How to control the Si5351 pls ?

**iw2fvo** · - 18th August 2016, 21:25

Good day to all in this forum.
I will like to know if there is an algorithm capable of controlling the Si5351 frequency generator.
I really do need just a simple frequency to data_out to set up the correct frequency on the generator...
Thanks in advance for any assistance.
Regards,
Ambro

**Charlie** · - 10th November 2017, 14:20

Did you ever get this working?

**Charlie** · - 12th November 2017, 16:30

Maybe I'll ask a broader question. The SI5351 device can generate a number of independent clocks, from a few Hz to over 100 MHz. Sounds pretty useful as a replacement for a VFO in radio projects. I'm pretty sure I can get it going, but the command structure is quite complicated, and I was wondering if anybody had implemented it in PBP yet? I have no problem with I2C or actually talking to the device, I was just hoping to save time on implementing the commands. There is an Arduino sketch for it, but reverse engineering that is probably more work than just starting from scratch. Has anybody taken a poke at this one?

**iw2fvo** · - 15th July 2018, 15:34

good day to all in this forum,

is there some implementation of PICBASIC code for controlling such a devices up to now ?

Thanks
Regards,

**Charlie** · - 5th December 2018, 14:19

The silence remains deafening :-)

**iw2fvo** · - 6th December 2018, 08:15

I think nobody has implemented it...

In the ham radio club everybody is very interested on this item.
Still hoping.
iw2fvo,

**F1CHF** · - 10th December 2018, 10:13

hello here is F1CHF
did you try a Google search with SI5351
I see some example
try
https://www.google.fr/search?q=SI535...w=1024&bih=679

Francois

**Charlie** · - 10th December 2018, 14:40

Francois - I did, and have a really good idea how to control it. I can't actually imaging doing anything anymore without online research.
However, the calculations require floating point math, which PBP does not support. If I want to go Arduino, I can be done this afternoon. There are lots of examples. In fact, I have an Arduino version running.
If I want to use PBP, it will be many weeks of coding. So I was hoping somebody had an elegant solution for the calculation already worked out that I could use as an example.

**richard** · - 10th December 2018, 21:57

big numbers and floats can be handled in pbp

http://melabs.com/resources/fp.htm

http://dt.picbasic.co.uk/CODEX/N-BitMath

**iw2fvo** · - 11th December 2018, 08:11

I see a lot of Arduino / C code on the web ...
nothing in PBP ! up to now .
Still hoping.
regards

**richard** · - 11th December 2018, 11:50

Originally Posted by Charlie

If I want to use PBP, it will be many weeks of coding. So I was hoping somebody had an elegant solution for the calculation already worked out that I could use as an example.

if you show us the calculations you are trying to achieve and where the results fall short . along with the code/methods you have tried then perhaps more eyes can offer some suggestions . its not absolutely necessary to be interested in or have any use for the chip to enjoy the coding challenges , this seems to be your best shot here imho.
I assume some sort of effort has been made by those seeking a handout

**Charlie** · - 11th December 2018, 15:51

Originally Posted by richard

i
I assume some sort of effort has been made by those seeking a handout

Nice - I love how we are all polite to each other in the forum.

**richard** · - 12th December 2018, 12:50

the fp library can be easily incorported into pbp and did the job admirably for this complex calculation
overheads were minimised by identifing and precalculating as many constants as possible
i have not provided the pbp code since the op of linked thread is still working on his own solution
http://support.melabs.com/forum/picb...erature-sensor
C representation for clarity of calculation strategy

Code:

'const float k[]={-611.715,0.032791,-7.2032e-7,9.9704e-12,-5.6892e-17};//k0 to k4 as constants
'float adc16=36636.0,tres;                 int res    run tot
'tres  = k[4]*adc16*adc16*adc16*adc16;//  -102.49    -102.49
'tres += k[3]*adc16*adc16*adc16;//         490.27     387.78
'tres += k[2]*adc16*adc16;//              -966.81    -579.03
'tres += k[1]*adc16;//                    1201.33     622.3
'tres += k[0];//                          -611.715     10.585
'result             //                      10.59

despite the sarcasm if someone provides a formula/method and some figures, who knows i might play along.
i really can't be bothered reading the data sheet and coming up to speed for a device i won't use ever.
its your call.

pic of usual crappy ds version of formula attached