Deviation adjustment

[VK2BEA]

There seems to be only a selection of wide / narrow deviation.

In wide, I can only achieve a 4 kHz peak deviation (which is low). Is setting the deviation a register setting?
Could this be made available to the user?

Michael

[armel]

Hi Michael,

Thanks for your message.

Are you talking about TX deviation ?

Currently, the firmware only exposes a choice between wide, narrow, and narrower bandwidth modes (except in the Basic edition). These options correspond to preset values written to register 0x43 of the BK4819, which primarily controls receive filtering — not the transmit deviation directly. That’s all for now.

But feel free to explore what other alternative firmwares might offer if TX deviation control is an important aspect for you.

Armel, F4HWN

[VK2BEA]

Yes, transmit deviation.
I presumed that changing the bandwidth on (wide/narrow) would also change the transmit deviation.
If it doesn't, then that would explain why it is low.
I think its register 40<11:0> RF Tx Deviation Tuning (Apply for both in-band signal and sub-audio signal). 0=min; 0xFFF=max

The deviation should be greater (5 kHz peak) on wide(er) FM.

Michael VK2BEA / NV3Z

[armel]

You're absolutely right Michael, according to the datasheet, register 0x40<11:0> controls the TX deviation. I haven’t experimented with it myself, and at this point, I don’t plan to make it configurable in the firmware.

That said, if you're comfortable with coding, feel free to tinker with it, it could be an interesting area to explore.

Armel, F4HWN

[VK2BEA]

Yes, OK I'll give it a go but I'm not familiar with the 'codespace' IDE, so this will be interesting 8-)

...

.... edit: some time later ....

I can run make

@VK2BEA ➜ /workspaces/uv-k5-firmware-custom (main) $ make

This works and I see the changes compile but

@VK2BEA ➜ /workspaces/uv-k5-firmware-custom (main) $ ./compile-with-docker.sh bandscope
📺 Bandscope compilation...
   text    data     bss     dec     hex filename
  60656      52    3376   64084    fa54 f4hwn.bandscope

always has the same binaries ...

@VK2BEA ➜ /workspaces/uv-k5-firmware-custom (main) $ md5sum compiled-firmware/*
6b5892988490b57f4cfc90cd4ece0aa4  compiled-firmware/f4hwn.bandscope
f7b2639f7832fac096b6179622834bb3  compiled-firmware/f4hwn.bandscope.bin
23cecf7afc422a1e31284ec263582573  compiled-firmware/f4hwn.bandscope.packed.bin

If the docker build is not using the source code I changed ... what is it using?

To mimic what the docker script does for 'bandscope' I compiled using ..

ENABLE_SPECTRUM=1 \
        ENABLE_FMRADIO=0 \
        ENABLE_VOX=0 \
        ENABLE_AIRCOPY=1 \
        ENABLE_FEAT_F4HWN_GAME=0 \
        ENABLE_FEAT_F4HWN_PMR=1 \
        ENABLE_FEAT_F4HWN_GMRS_FRS_MURS=1 \
        ENABLE_NOAA=0 \
        ENABLE_FEAT_F4HWN_RESCUE_OPS=0 \
        EDITION_STRING=Bandscope \
        TARGET=nv3z.bandscope make

and I see f4hwn.packed.bin in the root (which is the new build). (I don't know if that is what I should have done)

Anyway, I successfully fixed the low deviation using a larger value in register 0x40 than the default, so this is working for me. (the default was 0x4D0 so I made it 0x600 ... to scale it based on the deviation I measured)