To search within the Clifton Laboratories site, enter your search term below.
 

Introduction

Day 1:  1.5 Hours (Total time: 1.5 hours)

Stage 1 successfully completed. (Stage 1 installs the input power supply switching and filtering and the +5V regulator for the microcontroller.)

The 1.5 hours is measured from opening the box to finishing installation,  resistance testing and voltage measurement tests

Dario says "the solder masked and stenciled version will be a tremendous assembly improvement" and "a silk screened version will go much faster."

Day 2:  2.5 Hours (Total time: 4.0 hours)

Stage 2 successfully completed, including software test. (Stage 2 installs the RS-232 interface and the 18F4620 microprocessor.)

Problems found: Discrepancy between schematic and parts list for R517. [My error, schematic and part shipped were right; but forgot to change the assembly manual—manual now corrected. JRS]

 
Also, the home page for the recommended terminal program “Terminal.exe” remains unreachable. [It’s still down today (Wednesday, 19^th July 2006). Terminal remains available from an alternate site however. http://www.hw-server.com/software/termv19b.html  JRS] 

I took my time and made triple sure everything was where it is suppose to be installed. I also cleaned all flux residue with alcohol so my assembly takes longer.

Day 3:  2.5 Hours (Total time: 6.5 hours)

Soldering the AD9851 is difficult without the  proper tools and experience. I highly recommend this be a factory soldered component.

[I’m leaning that way, and probably will make the entire DDS daughter board an assembled and tested module. It’s definitely tricky to solder a 40 pin/inch surface mount chip with conventional techniques. JRS] 

I was not able to pick up the 30Mhz Xtal or the DDS signal at 8455. Is there any other way to verify the DDS is working?

 [Yes, if you have oscilloscope, look at the signal output on Pin 6 of the DDS header. Sample voltage and oscilloscope waveforms are provided with the annotated schematics under the Troubleshooting section. I also highly recommend using the frequency command mode to set the DDS to output a steady frequency that you can listen to with your receiver. I’ve revised the Assembly Manual to explain this test in more detail. JRS.]

[After further discussion with Dario, I believe there's a problem with the DDS Daughter board he assembled, so I've sent him a known good assembled board and will examine the one he built. This has reinforced my leanings towards making the DDS Daughter board a "wired and tested" assembly only. JRS]

[ I received the inoperable DDS Daughter board on 26 July 2006 and found that the 30 MHz oscillator module (the black rectangular "blob" seen at the  top of the DDS Daughter board in the photo) was defective. I replaced it and Dario's DDS Daughter board is on its way back to him. JRS] 

Day 4:  1.5 Hours (Total time: 8.0 hours)

Dario reported no problems with Stage 4.

[However, when I saw his photograph (at the right), it was apparent that U105 was installed 180 degrees rotated.

Dario was able to remove U105 and re-install it with the correct orientation, but U105 was destroyed, either by applying power with it in the wrong orientation, or by the heat of de-soldering.

I've sent him a replacement Gali-74 by overnight mail, along with a known good assembled DDS Daughter board.

He's continuing to install the remaining stages.

These mistakes are easy enough to make with a bare board, and the production boards with solder mask and silk screening should reduce this type of error to nil. JRS]

[Here's the specific error in inverting U105. The correct orientation has the tab towards the bottom of the photo, not the top. JRS]

Day 5:  1.0 Hours (Total time: 9.0 hours)

Dario has no specific comments in assembling Stage 5.

[Note Stage 5 also has a Gali-74 broadband amplifier and surface mount parts. The surface mount parts are spread out and installation should not be difficult.

I've made several changes to the Assembly Manual based on his comments. JRS]

Day 6: 0.8 Hours (Total Time 9.8 hours)

Dario finished Stage 6 in 50 minutes, with no specific problems. Stage 6 is the crystal filter section.

Day 7: 1.0 Hours (Total Time 10.8 hours)

Dario finished Stage 7, the log amplifier, buffer amplifier and ADC voltage reference. The build is now complete, except for replacing the Gali-74 amplifier (see Day 4) and installing the substitute DDS Daughter module. (See Day 3)

PCB assembly time 0.8 hours and cabinet assembly 0.2 hours, totaling 1.0 hours. The only assembly issue related to omitted assembly instructions and orientation of U502, the ADC voltage reference regulator.

[I've corrected the missing U502 information.

The remaining tasks are to calibrate the ADC reference and the time base. These should not  take more than 5 minutes each. I also told Dario that he is bolder than I am, as I recommend the shield not be installed until after the completed PCB has been verified as working since it makes working around the log amp module more difficult. JRS ]

The completed PCB

[Nice work, and the quality of Dario's craftsmanship can be better seen if you click on the image to see a full size view. JRS]

Day 8 Replacement Parts Installed and Z91 Functioning

Dario reports that the parts arrived today (Sunday) via Post Office and he installed them. All seems to be working and Dario supplied the sample screen capture at the right.

[Dario also found a bug in the display software that was, fortunately, easy to correct, and he is now using the corrected software. The image shows signals only over about a 10 KHz span because he is using a Racal RA6790/GM receiver, which has its IF output port after the normal receive filters. JRS]

26 July 2006 - Operation with Elecraft K2

Just above the 40 meter band, receiver tuned to 7415 KHz, 200 KHz span. The signals are strong shortwave broadcast band stations.

Same as above screen capture, but 100 KHz span. The signals are strong shortwave broadcast band stations.