I got a Si5348 a few months back. Board name : Si5348-EB REV 1.1.

Following the steps in the reference manual, without any changes  to the jumpers present on board, i connected the eval board through the provided USB cable to my PC after installing ClockBuilder Pro software. After successful driver installation, the board was recognized as "Si5348 Rev Unknown". Going into the Eval board planning GUI, i get this error :

DUT/Board Mismatch:

There is a mismatch between the device present on the EVB and the type of board. The device is a Unknown, but a Si5348 board is present (board ID is 48d).

Please contact Silicon Labs for assistance.

(An image is attached of the same in this post)

All the register reads were returning values of 0xFF. The Board ID Code as mentioned in the GUI is the only register which seems to have returned a valid value - 43. After clicking OK at the error dialogue, DEVICE_PN_BASE register as mentioned in the GUI has a value of Si16665.

I have verified all the jumpers to be in their right places, especially the ones at J36 which acts as a shunt for the MCU and the Si5348 SPI communication.

After this i tried interfacing with the board directly through SPI with an Arduino with the following pin configurations :

Si5348 -------------- Arduino Due

A0/CSB ------------> digital_pin_10

SDA/SDIO -------->  MOSI

A1/SDO -----------> MISO

SCLK ---------------> SCLK

GND ---------------> GND

The code is on another device but i can post it tonight. But i am fairly certain i was doing the SPI communication with the proper steps correctly -

Set Address -> Set Page number 0x00 -> Set Address -> Set register address 0x02 -> Read data -> Send dummy 8 bit values.

But for each command and also for the final read command, i keep getting 0xFF. Checked the signals coming out of the pins with oscilloscope just to be sure. Also swapped the MISO to MOSI just in case i messed up the connections. But no successful reads.

Then i removed all the shunt jumpers on J36 to try to use the I2C interface. With A0 and A1 pins grounded, i connected the SDA/SDIO to SDA of Arduino Due and SCLK to SCL of Arduino Due. Removed jumper from JP4 - I2C_SEL on the EVAL board to enable I2C. Ran the I2C Scanner test code on arduino - which is going to send a byte of data to all possible i2c address and displays any address to have acknowledged the byte transfer. This module always worked with any i2c device. But with the eval board no device was found. Tried this with the JP14 I2C_EN jumper open and closed but with no success. 

Also tried all the above with and without the provided SiOCXO connected with power on and off. How to rectify this problem? Or is my board dead?

Hi,

I am a bit puzzled by the propagation delay reported in the datasheet for the Si53340-45.
I think there is a unit error in the datasheet.
Shouldn't it be 650-1050 ps instead of ns?

Can you confirm that this is an error?

Best

 

Hi,

Is there a field programmer available for Si5335?

I have already created an orderable part number and we are procuring the same.

But if we have any hiccups during bringup, we want to keep a field programmer handy to program some blank Si5335 chips.

Also, are these chips reprogrammable or one time programmable only?

Thanks

Nandini

 

 

We are using the Si5338 in a new design. We are using the ClockBuilder tool to define our configuation. When having our 100MHz clock pass through to 3 outputs (CLK0-CLK2), everything operates properly. However, when we use ClockBuilder to create a 250MHz output (using a PLL) on CLK0 and CLK1, we don't get any outputs, even CLK2 which is just pass-through clocks. The only thing that changes between the 2 configurations is the ClockBuilder-generated .h file. Any help would be appreciated. I've attached the configurations here. Thanks.

Hi，I am using Si5345A-D-GM. When I generate a register configuration script via ClcokBuilder Pro software, there are about hundred of registers need to be configured.. I try to find the default value of the registers, to let the chip work by configure the minimum registers. In the Reference Manual, it says I can get the default values of the register map settings by accessing the Silicon Labs part number lookup utility, But I can't find it in the link. How can I get the default values of the registers? Thanks!

Hi,

tried to switch CLK1 output stage from MS1 to MS0 on-the-fly.

Used a function like this:    Si5338C_read_modify_write_byte(32, 0xa0, 0xe0);  // register, data, mask

Result: output CLK1 went off. Switching back with:  Si5338C_read_modify_write_byte(32, 0xc0, 0xe0); 

made the output working again.  Frequencys on CLK0 and CLK1 where low between 12 MHz and 25 MHz.

Are ther any prerequisites to do so or is switching on the fly generally not working?

Thanks!

Hi

I am programming SI -5341 connected to ZYNQ ARM using SPI

I am not able to read a default register. only 0xff is coming on MISO line.

We have checked MOSI and CLK on chipscope both are coming.

Please help me.

Are the control input pins (I2C_SEL, IN_SEL[1:0], RSTb, OEb ...) of a blank SI5340 device 3.3V tolerance?  Can SI5340 be set up in auto switch mode between available inputs?  Thank you

Are the control input pins (I2C_SEL, IN_SEL[1:0], RSTb, OEb ...) of a blank SI5340 device 3.3V tolerance?  Can SI5340 be set up in auto switch mode between available inputs?

HI,

I am planning to use Si5351A Programmable oscillator in my project.

27.12MHz crystal is already available in my current design. Can I make use of this 27.12MHz crystal as the reference clock for Si5351A whereas in the datasheet it is mentioned to use 25MHz 0r 27MHz crystals.

SO please let me know, I can make use of 27.12MHz crystal or I need to use 27MHz crystal only.

 

Thanks

Hi, 

I have two questions regarding the input/output specifications for the SI55307 and SI55304 series of muxes. 

1) Why can the CML input not be DC coupled? Is it correct that part will not function at all if attempted to be driven by DC coupled differential CML driver?

2) For CML outputs, the electrical specifications section shows no common mode voltage and shows only single ended swing, no other differential characteristics. The output termination section shows only AC couple termination, not DC couple for CML output. Can you provide any more specifications on these differential outputs? Do these parts not support DC coupled CML outputs? If not, can you please explain further?

As title, the antivirus Avira found that there is virus included in the downloaded ClockBuilder Pro, Version 2.28.1 [2018-09-24].

Which product should I use to get lowest spurious for multiplying 20MHz or 100MHz up to 600MHz.

Si5325 seems to work well, spurs below -70dBc but is no longer available.

Si5326 has high close in spurs due to jitter attenuator, but can mimic Si5325 performance if programmed as a 5325 (which I can achieve by setting RATE=LH on 5326 which is reserved, not sure if this is valid)

Si5338, which is recommended replacement for 5325, has *very* high spurs (-42dBc at 1MHz offset) generating 600MHz from onboard 25MHz crystal.

Have not tried any other parts.

Hi,

 

I am looking for alternative ways to solve clock design for ethernet devices. I have an noisy 10MHz input to Si5319 and expecting to obtain the 25MHz output. Need advise on the below question.

1. Is the Si5319 capable of deriving 25MHz from 10MHz?
2. When the dirty 10MHz clock stop, can it switch to 10MHz osc automatically and seamlessly without glitch? The glitch may disrupt the operation of target ethernet devices.
3. Is Si5319 or Microprocessor Controlled Parts able to preprogramed with desired output clock frequency or it require to reprogrammed everything it power up?
4. Any software/tools can help to derive the parameters for the Si5319?

Is it possible to directly bypass one of input clock signals to output pad without any modification for debug purposes?

If yes, how to program this debug mode?

 

Thanx, Jury.