The low-drive version with the 8417 driver tube did not have the band selector. It was a straight-up 11 meter leen-yar.
We use a fairly-simple pi-network to match the input of the two tubes down to 50 ohms.
This coil goes from the relay's input lug to the blocking cap that feeds the cathodes.
Takes a capacitor to ground on each end of the coil. Bear in mind this old pic shows a shorter coil than we have found to be best. A couple more turns works better. The input lug on the relay, where the coil is attached gets a 68pf 2000 Volt cap from there to ground.
A 27pf 2000 Volt disc cap from the other end of the coil to ground is the starting point. Squeezing the coil's turns closer, or spreading them apart will allow you to minimize the reflected power reading on the amplifier's input when it's keyed.
I also recommend squeezing a fan in beneath the tube sockets. Doesn't have to be a hurricane. A very-gentle breeze is sufficient. A fan meant for 120 Volts AC can be slowed down a lot by putting a power resistor in series with the coil. Slowing the fan's speed so you can't hear it with the cabinet in place won't reduce its effectiveness keeping the socket pins cool. A 12-Volt DC fan works fine, and a smaller one will do the job. A low-drive JB that has had the driver tube pulled out has a 7-Volt AC source that was lighting up the driver tube's heater. A half-wave rectifier and 1000uf filter cap would power a 12-Volt DC fan at a lower voltage and slow it down a bit.
The original Eimac data for the tube mentions the need for a gentle breeze on the underside of the tube socket. Doesn't take a lot. The JB2000 doesn't have any such airflow built in, and it's famous for melting the solder from the two filament pins. This exposes the socket's spring contacts to excessive temperatures and will oxidize the contact surface. Also ruins the contact's spring temper. Of course a loose contact spring makes a poor connection that generates even more heat. Bad tube sockets in a JB are like a downhill snowball. A tube just won't light the filament any more, so the operator puts in another tube. Same thing happens, but faster now. Figures out the socket has gone bad. Replaces the socket. So far so good, but now he puts the damaged tube that lights up only when you nudge it into the new socket. Burns up the contacts. The tube eventually won't light up at all, and a new tube goes in the new socket. The new, damaged socket. Lather rinse repeat. I call it tube socket ping-pong. A good tube with undamaged pins and a good socket with undamaged, clean, shiny spring contacts are the only cure.
73