Here are random answers to email that may be of use (or amusement) to builders of the SA kit. They are random paragraphs in random order... General Construction: Some of the parts (especially the ceramic disk caps) when inserted in the holes "shave" some metal off their leads and may cause shorts to ground. I spent several hours tracking down problems related to this. On the sweep board there was a damaged trace in the feedback loop for the Miller integrator that led to no oscillation, probably an isolated problem. You will have to plan carefully on some modules since you can fit 3 SMA's in the space of two BNC's. The depth of the connector into the box is probably the same but you will have to have the holes offset toward the top of the box to clear the board. Remember to leave clearance for the lid lip which is part of the RF seal for the box. On some of the boards like the VCO/1st IF you may want to build the board, set it in the case, and play with a couple of connectors to figure out the clearances. In any case this project is a great way to learn how to use a drill press. I used 1/4" 4-40 standoffs to mount the boards. For each module I built I did the following: 1. Set blank board on back of box, mark for mounting holes. 2. Drill holes, install 4-40 standoffs. 3. Verify that board fits mounts, slash one side of board with marker and make alignment mark on box. 4. Populate board. 5. Set board into box. 6. Plan connector holes and clearances to parts on board. 7. Drill holes for RF connectors. 8. Solder RF connectors to board. 9. Make power and DC control connections thru slide on connectors, mark B+ with a red marker, the boards do not have reverse polarity protection, you might want to add 1N4002 or similar protection diodes to provide a short in case of a polarity reversal, then just use a current limiting supply for testing. Torroids and such: Remember to observe the phasing dots on the bifilar coil windings! Some of the torroids which are non-critical used in LC filter networks on the power feed to a MAR-11 MMIC may have conflicting winding instructions between the Kanga board manifest, the article text, and the parts callout near the figure in the article. 110 Mhz Bandpass Filter: The 110 Mhz bandpass filter used to be a 850 Mhz IF unit from a cell site transmitter that I bought as parts at a hamfest. I just chopped up a random non-square piece of stock, drilled and countersunk a hole in it and eyeballed the .15 from a rule and sliced the slot out with an exacto knife. If you look at how the leads of the caps are dressed you will see that the lead on one of them runs parallel for quite a way on one side since I needed more coupling. The 110 filter is a pain to align. At least on mine the cover has an effect on the coupling. So sliding the cover on and off during adjustment is a must. The goal is to put the two "camel humps" together and get a single peak with the same shape on both sides. It is said that it can be done with a return loss bridge, but a spectrum analyzer with tracking generator makes it much easier. I was able to get the insertion loss down to 5.5 db on my filter. 300 Khz Resolution Bandpass Filter: The 300 Khz resolution bandpass filter is fun. I found that I had to remove some windings on the torroids to get the correct values. You may want to consider buying the AADE LC Meter IIB from Almost All Digital Electronics if you don't have a good LC meter. http://www.aade.com/lcmeter.htm I spent a week playing with the filter before I bought the meter and solved my problems. The other option is to use a silver mica cap and a pickup loop with a grid dip meter to check the values. 70 Mhz Low Pass Filter: The 70 Mhz low-pass filter seems to work without tuning no matter how you mangle it (1.3 dB loss max across the 0-70 Mhz range, 66 dB rejection by 75-80 Mhz) Power Supply: The power supply is a major part of success or failure in this project. AC ripple or variance on the DC will play havok with the unit. AC ripple shows up as the peak "bouncing" on a displayed signal instead of being stable and only changing when the source amplitude is changed. I used a AC/DC electronics +/- 12V @3.4Amp supply. I just turned up the OVP and the voltages on the outputs. This derates the current capacity but the unit draws ~ .5A on +15V and very little on -15V. In order to get rid of HF signals leaking in thru the power line I found it necessary to filter the AC line inputs with 32 mHy inductors. This took SW BC stations like CHU? at 3.333 Mhz from -60 to < -85 dBm. Connectors: I used various connectors. I used BNC's on the filters with the exception of the 110 Mhz bandpass filter which is SMA. Most of the RF connections between the modules are SMA bulkhead connections (hamfest parts, new is cost prohibitive.). I used BNC's for most control voltages. I used one RCA connector for the log amp gain control. I will use a shielded "soldered all the way around" Motorola RCA connector for the connection to the gain control with a LC filter on the power at the front panel. Misc. I enclosed the sweep generator since there is an AM broadcast station 1/2 mile south of my location that runs serious power on 1660 kHz. I added 4.7 uH inductors to the power leads inside the boxes on all the modules to get it off the power lines. My friends who run the local radio shop are joking that it has only taken $80,000 worth of test equipment to align the spectrum analyzer.