Like they say “there are many ways to skin a cat”, just like a test engineer, there are many ways to test a board in ICT, but to me, a “lazy” test engineer, who doesn’t like to work hard to fix bad programs or fixtures, I do it right the first time, short, sweet, solid and effictient. I’ve been with my company for over 12 years, I’ve been treated well and I love my job. My goal is to write a program, deploy a fixture with high yields at any CEMs, any where and never have to hear any problems so I can enjoy the nice weather in Silicon Valley.
Hp3070 is a great machine, I prepare the tests well, design the fixtures correctly with fixture vendors, my products will speak for themselves. In order to have all these things come to place, the design of the board has to be correct, sound, testable and manufacturable. It doesn’t matter what the design engineers believe in their designs, if they’re not testable, manufacturable, they’re bad.
It doesn’t matter if cad designers claim there’s no real estate to add testpoints, I can cluster test them to get higher yields if they give me what I need. I can write a program to have over 95% yields with pins/nets less than 60%. Yes, instead of giving me junk testpoints, give me the one that I can test, I’ll make it happen. I need to work closely with design engineers and cad designers.
HP3070 does have many features and codes, sometimes, when we don’t use a certain features or codes, we may forget. I documented some of the codes, tips or tricks as I encountered them for future references. If you’re in the same profession, you can read some of the ones I wrote. Just need to register to read them, it would be nice if you let me know who you are or share some tips with others with comments.
Thanks and hope the information is useful.