5-10 Output transformer
From the start I wanted to create an output transformer for the Mullard 5-10, not as a cheap option, but a quality product that would allow a 5-10 builder, or any design using EL84’s in a push pull with UL taps, to shine.
The result has been a true adventure in design and development.
I wanted the build for the output transformer to use layer winds. This is where we insert thin layers of insulation between winding layers. In the past this would have been used for inter-layer insulation, but in our case, using high grade enamelled wire this is not required. What layer winding does do is ensure that the winding is neat and tidy and uniform.
Many cheaper designs just wind one section at once, causing inter layer mis-alignment, and a loss in overall build quality and a drop in frequency response.
Above is an image of a layer getting towards the end, showing the inter-layer insulation.
The design incorporates a multi section design.
The Mullard 5-10 output transformer we ended up with has been put through its paces in a real 5-10 build.
The results are quite remarkable, for only 10W. It easily fills quite a large living room. The bass is fantastic, and the upper frequency scale is very clear and well defined.
The mains transformer for the 5-10 is able to supply the requirements for a mono build, at 100mA DC current. At full overload the HT current rises to 125mA. however the design of the transformer is such that this will prove no problem, it getting a little warmer if in this situation for a while.
I decided that as we had gone to so much trouble with this particular design, we should incorporate good solid electrostatic screen between the primary and the secondary:
The picture shows the copper screen shortly after fitting to the bobbin.
The screen also provides a level of safety, and the screen wire must be connected to a good earth, and the supply fitted with the correctly rated fuse.
The primary is actually wound in it’s own “pocket”, as is the 600V centre tapped secondary. This gives excellent insulation between the two.
I intend to do some experiments on the electrostatic screen, and I will post the results at a later date.