[caption id="attachment_927" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-5.jpg) Completed DDS module[/caption]
The DDS modules have an interesting charge. Their purpose is to generate very precise clock sources to the phase-locked oscillators (PLOs, or sometimes referred to by the different, but related Phase-Locked Loop PLL). The PLOs can be thought of a variable frequency multipliers. Often, they have a limit with respect to the multiplication value they can use. There are "fractional" PLOs that can multiply by some value plus a fraction. The way to get a lot of precise variability out of a PLO is to adjust the reference frequency by extremely small amounts. A DDS functions in an entirely different way than a PLO. A DDS is a "Direct Digital Synthesizer", which is essentially a sine-wave generating digital-analog converter (DAC). When a given frequency is desired the device determines the value of what the output should be at each DAC cycle for a perfect sine wave. Because the refresh speed of the DAC is limited, the output then passes through a low pass filter to remove the majority of the artifacts. A major advantage to the DDS is that they have very low phase noise. A PLO, in general, will multiply the phase noise of it's input. This make using the DDS as input to the PLO a wise choice.
[caption id="attachment_923" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-1.jpg) Removing extra solder mask[/caption]
Anyway, onto construction… This particular revision of the PCB has some extra solder mask on the voltage regulator. This same error is present on all boards with this voltage regulator package. To remove it you can just scrape it off. It's worth noting that the solder mask comes off better when it's chipped off.
[caption id="attachment_924" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-2.jpg) "tinning" the exposed pad[/caption]
I'm always conscious of the speed that copper oxidizes, so I tinned the pad right away. You probably already know that I used solder flux, don't you? I did.
[caption id="attachment_925" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-3.jpg) Squarer circuit before modification[/caption]
Recently, Scotty suggested the addition of a few components to the output of the DDS. These include a pair of capacitors and an inductor. This image is the "before" shot. He suggested a method for changing it that is slightly different that the way I did it.
[caption id="attachment_926" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-4.jpg) Removing solder mask and trace segment[/caption]
I started by removing all the solder mask on that segment of the trace. Then I cut a small gap in it for the inductor. I also removed the resistor that was there before.
[caption id="attachment_936" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-1-3.jpg) Modification complete[/caption]
As you can see in this photo, I crammed the resistor back in, as well as the inductor. Finally, I bridged the gap between this mess and the border with the requisite pair of capacitors. I didn't mean to get 0402-size capacitors, but I did anyway. They are ridiculously small.
[caption id="attachment_932" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-1-2.jpg) Bad modification[/caption]
I wasn't going to include this photo because it didn't turn out well, but this is the other DDS module. On the capacitor on the left you can see a bit of solder shorting the capacitor. They are so small that I couldn't see it with the naked eye. It took looking at this photo before I saw it.
[caption id="attachment_930" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-8.jpg) Squarer output[/caption]
This is the output of the squaring circuit with the modification through a 20 Mhz scope probe. It's no where near the waveform that Scotty has, but it's also unloaded and his O'scope has 100 Mhz of bandwidth. One thing I love about this image is that it's an almost textbook example of a band-limited square wave.
[caption id="attachment_929" align="aligncenter" width="540"]!(http://alternet.us.com/wp- content/uploads/2010/05/DDS-7.jpg) Extra DDS outputs[/caption]
This is an example of what the extra outputs of the DDS look like. This is an image contains at least 3 different traces of the scope. I couldn't get a solid trigger to capture just one. I have no idea what to expect, here. For now it is what it is. Looking back, the main page for the DDS says that alias frequencies and harmonics are to be expected. Finally, I noticed that it also says to remove R3 which attenuates this output. I haven't removed this resistor yet, but I will soon.
I'll be working on the PLO soon! I'm getting really close!