I am also very interested in the 10 or 12 watt kit. Including 8 and 16 ohm output taps would be a big plus for me.
The transformers can be wound with taps. My best 10 watt is EM7 triodes push-pull with little or no feedback.
I also have a design that combines this 10 watt amp for the woofer with a Single Ended EM7 for the tweeter. Each amp is mono and includes the crossover at the input. Therefore one need run one signal cable to the amp beside the speaker and two short runs to the drivers. In any multi-way speaker the weakest link is the crossover that follows the amplifier as it has to handle low impedance drivers. Although some omit the series woofer choke due to its expense it is a big mistake to let the woofer roll off on its own as the response is very ragged in this region. It is far better to separate the frequencies
before the power amps with simple resistor-capacitor filters. Then each amp and its output transformers can be optimized for its frequency range. A volume control is included to match the generally more sensitive tweeter with the woofer. This volume control is a handy way to change the tonal balance of the speaker at will. One of my customers who has this says he can't imagine living without it. He simply trims the highs based on what he hears on a particular recording to suit that recording. If you have never had this option, well you don't know what you are missing. Haven't you often wanted a little more or a little less treble from some recording?
This also fixes a common problem when driving a two way speaker with SE amps. Many two way speakers have impedance peaks or dips around the crossover point or some other place where the designer inserted a network to trim the frequency response to his desires. Since most SE amps have rather high output impedance this causes the speaker's response to be amplifier dependent. To hear what that speaker designer heard you would have to use an amp with his output impedance (Load Z/damping factor). Now, if we have a simple two way system (woofer and tweeter) we have two drivers whose impedance is generally more constant when seen directly than through a crossover. In this case the damping and difference in damping is far less critical. This is the correct way to do things and it's really quite simple. Do you know that most single drivers are constant impedance over their usable range due to the simple fact that they are mass loaded? This is what Rice and Kellogg realized in 1924 when they invented the moving coil loudspeaker for General Electric.
This is available for $1500 per mono two-way amp using 3-13EM7 tubes and 1-6BQ7 with solid state rectifier. $1900 with a 5Y3/5AR4 family rectifier and choke input filter. Anyone making a two way speaker should consider this option over expensive exotic crossover caps and inductors. I think it demonstrates a lack of imagination to do things the conventional way (passive crossover at the speaker) when for about the same money it can be done a much better way.