#synth ref frequency #synth


DC1MAK
 

Hi Hans, hi all,

somwhere i read about, that the reference clock of the Silab synth in a U3S MUST be at 27 MHz.
Of course this depends on the firmware of the U3S.
But i think, i heard that there is another cause.
Something with the 1.4648 Hz spacing of the wspr tones ?
Does anybody know about it and can explain that
to me ?


F1BFU - Fr - 79
 

This is explain in assembly manual of Si5351A of QRPLabs.

"The SiLabs Si5351A chip is a cousin of the famous and popular Si570 chip, but is much less expensive. Unlike the Si570 however, the Si5351A has no quartz crystal inside. An external reference oscillator or crystal must be provided. The reference frequency may be 25MHz or 27MHz. In this module, a 27MHz crystal is used. This frequency is chosen because it is possible to configure the chip to produce the exact 1.46Hz tone spacing for WSPR, on any amateur radio band from 2200m (136kHz) to 2m (145MHz). A 25MHz crystal cannot provide WSPR tone spacing on the 2m band. In summary, the Si5351A chip synthesizes output frequencies in three stages. The block diagram of the chip is shown to the above (from the Si5351A datasheet). First, a crystal reference oscillator is multiplied up to internal frequency in the range 600-900MHz. There are two PLL’s in the chip, each may be selected to choose a different internal frequency. At the second stage one of the PLL frequencies is divided down to each of up to three required output frequencies. Both the upward multiplication to the internal PLL Output frequency, and the division down to the output frequency, use fractional ratios – an integer plus a fractional part consisting of 20- bit numerator and denominator. Optionally a third division stage may be configured to divide each output by a power of 2, for a maximum division ratio of 128. It is used for generating low frequency outputs between 8kHz and 500kHz."

73 QRO
Gilles F1BFU /Fr

Le mer. 21 nov. 2018 à 21:38, DC1MAK <dc1mak@...> a écrit :
Hi Hans, hi all,

somwhere i read about, that the reference clock of the Silab synth in a U3S MUST be at 27 MHz.
Of course this depends on the firmware of the U3S.
But i think, i heard that there is another cause.
Something with the 1.4648 Hz spacing of the wspr tones ?
Does anybody know about it and can explain that
to me ?



--
Gilles - F1BFU/FR


Alan G4ZFQ
 

This is explain in assembly manual of Si5351A of QRPLabs.
In this module, a
27MHz crystal is used. This frequency is chosen because it is possible to
configure the chip to produce the exact 1.46Hz tone spacing for WSPR, on
any amateur radio band from 2200m (136kHz) to 2m (145MHz).
What we do not know is just how sensitive the WSPR decoder is to frequency accuracy.
I use 26MHz TCXOs and have decoded my 2m transmissions with no problem.
There was some discussion a few months ago about using even a 10MHz reference for the Si5351 but nobody thought to try WSPR using that.
WSPR has to cope with borderline HF propagation, TX and RX drift.
Without knowing the tolerances of the WSPR decoder it is wise to keep as close to the specifications as possible but I suspect the tolerances are wider than many think.

73 Alan G4ZFQ


DC1MAK
 

merci beaucoup Gilles, many thanks Alan,

i knew, ther was something written somewhere....
in the next weeks i'll try different clocks with
my Arduino-based transmitters....
I'll report my experiences....

73 de DC1MAK micha


Alan G4ZFQ
 

somwhere i read about, that the reference clock of the Silab synth in a U3S MUST be at 27 MHz.
But i think, i heard that there is another cause.
Something with the 1.4648 Hz spacing of the wspr tones ?
Hans has said that a 27MHz is required to make the tones exact.
I do not know if anyone can say what amount of error is permissible with WSPR but I use a 26MHz TCXO with no apparent disadvantage. (But it is difficult to be certain, maybe I do get missed sometimes?)
The U3S will operate with a wide range of frequencies, I've tried 10MHz.
But I do not know of anyone properly testing for signal purity.

73 Alan G4ZFQ


Hans Summers
 

Hi Alan

27MHz was required for my algorithm to get the WSPR tone spacing exact on 2m. On all lower bands than that, 25MHz or 27MHz both produced exact solutions. Therefore I chose to use the 27MHz crystal in QRP Labs kits. 

26MHz will be fine also, all across HF. Holes in coverage of my algorithm (for exact WSPR tone spacing) only started to appear beyond 40MHz as I recall. So I couldn't say for sure that 26MHz will produce accurate tone spacing for WSPR on 6m, 4m and 2m bands. But you said you used it successfully on 2m so I think it the proof of the pudding is in the eating :-) 

73 Hans G0UPL

On Thu, Feb 28, 2019 at 7:40 PM Alan G4ZFQ <alan4alan@...> wrote:
> somwhere i read about, that the reference clock of the Silab synth in a
> U3S MUST be at 27 MHz.
> But i think, i heard that there is another cause.
> Something with the 1.4648 Hz spacing of the wspr tones ?

Hans has said that a 27MHz is required to make the tones exact.
I do not know if anyone can say what amount of error is permissible with
WSPR but I use a 26MHz TCXO with no apparent disadvantage. (But it is
difficult to be certain, maybe I do get missed sometimes?)
The U3S will operate with a wide range of frequencies, I've tried 10MHz.
But I do not know of anyone  properly testing for signal purity.

73 Alan G4ZFQ




HF
 

Hi Alan,
Did you ever figure out what the decode sensitivity to errors in the tone offset frequency is?  I'm trying to scheme an inexpensive transmitter for WSPR on the 70 cm and maybe even the 33 cm band.  I think I can get the frequency stable enough but want to know how coarsely I can set the frequency shifts.  And, what if some shifts are different from each other?
Cheers,
Halden VE7UTS


HF
 

Hi all...
The Si5351a datasheet says its reference frequency has to be 25 or 27 MHz.  Others here have commented on using 26 and 10 MHz.  What are the disadvantages of using other frequencies, other than the inconvenience of scaling the divider settings to compensate for use of a non-standard reference?
Cheers
Halden VE7UTS


Hans Summers
 

Hi Halden

SiLabs engineers confirmed somewhere on their forum that although the datasheet says the reference crystal should be 25 MHz OR 27 MHz, in fact anywhere in the range 25-27 MHz os what they really meant. So according to that you can use 26MHz without disadvantage. 

Use of 10MHz works but is way outside spec; aside from reducing the upper frequency limit you can attain, it also degrades the spectral purity. It doesn't matter on low frequencies but on VHF it seemed quite bad - there was a post in this forum ages ago about it but I can't find it now. 

73 Hans G0UPL 



On Thu, Mar 25, 2021, 05:36 HF via groups.io <incorridge=yahoo.com@groups.io> wrote:
Hi all...
The Si5351a datasheet says its reference frequency has to be 25 or 27 MHz.  Others here have commented on using 26 and 10 MHz.  What are the disadvantages of using other frequencies, other than the inconvenience of scaling the divider settings to compensate for use of a non-standard reference?
Cheers
Halden VE7UTS