Si5351

[brandon7861]

I've had nothing but bad things to say about using these for crystal replacements, but at $2.77 a piece, I couldn't help but add a couple to the last mouser cart.

What puzzles me is the excessive packaging and humidity warnings. The package sounds like it has desiccator beads rolling around in it and I an now afraid to open it because I am not ready to use them. If mouser is that worried about moisture, what am I supposed to do after it is on the board? Bury it in wax? Those cheap modules on amazon don't do that. How can moisture be an issue in storage and not in the application?

My next bit of confusion was that the highest stability crystals Mouser had were 10ppm. I assume that is a slow drift and not all over the place, but if thats the best crystal I can buy (at the same price as a complete module on amazon), what crystals did they use on those cheap modules? 50ppm? 100? Looks like I will I need to add a micro just to keep updating this thing as it drifts. If I need to build a board with a programming interface and a bandpass filter, I am not sure it would fit in a radio.. I would have ordered the ones with a built in VCO, but I am not too good with soldering things with pads underneath the part, so I didn't. I think this is an AM only project.

 

[TM86]

I used the Si5351 and some support parts to replace the PLL in a Sidebander VI. Got good reports on the air for the few contacts I was able to make.

 

[brandon7861]

I used the Si5351 and some support parts to replace the PLL in a Sidebander VI. Got good reports on the air for the few contacts I was able to make.

How was the frequency stability, did you have to keep adjusting it (manually or automatically) to hold a conversation or was it good on its own with a set program?

 

[TM86]

It was good once set. I didn't do a 24 hour test or anything, I don't have the patience for that.

One thing I think helped a lot was an idea I stole from Nomad. I set the chip to generate a higher frequency than needed, then divided down to what I wanted. The idea being to reduce jitter. I also fed the divided output through a three section low pass filter to make it look more like a sine wave.

 

[brandon7861]

One thing I think helped a lot was an idea I stole from Nomad. I set the chip to generate a higher frequency than needed, then divided down to what I wanted.

I will remember that trick. Thanks for mentioning that, and thanks to Nomad as well! The snow will be 3' deep before I get to trying it though.

 

[nomadradio]

My favorite quick divide-by-two circuit is a type "D" flip-flop.

"D" as in "Data". When the clock pin is triggered, the logic state of the D input appears on the Q output. Usually there is also an inverted copy of that output, labeled "not Q". Normally indicated by a bar above the letter. Don't know how to do that on this forum.

The inverted output connects to the D input. This way, every time the clock pin is toggled, the opposite logic state appears on the input. Provides a divide-by-two with a near-perfect 50/50 duty cycle.

Just need a chip fast enough for the frequency you want.

73

 

[DTB Radio]

I've never had any issues storing these things out in the open. They are NOT that delicate, lol! As long as you don't expose them to anything more than normal humidity, I doubt you would ever have any issue at all.

As for drift, I've used these in a number of projects, and I've not had any appreciable drift issues. They can be fine-tuned in software to compensate for slight differences in control clock frequencies, but thats a 1-time thing and then no further correction is needed.

As was mentioned already, using a good filter will go a long way towards cutting down the harmonic noise in the square-wave output, but I've tried a few straight-up and not had any appreciable noise problems.

I've had nothing but bad things to say about using these for crystal replacements, but at $2.77 a piece, I couldn't help but add a couple to the last mouser cart.
View attachment 66680
What puzzles me is the excessive packaging and humidity warnings. The package sounds like it has desiccator beads rolling around in it and I an now afraid to open it because I am not ready to use them. If mouser is that worried about moisture, what am I supposed to do after it is on the board? Bury it in wax? Those cheap modules on amazon don't do that. How can moisture be an issue in storage and not in the application?

My next bit of confusion was that the highest stability crystals Mouser had were 10ppm. I assume that is a slow drift and not all over the place, but if thats the best crystal I can buy (at the same price as a complete module on amazon), what crystals did they use on those cheap modules? 50ppm? 100? Looks like I will I need to add a micro just to keep updating this thing as it drifts. If I need to build a board with a programming interface and a bandpass filter, I am not sure it would fit in a radio.. I would have ordered the ones with a built in VCO, but I am not too good with soldering things with pads underneath the part, so I didn't. I think this is an AM only project.