SOME OF THE POD WORLD IN PICS                          THIS IS PODWORLDINPICS3                     IF INTERESTED IN EXPLANATIONS, ASK US. IF NOT, WHAT CAN WE SAY! AFTER ALL,               MOONMAN TOOK THE PICS, WROTE THE PROCEDURES, AND TRAINED THE PEOPLE  AS A DESIGN, PROCESS, MANUFACTURING, QUALITY, RELIABILITY, AND LEAN MFG. ENGINEER AT OVER 100 COMPANIES IN THE PAST 25 YEARS. THIS PART OF THE POD SITE IS DESIGNED FOR INTERACTION. THOSE WISHING TO CONTRIBUTE ANYTHING OF VALUE, AS PICS OR TEXT, PLEASE DO SO AND CREDIT WILL BE GIVEN. ALL THE FOLLOWING PHOTO DOCUMENTS ARE AVAILABLE ON THE POD FTP SITE AND ARE FREE.      WE ALL NEED IMPROVEMENT IN OUR WORK AND LIVES. WORKING CONCURRENTLY, WE CAN DO IT BETTER.      PODWORLDINPICS3 CONTENTS: CLICK ON LINKS   MOISTURE SENSITIVITY & SEALING LIQUID SOLDER MASKING PANELIZATION AND DEPANELIZING QUALITY - DEFECT PREVENTION AND DETECTION MECHANICAL & AND ELECTROMECHANICAL ENGINEERING AND DESIGNS ESD ENLARGED IMAGES WITH EXPLANATIONS

	MOISTURE SENSITIVITY & SEALING
	LIQUID SOLDER MASKING
	I have only 3 images for this procedure. You need to download the entire document from the POD free download site. This process saved considerable time and money compared with selective wave solder palets.
	PANELIZATION AND DE-PANELIZING
	There is little mechanical shock involved in the routing process when well managed. This is not true with "cookie cutting." Don't forget, PCB fab panelization almost always is different than that for assembly. THAT'S ALL I HAVE TO SAY ABOUT THAT
		There's a whole lot more in the free download section on the main page - about this process and many others.
	QUALITY - DEFECT PREVENTION AND DETECTION
CAUSE	EFFECT AS DEFECT	EFFECT AS NO DEFECT
	VERY IMPORTANT IPC STUFF STARTING WITH 3 HOLE TEST REQUIREMENTS BECAUSE X-SECTIONS ARE THE ONLY WAY TO SEE THE MOST IMPORTANT BOARD FEATURES, DEFECTS & OTHER INTERNAL ELEMENTS
	YOU'VE NOW SEEN OURS SO LET'S ALL SEE YOURS.
	MECHANICAL AND ELECTRO- MECHANICAL ENGINEERING AND PACKAGING
	What's good or bad about the designs?
Biggest problem here was fixing the IMC filter to make it solderable to the pads in the next photo.	We had to get IMC to mount their filter on a PCB substrate with castellations so we could solder it to the pad shown.	Filter installed with a LCCC type designed and assembled part but, as the CTE was the same for board and part, no problems.
No way to get heat from inside to out and even the fins were not properly designed.	You can observe no neck-downs designed for constantly cracking ceramic chips.	We designed neck-downs for this design iteration but company would not recalculate for RF requirements even when what we did worked physically.
POD presented another design that we made successfully at another company. The other company still wouldn't listen.	POD presented another design that we made successfully at another company. The other company still wouldn't listen.	Another non-listening, ignorant, company sleeps with the fishes.
	POD did it right by getting involved at the earliest design phase. A complete avionics package with LCD display and cold box.
