The genericism of Stabilizer is a limitation to the performance when using Pounder (input stage limited in slew rate, distortion noise to achieve flexibility, layout is tight and complex/high viscosity, component choice trade-offs, software/assembly/debugging complexity).
I'd like to consider a single board hardware variant of Stabilizer+Pounder with the following deliberations:
Single board (debugability, reduced complexity of mechanics, software etc)
Use CPU's ADC and DACs, no external ADCs, DACs
no PGIA, merged IF amp and ADC gain+shifting stages
no EEM/LVDS interface
design and layout revised for 70 dB RF isolation (input-output and between channels, i.e. at the level of the DDS)
potentially a variant with an FPGA as coprocessor: then it would be a consolidated version of FastServo+Pounder, i.e. Zynq again with external ADC
The genericism of Stabilizer is a limitation to the performance when using Pounder (input stage limited in slew rate, distortion noise to achieve flexibility, layout is tight and complex/high viscosity, component choice trade-offs, software/assembly/debugging complexity).
I'd like to consider a single board hardware variant of Stabilizer+Pounder with the following deliberations: