By looking at the schematic I would imagine using an 10uH RF choke could be a drop in replacement.
I actually had the chance to build that circuit years ago on a breadboard and had terrible luck with the resulting modulation, and the output power is nowhere near 10 watts as many sites suggest. Given that it biases the IRF510 as linear it only gets around 6 volts of RF out (using a 12v power source) which is closer to about 1.5 watts carrier.
Thanks Kage for your aid ! This is the Corsair 2, it is new. I known that the old circuit was problematic, but this would be ok.
I think that i will use a donut, because the transistor should go out with 45W (or 35) and i think that the commercial 10uH could burn or heat up (?)
Before another question, why a rf choke is used ? For filter the any interference ?
When i will finish i will be happy to send you the PCB, i have 2-3 copper board to use, and this is my first PCB, so you can have "one of the first PCB by Corvette"
The current across Q1 and Q2 and L3/L9 would be low enough that most small RF chokes would be suitable, especially with so few windings (<10uH). Even L1/L2 could be made small, like those chokes that are the size of resistors. Low value RF chokes/ferrite beads like L3 and L9 are usually used across VCC lines going to each major part of a circuit like that to suppress potential RF feedback flowing along the positive voltage line from the final RF transistor and to suppress parasitic oscillation between transmitter stages. Another common place small ferrite beads are found is on one of the legs of a transistor to prevent parasitics from breaking out at VHF/UHF which can cause catastrophic transistor failure or very bad interference to other parts of the radio spectrum. The majority of current will be going to Q3 since that is the RF output amplifier stage and L5 will be passing all of that so that would be the weak point if not using a decent gage wire for that coil at higher output power. Given the design only puts out <5 watts of power it is not really of concern either.
For L1/L2 i will use commercial inductance in plastic package.
As you suggest i will use also < 10uH commercial inductance for L3/L9 So the only homemade inductance that i have to make are L5-L6-L7
According to you what is the ideal thickness for the wire ?
I will use this post for speak about Corsair 2, it's ok for you ?
No idea what gage. Is there some documentation about this design around? You could ask Dave Martin since he designed the circuit but I don't know how you would contact him. Probably search out his email or ask on the pirate radio homebrew group on yahoo. When I tested this design on breadboard a while back I wound my coils by finding the resonating point where they worked best on toroidal cores I had in the spare part bin. I am not sure if changing wire diameter is going to have too large of an effect using the original design parameters as suggested in the schematic. Just give yourself some slack wire and test/trim until you get maximum power output would be where I would start and make sure to check the output on an oscilloscope if you have one handy.
I have the mail contact of Dave, but I do not want to bothers him, or better, maybe i will do but if i will have some big problem.
About the wire i found this image:
This is the GRANADE TX i am sure that you know yet. The power is the same of the Corsair 2 and its use a tiny wire in toroids (2 mm ?) the same that you can find in the coaxial cable of your TV.
EDIT: i have found this in dave's blog: All the above coils are wound using 22 - 24 SWG, enamel covered copper wire.
I have some spare toroidal cores (all grey) find in some power bench, i can use it, but i don't understood how have you make your measure.