Sorry in advance for the long posting! Perhaps it would be wise in any replies to delete my message.
A few days ago, Nels W0TUP raised an interesting question with me regarding using the v9.0 receiver with the switched BPF module built using the 6m filter option and Fred PE0FKO's v15.10 firmware.
What on the face of it should have been a really simple answer, in fact raised a number of interesting points and indeed, perhaps the need to re-arrange the order of the filters on the module, but only when building it with the 6m option.
Just to recap, when built for HF only operation the module has 4 filters referred to as Bands 1 to 4.
Note in the set-up for v15.10 firmware these are called Bands 0 to 3 (or filters 0 - 3 as this reflects the binary value of the two bits used to select each one. For the rest of this message I will refer to them as per the BPF circuit diagram, Bands 1 - 4.
In HF use:
Band 1 = 1.8 to 4 MHz
Band 2 = 4 to 8 MHz
Band 3 = 8 to 16 MHz
Band 4 = 16 - 30 MHz
The best option to use in the v15.10 firmware is the 'ABPF' Auto BPF mode, as this allows full compatibility with all existing SDR programs including Rocky as the BPF switching is controlled by AVR chip dependent on the LO frequency selected. All you need to do is set three filter cross over points, currently by default set at 4, 8, 16 MHz. These cross over points need to be in ascending frequency order.
When built as suggested with the 6m option filter:
Band 1 = 6m
Band 2 = 3.5 to 8 MHz
Band 3 = 8 to 16 MHz
Band 4 = 16 to 30 MHz
I guess that order was suggested to allow the Band 3 and 4 filters to remain the same in each version. It can be seen that it is no longer possible to set the three ABPF cross over points in ascending frequency order, therefore causing a problem.
There appears to be a number of possible solutions. I have discussed these with Fred PE0FKO, and whilst it could be possible to do this within the firmware, this is not really a desirable option, as the number of people using the 6m option is likely to be small compared to the general use, and the desire to keep any new releases of the firmware to an absolute minimum.
One simple solution is to set the first filter cross over point to greater than 50MHz, therefore forcing the first filter only to be used. This of course limits using the v9.0 RX on 6m only, which is not really desirable.
A second, more logical solution is to build the BPF module with Band 4 as the 6m option, and the others shifted down by one position. That then would allow you to set the ABPF three cross over points as 8, 16, 30 MHz, and would allow correct operation on all bands.
One other point to remember when using the 6m option is that depending on the maximum frequency your Si570 will run at, it might not be possible to simply use the x4 multiplier on the LO frequency. Some of the CMOS devices no longer go up to 200MHz, therefore you would need to use sub-harmonic sampling, 4/3 (1.3333) multiplication factor. As you cannot set this on a band by band basis, this applies to all bands.
The other option available to use the BPF module and 6m option has severe limitations.
You can use the 'BPF' mode within the firmware and define the start and stop frequencies for each band, however this will limit your use of the v9.0 to Winrad and possibly some versions of PSDR. You would not be able to use Rocky with this option.
I believe the best solution is to build the BPF module with the filters arranged as:
Band 1 = 3.5 to 8 MHz
Band 2 = 8 to 16 MHz
Band 3 = 16 to 30 MHz
Band 4 = 6m
As the switching arrangement is common for all bands, I cannot see this change having any adverse effects.
The only 'work' needed would be to create an updated circuit diagram and revise the builders note to reflect these changes.
Sorry for the long posting, but some things perhaps to be thinking about if you intend to build yours with the 6m filter option.
73, Bob G8VOI