Building an Output Transformer for a Tube Amp
By John Fisher
As anyone who knows about tube amps will tell you, "The output transformer is the most expensive part". Not is it is only the most expensive part, but is is also probably the hardest part to come by. Parts for tube amps in general are a lot harder to find then other more common electronic parts. Transformer theory for power transformers is pretty straight forward and it is fairly easy to find information for winding a power transformer of any type and size. I have always been intrigued by output transformers and it seemed very hard to get any specific information on how to make one. Unlike power transformers where you can make one that will work and put out the voltage with sufficient current needed, output transformers have other factors that depending on it's design, will effect the sound that it puts out. I am not going to pretend that I know all about them because I don't . (At least not as of this writing). I also will not fill this article with a lot of mathematical theory as I am just trying to figure out myself but will include more on this in Part 2 of this article. But I just wanted to share the specifics and experience on how I made my output transformer.
After much frustration I found someone on the Internet to help me with some specifics on building one. At least a good starting point. I want to thank P.G. Doynov who helped me greatly with his expertise. He gave me a good starting point on building my output transformer and answered many questions that I had on the subject. Here is his web site :
This man knows his stuff and has a good supply of rare tubes and transformers and other valuable tube related components, for anyone who is interested.
With just under $5 worth of materials and an evening of work I was able to successfully build my output transformer. I wanted one for putting into my Fender Deluxe guitar amp (5D3) clone that uses 2 6V6 outputs tubes in push pull. This amplifier is about 15 watts.
Above is the schematic of the output transformer. The primary wire could be from 0.16 to 0.18 mm in diameter, enameled copper wire and the secondary can be 0.69 to 1 mm diameter. Mr. Doynov told me that thicker wire will result in better bass response but it will depend on if you can fit it on the core OK or not. I used the dimensions in the above drawing and it had good bass response for a guitar amp in my opinion. I used a core that was 1" square in "E" and "I" laminations.
The Winding Plan
Following is how I wound my transformer. This is something that I am just learning, so I know there are many improvements that can be made and I am sure that I will eventually write up a sequel to this article as I learn more about it and gain more experience. I first want to point out that in winding the transformer it is very important that things are isolated well as if things are done in an unsafe matter not only can it ruin your amplifier, it can also kill you!
I started with a plastic bobbin that was made for the 1" square core for this transformer. Although interleaving is recommended in output transformers, they are used less extensively in guitar amplifiers as compared to "hi fi" tube amplifiers. Interleaving is where you wind part of the primary windings and then some of the secondary windings over it and then some more of the primary winding on that and so forth to where the primary and secondary windings are interspersed in sections within each other instead of just winding the primary windings all together and the secondary windings on top of that. Interleaving the windings is to help give the transformer a better high frequency response. In my transformer I did a minimum of interleaving. I first wound 66 turns of the secondary windings. I wound it neatly side by side and tight and then before I started the the next row I painted the windings with hot wax and then wrapped some very thin insulation paper around the layer before continuing the next layer. I got the thin insulation paper or capacitor paper at a transformer supply place. After wrapping a layer of paper tightly around the coil and gluing it together (kind of like rolling a cigarette) I started the next row of the secondary in which I was able to finish the 66 turns. I then ran the ends of the wires outside the bobbin giving them about a 6 inch length. (This will also be where I make my 4 ohm speaker tap. I was now ready to start the primary windings. Before starting the primary windings I put a layer of thicker transformer paper around the first section of the secondary windings. It is extremely important that this is done right and that the primary is insulated well from the secondary windings. I then started the primary windings which use wire that is a lot thinner. Ideally it is good to wind the thinner wire in a neat row but because of the crudity of my setup I didn't get it perfectly wound side by side so I basically did the best I could and then after about 200 turns of the primary I then painted it with the hot wax and covered that layer with the thin paper before starting the next row. After 7 or 8 rows of the primary I got up to 1500 windings which is where the center tap of the primary coil is. I then ran a loop of wire out for the future center tap and then continued on with the next half of the primary windings. After the primary windings were done, I then wound the remainder of the secondary windings over that making another tap where the 8 ohm speaker connection will be. I then connected this in series with the first section of the secondary windings. After I was done all the windings I covered it again with the thick transformer paper. I then got each wire and soldered a different colored stranded wire to it as leads that will run out of the transformer. After taping them separately to the transformer to secure them individually and away from each other, I then put one more layer of paper around the whole thing to further secure the leads and to protect the transformer coils.
Above is an illustration of the top view of the coil that hopefully will make things clearer. Part of the secondary is wound first and then the primary is wound on top of that and then the rest of the secondary is then wound over the primary. The 2 secondary windings are then connected in series. This is a very basic interleave. (Important) Always wind each section in the same direction. If not, some coils will cancel out other coils reducing voltage.
My Simple Winding Setup
I made a simple but effective winding setup using a hand drill and a digital counter that I built. I like using the hand drill because I feel I have a fair amount of control as I can alter the speed of the turning at my own wit plus with my other hand control the tension. Although a more complicated machine could probably do it faster and probably more consistently, the hand drill works and I can wind things quickly. I got this idea from Mar Hammer for also winding guitar pickups. The thing that took time doing the transformer was not the winding but taking time to paint each layer with hot wax and cover each layer with the paper. The actual winding of a few thousand turns is actually quite easy and fast. I bought a hand drill on the street for under $10 that worked find. I did realize later the it is important to make sure the hand drill you get works smooth and that there is no wobble in the axle.
I simply screwed the hand drill to a board where I also mounted the digital counter I built.
Above shows my very crude assembly. The counter has 2 wires that go to a reed switch that is placed close to the chuck of the drill. The drill chuck has a little magnet glued on to it so that everyt ime the drill makes one turn the little magnet will pull on the reed switch making a contact and triggering the digital counter.
Here is the transformer bobbin in the chuck. I just glued a reed switch on the little piece of wood with a glue gun to hold it in place near the chuck. The bobbin is held with a little square block of wood pressed into the center and a thick nail with the head cut off in the center of the block of wood which is inserted in the drill chuck. The trick is to get the nail well centered and straight so that it will turn evenly while winding. The lead wires coming out of the bobbin are temporarily taped up out of the way while winding. It is important to test out the counter well before starting, to make sure that it is counting correctly and that there is no false triggering. I am sure that this can all be done better and that this method can be improved but it worked good for me and I am inspired about the possibilities.
Here is the digital counter section and circuit board that I just glued on the wooden board. I also have a reset button for resetting the counter. I didn't take time to make it fancy in any way. Maybe later!
This is the same circuit that I showed in my pickup winding project that I also used for this project. This is a digital counter that can be duplicated for as many digits as desired. It is cheap to make and easy to find the parts. I made 3 digits on mine which was practical and enough for me. I don't have a PCB layout for this but I just drew one up and surface mounted the parts. You can probably find some ready made perforated boards or other similar ready made boards that will probably work fine for making this. This is the basic circuit excluding the reset switch and the bouncless switch.
Here is the bouncefree switch that I used. This is essential in preventing any false triggering in the counter as there has to be a small amount of time between each pulse before it can be triggered. I used a .01uf capacitor for C1 to get about a max of 10 turns per second triggering properly. I don't think that I would wind anything faster then that. I replaced the push button with the reed switch.
Here is how to make the counter reset to zero which is not indicated on the first circuit. All the pin 2 and 3's of each 7490 IC should connect together.
Here is a suggested power supply for the Digital counter. I would use a transformer that is 500 mA or larger. I would also advise putting a small heat sink on the 7805 voltage regulator.
This illustrates how the magnet will close the reed switch every time the drill makes one turn. I glued on the magnet to the drill chuck.
Above shows the E and I laminations that I used. For a "Hi Fi" output transformers it is common that the laminations are assembled in such a way as to create an air gap in the core. This creates less distortion but causes a slight decrease in the output power of the transformer. I will explain more about this later but in this transformer that is used for a guitar amp, I laminated it without an air gap but by staggering the laminations as in most power transformers.
How Does It Sound?
I was very pleased with the sound when I tried it in my amp. I still have to try it compared to other transformers but I have no complaints in the way it sounded and worked. I plan to also install bell caps and supports when permanently installing it in my amp. I Played for a while loud and soft and visually I can see that the tubes were working properly and there were no apparent problems. The test will be to take it on a live performance for a longer period of time and see how it does. I consider this project a success and I am inspired about making and experimenting more. I would like to try and make another one using more interleaving to see how this might effect the sound.
Stay tuned for part II