The 10m band is 1.7MHz wide and I knew I couldn't possible make a fixed PI network to cover it. A Q of 8 would give the 3dB bandwidth of 3.6MHz and probably it will be barely enough to cover the band. With the minimum capacitance of 30pF on the plate-side plus some adjustability, the minimum Q has to be about 29. That is the 3dB bandwidth of 1MHz and about 500kHz of usable bandwidth as a no-tune final.
At Q=29, the parameters for a PI network matching 3900:50 are CP=41pF, L=816uH, CL=343pF. The capacitances are similar to those of the 15m network. It is better to wind the inductor for this frequency with a thicker wire, but I only had 12 AWG. Since the average output is going to be less than 200W (the peak being 350-400W), I thought it is not too bad. The coil was wound on a 1" ceramic form salvaged from a broken drake.
Testing and tuning was much easier after doing it for the 15m network. The minimum matching frequency was too high, so I padded the CP with a 12pF 3kV ceramic cap. Again, a cold alignment using 3900 ohm termination between plate and ground was spot on when actually tried after assembly.
The frequency range without significant output drop is 28.300-28.900MHz.
This is what my 600L puts out.
| Band | Drive Power | Output Power |
| 160m | 4W | 350W |
| 80m | 4W | 350W |
| 40m | 4W | 320W |
| 20m | 4W | 180W |
| 20m | 7W | 300W |
| 15m | 4W | 210W |
| 15m | 6W | 300W |
| 10m | 4W | 250W |
| 10m | 6W | 300W |