[NTLK] SER-001 (The real story...)

Everyone,

Although I have officially left this arena, I decided to clear up some misconceptions regarding the SER-001 board once and for all.

The board has gone through 3 revisons since its' inception in 2001.

The first revision was the odd shaped board. It was shaped this way to avoid any interference with the peg in the case bottom. This peg was meant to engage in the modem connector and provide additional support.

I realised that a simpler shape would allow easier panellization (manufacturing) and so the rectangular 2nd revision was born.
In order to avoid the peg a slot was milled in the board at that position.

The 3rd revision has provision for a different IC. If the Linear Tech part should become hard to source then I could use an alternative Texas Instruments chip. It has a different pinout and footprint.

As far as the bulging/cracking issue that has become a topic of disscussion recently, I offer the following:

Despite what has been suggested I DID take time to examine the problems of interference. The problem I faced when trying to fit things together was due to the fact that I had to make it fit an existing case with existing components. I did not have the luxury of custom components so the design is a compromise from the start.
That is the reason the minidin connector was modified with a milled step. The reason I did not mill the whole top was twofold: If you look at where it fits you will see that the interference is limited to the front of the connector only and milling the complete top surface would render the connector very weak.

The black spacer has nothing to do with bulging and everything to do with ensuring a secure fit. It must be fitted.
The minidin is offset from the monting screws. If you use only one screw then, over time, you will break the connection internally as the board will tend to rotate around the mounting screw. Two screws will securly anchor the board because they are on both sides of the minidin connector.

Actually, the real culprit is one that was meant to ensure that the board connector is not misaligned - the end shrouds. In fact, it is one of these shrouds that is the problem. The bottommost shroud hits a capacitor on the main board due to the fact that it is slightly too thick. This means that the connector cannot fully engage and there is a potential for bulging. The earlier verions didn't have these shrouds so they don't exhibit the same issue.

This connector was chosen as a substitute for the JAE connector that was not available at the time - close but not exact. If I had been able to get the JAE connector then this would not be an issue.

The solution therefore is to remove the bottom shroud thus eliminating the interference and removing the tendency to protrude upwards and cause case bulge. The shoud is easily removed by bending away from the connector and working back and forth until it breaks off.

At the end of the day, this was not a perfect solution but it enabled lots of you to not have to worry about carrying a dongle around and has saved many interconnect port disasters.

Finally, if you want to be critical of me or my design - that's fine, I realise it could have been done better. I just wanted you all to be in possession of the facts.

Regards,

PCBman

-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/