	ESD BY IPC & POD

ENLARGED IMAGES WITH SOME EXPLANATIONS First image, top left, shows optical x-section of  42 layer PCB with DIMM modules soldered in place. Second image shows titanium nozzle I machined to use with a very hot air machine I developed for removing DIMM modules without damaging the $75,000 MLB Third image, on left below, shows similar MLB  with 1189 "pin" CCGA's mounted on board's surface. Last image, below right, shows section of relatively simple MLB using "pin in paste/intrusive reflow" soldering technology First image, below right, shows one of my favorite little solder/desolder machines as the Air-Vac SRM4. It does about everything I need often in place of the venerable Wenesco. At this time, we're adding a little liquid flux to facilitate the next step. The next step is to heat the board's bottom side then apply low pressure air, from a retractable nozzle, to the surface where the holes need cleaning of solder in them. This is done to ensure holes the specified diameter for insertion of the replacement connector and subsequent re-soldering via the machine's solder pot. The second image, on right, shows another high layer MLB with CCGA's mounted all over it. It took over 6 months to replace all parts with later revision types using an SRT-1100 repair machine. The third figure, lower left, shows the 42 layer PCB again. This time it is mounted on my custom built repair machine with which we replaced all the DIMM's with newer revision parts. The last image, bottom right, shows some of the components I fabricated in Celestica's 1st class machine and fab shop - lots of fun! Ah, my machine when "completed" showing a probe held over the titanium nozzle, with hot air being forced from four Liesters, not yet shown providing a very high reading of 918 degrees as shown on the digital thermometer in figure 2 above right. Image 3 shows my new machine's humble beginnings. I needed a pre-heater to bring the board temp up to soldering range so I went to K-Mart and bought a "blue light" special as a $19.95 electric pancake griddle capable of reaching a sustained temperature of 450 degrees. Also, I had created my version of the SRM4's hole blower cleaning capability resplendent with their rubber boot to evacuate solder from the holes to be re-soldered with a new connector. Under the "griddle" is visible a single Liester "hair dryer." The last image, below right, shows the same board with a CCGA attached that was not affected by the de-soldering/soldering repair operations. As a process development and advanced manufacturing engineering consultant, I was allowed many creative tasks. The first was assigned as removal and replacement of a micro-pax connector with pin dimensions just a bit larger than the insertion hole diameter. No one had been able to do this job going on two years and the project hinged on this process capability's success. With the pin alignment gage, first figure, the SRM4, some seriously steady hands, patience, and the specified soldering profile, the very small connector was replaced, figure 2 right, effectively as shown in a later x-section. Figure 3, lower left, shows the SRT1100 placing/re-placing a CCGA. The last figure, below right, shows another very interesting connector as a BGA style on .8mm centers. With the pick and placement "cap" on, it was easily placed and later soldered.   The first two figures show a Cray 50 layer test board. It is 17" x 21" and cost lots of money (about $100,000 each for the protos). Cray requires a manufacturing plan, starting at the design level, addressing rework, repair, and modification process success. The large termination areas are for 3200 I/O MCM's. The smaller areas are for the aforementioned 1189 I/O CCGA/s. There's lots more to explain but I'll just say, for now, the components no longer were placed in vats of coolant. The coolant was piped to and sprayed on them via some interesting plumbing consisting of pumps, conduits, manifolds, and housings for the nozzles applying the mist. This is a series of pics showing the rework, repair, and/or modification process. The lady with the nice hands made a nervous mistake, in my presence, and applied solder paste using the wrong micro-stencil. She then needed to remove the paste from the mis-printed CCGA site and start the tedious process over eventually allowing the subsequent placement, into the paste, and soldering operations yielding perfect solder joints. The stencil fiducial recognition and alignment process, on the DEK printing machine CRT, clearly is shown in the first image, top left, followed by the other images showing stencil alignment on only the pads with no green (board) showing so perfect solder paste printing is assured. A series of images is shown taking us from squeegee selection through the solder paste printing process and, ultimately, through reflow and determination that acceptable solder joints are assured as specified. This is a very interesting and critical set of images with first, top left, showing a x-section of a non wetted J-Leaded device solder joint. The reason for this defect is shown in the second image, from top left, as an EDX photo showing 18:1 oxide to metal ratio lead contamination rendering solderability impossible. The other photos show solder termination areas on PCB's. Most show impossible situations not allowing any type solder wetting at their surfaces. Others show solder mask failure often rendering solderability impossible especially as indicated on the last photo - left to right bottom row. For those never having seen "black pad," the second row's third figure from the left shows this phenomonom. The pad really is black but the top lighting renders it otherwise though clearly showing no possibility of solder wetting or solder joint formation. More explanations, if needed are available from POD and be sure to read my article on alternative, to HASL, solder termination area coatings. You all have seen an x-ray image from a HP 5 DX laminographic machine. This particular  one shows a serious problem rendering the entire assembly unacceptable. Email us and we'll explain why and how to avoid this problem. This is my favorite set of images requiring a PCB to be "flat," or relatively free of bow and twist, as indicated in the text above the images. I designed and had fabricated many MLB's with unbalanced constructions that exceed the minimum requirements when properly engineered using DFM/CE. Ask us about this and many other difficult and demanding designs - especially for RF design types. What else can I say but this is my first blind/capped via in the 1980's. It needed some work but it worked and got better over time - don't you think? LET'S GO TO POD WORLD IN PICS 4