Reply
Highlighted
Posts: 3
Registered: ‎08-26-2017

Si468x DAB_DIGRAD_STATUS never shows 'valid'

Hello all,

I'm new of this forum and I'm trying to figure out how to solve an issue we find on our DAB Si468x tuner board.

We, my collegue and myself, developed a board featuring Si4684 tuner, STM32F4 MCU and STL500 codec plus some other ICs that are currently shut down, to avoid influence on tuner performance during debug.

We were able to correctly program the Si4684, system answers that is up and running, we were able to load and read DAB frequencies as well as read default ones.

When we trigger a DAB_TUNE_FREQ and then, after waiting proper INTs, DAB_DIGRAD_STATUS what we always have is an invalid tuning, with ACQ = 0 and SNR=0 (or eventually SNR=-10).

This happens with all the frequencies listed in EU table (we live and work in Europe and we are interested in EU transmission, as for now).

Our board routing is the same shown in figure 4 of AN851, since we are currently using headphones as antenna.
RSSI measurements vary from 20 to 34, depending on frequencies.

We are now forcing Si468x to tune on one specific frequency for which we know that the transmissions are fully working (we can receive and hear music using commercial DAB tuner) without any ACQ flag.

We are following the AN649 DAB flowchart step-by-step. We double checked every command and every answer and everything seems perfectly coherent with the application note.

One thing more: we loaded FM firmware, we tuned the receiver on to a known station and we were able to receive audio and RDS informations, such as PI (RDSSYNC and PIVALID were '1's).

After several attempts and tests, I then found this http://community.silabs.com/t5/Other-Products/Si4684-no-DAB/td-p/171750 which seemed related to this issue. I tried modifying bias current but nothing happened.

 

Has anyone ever tried using headphone GND as antenna for DAB?

Which front end matching network do you suggest for headphones?

 

Many thanks.

 

Kind regards,

 

Giacomo

Posts: 77
Registered: ‎02-15-2015

Re: Si468x DAB_DIGRAD_STATUS never shows 'valid'

Can you check (using an oscilloscope), whether the crystal oscillator is working after booting up the DAB-Firmware?

And, if it's working, check if the amplitude is the same as when booting the FM firmware.

 

Best regards,

Andi

Posts: 3
Registered: ‎08-26-2017

Re: Si468x DAB_DIGRAD_STATUS never shows 'valid'

Hello Andi,

thank you for your reply.

 

We checked OSCH during a debug session finding waveforms that were pretty much the same for FM and DAB.

Here are the screenshots:

 

DAB

DAB XTAL.png

 

FM

FM.png

 

Do you think this waveforms are correct?

 

Many thanks!

 

Best regards,

Giacomo

Posts: 77
Registered: ‎02-15-2015

Re: Si468x DAB_DIGRAD_STATUS never shows 'valid'


GCalanchi wrote:

 

...

Do you think this waveforms are correct?

 

 

...

Yes, they look correct.

Did you try to connect a "regular" antenna to RF input instead of using the a headphone antenna?

 

Best regards,

Andi

Posts: 3
Registered: ‎08-26-2017

Re: Si468x DAB_DIGRAD_STATUS never shows 'valid'


Andi-84 wrote:

Yes, they look correct.

Did you try to connect a "regular" antenna to RF input instead of using the a headphone antenna?

Best regards,

Andi


Well, unfortunately we have not tried yet with the suggested wire antenna network (i.e. 33 pf series capacitor, 18 nH + 2.7 pF VCO spur filter, 120 nH parallel inductor and 22 nH inductor to be shorted with VHFSW).

We need to wait for the components to come from our broker. We designed the board with two ways mounting components (HP antenna and wire antenna) so we should be able to replace one matching network with the other quite quickly, fortunately.

 

We will come back as soon as we try the new matching network.

 

Many thanks.

 

Best regards,

 

Giacomo