Some of the mor4e frustrating repair projects come from assuming that a new part is good because it is new. Out of tolerance capacitors and wrong marked resistors can lead to an extended diagnostic chase, that is certain. Mismarked transistors also, particularly when a PNP is found marked as an NPN. So for some things that need to be correct the first time they are switched on, validating the "ggodness" of every part winds up being a real time saver.
It might be heat from the soldering iron. I remember those ceramic terminal strips, and several components might be soldered into those silver-plated notches. It might require a lot of heat to overcome that much thermal mass to melt the solder and replace the diode.
BrainiacV -- from these comments, I get the feeling that this problem is not uncommon at all. I guess troubleshooting should come without any assumptions, including the assumption that the parts -- even new ones -- are not faulty.
Chiming in not as an expert but as an enthusiast of the classic 500-series Tek 'scopes: The Type 547 isn't a storage 'scope, perhaps the author meant the Type 564? I've owned both and still have the 564, which still stores a trace! Not bad for a 50-year-old instrument.
You make a great point, Eric - especially for a large company like Tektronix that has the resources to fully address these types of issues - they probably have a large stock of those diodes and they certainly have a QC department that can handle the testing. It may be faulty parts coming from a certain supplier and taking the time to find the root cause could possibly prevent not only QC failures at the factory but operating failures in the field, depending on the mortality rate of the part in question. In the case of the scopes - it is very frustrating when using a piece of test equipment and having it fail while testing or troubleshooting.
Another variable to consider when seeking a root cause, Tekochip, is the source of your parts. I was building a microcontroller based wind rose for a project in college and I had an LED display that would light 4 different colors of LEDs to indicate 4 different ranges of wind speed, with the LEDS trailing within four circles to indicate wind direction. So...I had four circles of LEDs, each a different color. The smallest circle was yellow with a 0-5 mph range, the next circle was green...hopefully you get the idea. the outer circle was red and indicating over 20 mph. I etched my own circuit board, carefully drilled the holes - it was an intensive labor of love. I went to the local electronics store that all of us students bought our supplies at. After painstakingly soldering in all of my LEDs and beginning running tests - there was a huge variance in the color of my red LEDs - some showed up a very orangy-red which was very distracting to the balance of the display. I desoldered the ones that were orange and returned to the electronics store which often bought "surplus" parts. This time I brought a battery and tested the LEDs right at the shelf until I found enough red LEDs that were truly red, to complete my project. This taught me that when you buy at a store that carries surplus or nonstandard stock, to be especially careful to test the parts you are using to make sure they are operating within spec (or in the case of the red LED, what you expect).
Agreed, this is the type of fake I've seen. Maybe the die bonds are weak or the temp. specs. don't live up to the real part, but sitting on the bench in the lab they pass all the tests and make the purchasing guys happy.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.