Wednesday, December 5, 2007

Talk with Westpak on Damage Boundary Curve Shock Testing

Goal: The goal of this conversation was twofold:
(1) To understand why the current shock test is considered by Westpak to be meaningless.
(2) To figure out a meaningful way to shock test the meter to destruction.

Summary of Conversation:
(1) Went over Damage Boundary Curve test methodology covered by ASTM D3332 Methods A and B.
(2) Went over applicability of said test for current products
(3) What is entailed in our current test, and why it may or may not make sense.
(4) Got an idea for how number of G’s relate to real life.

Conversation in Detail:
(1) Damage Boundary Curve Methodology.
a. Comparison of Method A vs. B (see table below)



b. Curve generation
i. X-axis is formed by Method B
ii. Y-axis formed by method A
c. Purpose of Damage Boundary Curve Method: knowing what the characteristics of the bare product is, design the packaging to bring the area within the “Damage” region within the “Damage Free” region. See picture below:


d. Duration of test: 2-3 days

(2) Applicability of above test on current product
a. Most likely cannot perform Method A test. We know the bare product can withstand drop test, which is equivalent to around 300 G. The machine can only do a maximum of 120 G for Method A.
b. Best proposal is to increase height on shock machine on Method B to up the max G’s to see what G’s the device fails.

(3) What is entailed in our current shock test, and why it may or may not make sense.
a. We do 300 pulses in each of the 6 directions
b. Each pulse is a half sin pulse performed at max peak of 30G, and 100G, at 6 ms.
c. Each pulse at 30G is equivalent to smacking the table with your hand.
d. This may or may not make sense as a lifecycle test.
(4) Idea of G in real life. A hand held device dropped from 1 m will experience 300-400 G.

Wednesday, November 28, 2007

Manufacturing Tolerances

This document is a repository of different manufacturing tolerances. Included are tolerances listed for:

A. Injection molding
B. Sheetmetal
C. PCB Manufacture
D. Rubber Compression Molding
E. LCD fixed icon displays
F. Padprinting/silkscreening
G. Foam gaskets

I hope to add: IMD (in mold decorating), machining tolerances, magnesium-alloy and more...

http://spreadsheets.google.com/pub?key=pC0_188K00LjPE_T9BcdEHQ

Monday, November 26, 2007

PCB Construction


The following picture shows how a PCB is constructed. The important insights are:

(a) Soldermask provides openings that are bigger than exposed copper so that the copper will never be covered.
(b) There is only one layer of FR4 on this two layer board.
(c) Through vias are created by drilling holes through the two layers of copper and filling the through via with copper. A through via connects one layer of the board with the next. There can also be blind vias (more expensive). Also vias are designed to rest on teardrop shaped copper pads. The reason for the teardrop is for greater tolerance for the drilled hole location.
(d) Soldermask masks off the circuits on copper layers.
(e) Gold is adhered to the copper with an intermediate layer of nickel.
(f) Both the copper circuitry and soldermask is created using imagery designed by the layout person.

Saturday, November 24, 2007