COMPLETED.
I then connected voltmeters to the power supply and output tube cathodes as instructed on the checkout page in the Simple SE manual. While standing a safe distance back from the amp, I used a remote power strip to power it up. The voltmeters revealed normal operation. I put in a CD and hit play. As expected it worked. I went through about 2 dozen different output tubes. This picture shows Electro Harmonix KT88's. I was impressed with their ability to handle power. I stopped the test twice and changed the cathode resistor to increase the tube current. I have decided that I will use these tubes in my own amp. I also called to purchase the remaining tubes. ( I now have two amps using these tubes)
Here is a picture taken in a darkened room of the KT88's at full tilt. This pair is operating above the recommended maximum ratings for a KT88. I would not recommend doing this, but I got these tubes for $10 USD each, so why not! The amp had been playing along happily for about two hours when IT happened. ONE CHANNEL JUST DIED! Did I get carried away with the tube current and blow a tube? NO, the tube current was still the same. I leave the voltmeters connected up for the first few hours of use on a new amp, and when evaluating different tubes. This makes it easy to see something that is not right. I got out my scope, and verified that signal was present on the grid of both output tubes, but the signal on the plate of the left channel was low and distorted. This would indicate a short in the OPT or on the secondary of the OPT. Some careful examination revealed the problem.
This was here all along. I went back and looked at the photos shot during the assembly and it was visible in the original images (before I reduced the resolution for the web). It just waited until I thought all was well to actually short out! In this case no harm was done. I fixed the short and the amp was happy. If this short was on the secondary of the power transformer, the fuse would have instantly blown. If someone then put in a bigger fuse (we know better, right) the power transformer would have been damaged.
I should have caught this when I visually inspected the amplifier before powering it up, but I was in a hurry to test some tubes. This shows that even a veteran amp builder can make a mistake, so DON'T SKIP THE INSPECTION AND CHECKOUT STEPS!
Another unfortunate side effect of firing up a new amplifier also happened to this one. It has been said (and confirmed by me) that if you start using a new amp before it is done, it will never get completed. Due to my work schedule (10+ hr days) and family commitments (traveling 6 out of the last 8 weekends), two months elapsed before I had time to work on this amp. During this time it went on another small detour. The amp (in its unfinished condition) was invited to appear at a listening session where it would be surrounded by amps costing 10 times as much. See what happened here. Now finally back to the build.
The Front Panel:
My original plan was to make the typical front panel out of sheet aluminum. This is how I usually make front panels, but I changed my mind twice during the two months of inactivity. During this time I learned about several types of thin adhesive backed aluminum that can be printed on directly or "exposed and developed" photographically. I first planned to use a piece of this over a sheet of Lexan for the front panel with a window to show off the cool blue glow of the EH KT88's. After thinking about this for a while, I realized that I would need to move the large cap that blocked the view of the tubes. This would snowball into a lot of components being moved, so this idea is put on hold for another amp build. Then I changed my mind again and took the easy way out. I decided to use a piece of PC board with the aluminum sheet on top of it. The PC board is far easier to work with than aluminum, and it will be covered anyway so I decided to use PC board material. The front panel is supported on all sides by the cabinet, so it is sturdy enough for this application.
I cut a piece of PC board material to the proper size.
Next I placed the knob and switches on the panel to determine proper placement. There are two brackets that will be used to mount the panel.
The center of each item that will be installed is marked on the PC board. Then I drilled all of the holes. I chose switches that are mounted in a single 3/4 inch hole. This is conveniently the same size hole as a 9 pin tube socket, so a chassis punch could be used. I wasn't smart enough to think of this until after I drilled the 3/4 inch holes.
I decided to use glue to attach the brackets to the front panel to avoid having screws visible on the front of the amp. J-B weld will stick metal things together permanently. I used it to glue the cast aluminum oil pan on my car back together. Tape will hold the brackets in place until the epoxy cures (overnight).
I test fitted all components to be sure I did this right. They can not be permanently installed until the overlay is on.