NBFM from DDS chip

[MikeN]

Hi

Has anyone given thought to producing NBFM using the AD 9854 or
similar DDS chips, for multiplying or heterodyning up to 2m or 70cm.

This could be an interesting project. I see Analog Devices has an
application note AN-543 which describes a stereo WBFM generator with
output on 90.3 MHz.

73 de MikeN ZL1BNB

 

[Tim Wescott]

Hi

Has anyone given thought to producing NBFM using the AD 9854 or
similar DDS chips, for multiplying or heterodyning up to 2m or 70cm.

This could be an interesting project. I see Analog Devices has an
application note AN-543 which describes a stereo WBFM generator with
output on 90.3 MHz.

73 de MikeN ZL1BNB

I've been considering it for years; it should work OK as long as your
sample rates are high enough. In fact there are nice, small DSP chips
now that should implement a good FM receiver to go along with it for a
complete transceiver.

It's never been a paying project for me, and it's never made it to the
top of the "home brew" pile, so it stays in "would be cool" land.

 

[Steve Nosko]

I've been considering it for years; it should work OK as long as your
sample rates are high enough. In fact there are nice, small DSP chips
now that should implement a good FM receiver to go along with it for a
complete transceiver.

It's never been a paying project for me, and it's never made it to the
top of the "home brew" pile, so it stays in "would be cool" land.
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

I wouldn't be surprised if this wasn't the algorithm used in DSP generation
of FM.

It was used to generate four part music, in software, back in the early 80's
with a 0.9 MHz 6809 CPU.

 

[Paul Keinanen]

Has anyone given thought to producing NBFM using the AD 9854 or
similar DDS chips, for multiplying or heterodyning up to 2m or 70cm.

Doesn't the smaller DDS chips have serial loading of the phase
accumulator addend ?

To get communication quality audio, at least 8 kHz sampling rate is
required. If the phase accumulator addend is 32 bits, the serial clock
would have to be more than 2 MHz.

On 9854, the 14 bit phase offset word could also be used, thus
reducing the serial data clock to 1 MHz, well below the maximum 10 MHz
data clock frequency.

Maybe the FSK ramp rate could be used to reduce the clock rate needed.

This could be an interesting project. I see Analog Devices has an
application note AN-543 which describes a stereo WBFM generator with
output on 90.3 MHz.

A stereo WBFM generator would require at least 120 kHz sampling rate,
which would be quite trivial with parallel tuning words, but would be
quite demanding if 32 or 48 bit tuning words needs to be loaded each
audio sample time.

Paul OH3LWR

 

[Energy Conserved]

Hi

Has anyone given thought to producing NBFM using the AD 9854 or
similar DDS chips, for multiplying or heterodyning up to 2m or 70cm.

Texas instruments: TRF4400

Single-Chip RF Transmitter for 433 MHz ISM Band
420-MHz to 450-MHz Operation
FM/FSK Operation for Transmit
24-Bit Direct Digital Synthesizer (DDS) With 11-Bit DAC

regards