Re: Writing incorrect data to the Si5351a registers - Destruction of a Synth module??? #synth


Hans Summers
 

Hi Colin

I very much doubt that it is possible to destroy an Si5351A by mis-configuring it. I have never seen such a thing - and believe me, I have mis-configured PLENTY of Si5351A chips :-)

73 Hans G0UPL

On Tue, Jan 16, 2018 at 11:27 PM, Colin Buckup via Groups.Io <rotbaron77@...> wrote:
Hi gang,
Mea Culpa! Mea Culpa! Mea Culpa!
I believe I destroyed a perfectly fine Synth module... Question remains exactly how it happened???...

Here is the story:
As part of my multi-band 5W CW transceiver using an Arduino and various QRP-labs modules, I was tinkering with the code running on the Arduino to control the synth module on the QRP-Labs Arduino shield. Having gotten the basic part of frequency generation to perform as planned, I advanced the project and included the GPS disciplining function, using the 1PPS from the GPS as interrupts and the 16 bit counter on the controller to calculate the real oscillating frequency of the 27MHz xtal used for the synth. So far so good, all still working fine and disciplining the synth module was working well and results were great. (by the way, had to re-route some of the signals between Arduino and the shield to handle the correct interrupts and counters, but that's ok).

Now here is where things went south, yours truly made a stupid coding mistake where I was subtracting an already negative value instead of adding it for the compensation... Depending on which side of the curve the process started, it would either work fine or... it would go into a runaway mode where I was compensating the programming of the Si5351a with wrong values. Every time the process ran (I had it on a 20s GPS cal loop) it would correct in the wrong direction further and further and write more and more garbage into the Si5351a.

Another factor to the problem is that due to weather the air is extremely dry here right now and static is in the air, clothes, fingertips, etc... have been zapping myself all day long...

So: having said all that, I had the project running for several hours on a breadboard and all of a sudden I notice that the frequency being generated is very off... I pulled the synth module from the breadboard I was using and plugged it into my QRP-Labs VFO kit. Turn it on and... Argh... that dreaded error appears on the screen: Si5351a problems...

OK. I found my error in the code, that's now fine, and I actually put a safety in place where the correction can't go above or below a certain threshold to avoid this in the future. What I don't know for sure is if I blew the synth module by writing stupid values into the frequency programming registers or if I fried it by static discharge...
Also: If it was by register value error, is there a way to recover from this? Or do I need to order a new synth module?

Hans, or anyone who might have experienced something similar, input is welcome!!!

And, YES, it is absolutely MY fault one way or the other. Mea Culpa!
Joys of kit building and tinkering, but wouldn't change it for anything!!!

73 de N5GG
Colin - Dallas, TX


Join QRPLabs@groups.io to automatically receive all group messages.