You may recall my disdain for the designers of the XM interconnection scheme on my Trailblazer. Here is a brief summary of the effort to repair the actual antenna in the event you find this sort of thing interesting. If not, carry on
The failed antenna reads around 15 ohms across it’s output. A good one reads about 7Mohm.
First pic is with the top cover removed
Second pic is with the antenna guard removed
Next, the board assembly removed, coax had to be removed to separate from housing and facilitate debug of circuit board. I was surprised to see so much circuitry without any conformal coating or other protection.
Now, a back of the napkin, first pass schematic to begin reverse engineering the circuit. All I want at this point is to find enough to begin the process of isolating various circuit chunks (the divide and conquer approach). No desire to reverse engineer more than I have to.
Removing resistors in line with various active components helps to isolate them. These resistors are 0102 SMT components, the size of pepper specks…
Second circuit isolated and the culprit is found!
The package marking V67 leads to the actual device, a NE38018. Once removed, the gate to drain is found to be shorted. This is a 13GHz HJ FET used in GPS and other high frequency designs. Unfortunately, it is also obsolete! A cross reference was found pretty quick (NE 3508M04) and readily available from Digikey/Mouser etc for $1.92 qty 1.
Parts on order, will be here tomorrow morning. Should be back and running shortly after that…
ebay has some NE38018s available, although the photo they display is incorrect.
Recommended replacement parts are frequently NOT exact replacements, but will perform the same function, perhaps with different bias. so you have to compare data sheets carefully.
Already did that before I considered ordering the parts. The replacement supercedes the original part and is more than suitable based on datasheet comparison.
Hey Bill, been meaning to mention- I ran across some postings on electrical topics in another forum by some guy named Bill Russell too bad this isn’t Electronics Talk, we might have some interesting discussions…
At the risk of dating myself, when I started out, we made a lot of eyelet boards for point to point wiring on prototypes. So I’ve lived through the transitions to surface mount and beyond (e.g. chip on board etc). It’s the same thing, only smaller
We still use through hole for high power (thermal dissipation), high voltage (creepage and clearance) designs. High voltage on the order of 180kV and up to 750kV.
The antenna is repaired and functioning again. It cost less than $2 and about 20 minutes of my time. A new antenna is about $200 (parts only) at the dealer, one can be had on EBay for around $40.
I cannot imagine spending $200 on something like this. FIY has saved me countless $$ over the years on everything from electronics to home and car repairs. And sometimes, it’s even fun to do
The discussion about the high freq FET reminds me years ago, before cable tv, they had a sort of 3 GHz over-the-air cable tv you could subscribe to. To receive the signal surreptitiously you needed a 30 inch threaded rod and some spacers and washers and coffee can as the antenna, and a strip-line circuit board, to which you had to solder one HP GAS-FET smd transistor. I spent a lot of time figuring out to do solder that damn transistor. It turned out after many failed attempts I needed to use a special type of silver solder to get it to work correctly.
George, I believe you’re describing a Yagi type antenna. I made one myself to receive Selec TV back in the day. I used an extruded, plated aluminum can but many people had started resorting to coffee cans to save money as they did not significantly impact the quality of the signal if you were close enough to the transmitter.
My best recollection of silver solder is with the old Tek scopes. They had some inside to be used for repairs. It was required to prevent damage to the terminal strips.
My only experience with silver solder was with my dad and I welding a patch over rust on my 59 Pontiac. It was like a gas welding rod though. You had to get it very hot to melt it even though it was just wire, but then held like a gas weld would. Maybe there are different types but can’t imagine doing any electrical soldering with it.
One reason could have been some level of oxidation on the pins. Without a highly active flux core solder, it would be tough to get flow. Silver solder requires greater heat = burns off oxidation and allows for better adhesion. I don’t recall anything special about the gallium arsenide HJ FETs back then that had special plating on the pins…I do recall having to use a very corrosive solder flux on the boards I built and diligently clean it off afterwards. Of course, most fluxes were way more corrosive back then…
Yeah, Bing silver solder is an excellent mechanical bond but does require a lot more heat. Kinda like brazing versus welding eh? That’s cool you and your Dad did that kind of stuff together!
Why silver solder in Tek scopes?
Because its coefficient of thermal expansion better matches that of the ceramic connection strips.
I worked on many, many Tek instruments back in the day and saw the result of a previous technician using regular lead-tin solder: cracked ceramic.
I loved the aesthetics of those, and hand wired Zenith TV’s too.
And then there are those eccentric 3-dimensional “birds nest” breadboards Robert Pease would build:
too bad this isn’t Electronics Talk, we might have some interesting discussions… 
