The following companies are among the growing number of PCB fabricators installing ECT 9090 Series Universal Grid Test Systems.

Continental Circuits, Phoenix, AZ, has purchased a fourth 9090 Series and third HyperSort" automatic unloader and sorter. Continental reports that the addition of the HyperSort has more than doubled the already substantial throughput of the 9090. Continental manufactures high-volume, very fine pitch printed circuit boards in a multi-facility plant in Phoenix, AZ.

Chem Lab, Houston, TX, has installed their first 9090 bare board test system. Chem Lab manufactures high-density, quick-turn PCBs. They cite high throughput and the convenience of ValuGrid" fixtures in helping them provide rapid testing of quality printed circuit boards.

IBM, Endicott, NY, recently installed their first 9090 Series bare board tester. The system was purchased to test high-density boards in a volume production environment. IBM Endicott manufactures very high density PCBs for its sister companies around the world.

Herco Technology, San Diego, CA , has ordered their first 9090 board tester and HyperSort automatic unloader and sorter. Herco reports that the ValuGrid fixture is a dramatic improvement in fixturing pointing accuracy and predicts a substantial reduction of fixturing costs. Herco Technology manufactures high volume circuit boards.

Lone Star Circuits, Rowlett, TX, has installed their first 9090 bare board system. Lone Star is a manufacturer of quick-turn, high-technology printed circuit boards.

Tech Circuits of Wallingford, CT, has purchased their first 9090 Series bare board tester. Gary Ferrari of Tech Circuits, an active member of the Institute of Printed Circuits (IPC), says he is looking forward to performing tests to the IPC-ET-652 specification, a standard feature of the 9090. He expressed confidence in building ValuGrid test fixtures from IPC-D-356 data format.

Sheldahl has installed two 9090 Series bare board testers at their Northfield, MN facility, Sheldahl selected the ECT system based on our innovative fixture designs and our testing solutions for high technology flexible circuits.

New TeleFixture" Service: Turnkey ValuGrid" Fixturing

A new fixturing service has been introduced to provide ECT customers with fast turn-around for ValuGrid fixture-build data or completely assembled, debugged fixtures. The program, called TeleFixture Service, was initiated following successful tests to verify program integrity.

To use TeleFixture Service, the customer simply provides IPC-D-356 board data via modern. ECT will use a proprietary software program to translate the data for building a 3.75" drop-pin ValuGrid fixture with advanced features, including fixtures for our DenseGrid" Adapter. The customer then has a choice of two options.

Option 1: ECT will provide to the customer, via modem, drilling and routing files with on-screen operator instructions, the fixture bill of materials, and the netlist program for the 9090 Series tester.

Option 2: Includes everything in Option 1, plus drilling, assembly and fixture prove-in against the netlist program.

ValuGrid fixtures are designed with speed pins for fast, accurate PCB placement, and gutter pins for precision mating with the HyperSort" Automatic Unloader/sorter. The fixture also features QwikStat" plate spacers.

Contact your ECT representative for detailed information on TeleFixture Service.

The Impact of Test Area Environment on Test Reliability

The voltage of this static electrical charge can be in excess of tens of thousands of volts --- voltage typically found at a power company substation, but with a minimal amount of potentially dangerous current.

This type of casual static charge can seriously effect electronic components. Digital electronic systems typically run on 5 to 15 volt DC. Most systems run at a current much lower than that found in a static charge. In high voltage electrical testing (100 - 25OV), very low currents are used to prevent damage to the product under test. Test system electronics, and Computers, can't handle a large static charge.

Electronics, in ECT 9090 Series testers, counter electrostatic discharge (ESD) with a proprietary, custom-designed IC switch from AT&T. It has an internal circuit to deflect ESD away from switching and measurement systems. But everything has its limits. Extremely large or continuous charges will weaken devices and eventually cause failure. The ECT switch design provides ECT users with the highest reliability available.

This does not mean that no static precautions should be taken, however. Static can be reduced by controlling humidity and temperature in your test area. Limiting static can save money and increase system up time. The test goal should be zero failures and 100% up time.


The Humidity Factor
Wet air is conductive. Since the test system is actually measuring conductivity, the test system will measure the conductivity of excessively moist air as it tests the circuit board, or performs self-diagnostics. In either case, the tester will indicate leakage. Leakage is the state of being neither totally conductive (shorted or connected) nor totally isolated (open). It's somewhere in between.

Running the test system in an area where humidity is above the specified threshold may result in a leakage failure message. Sometimes you can lower the isolation test threshold and proceed, but customer requirements may prevent this. Out-of-spec high humidity will not ordinarily cause permanent damage to the tester unless condensation is present.


In this case, the conductive water will short the electronics and can cause severe damage. This can happen when installing or moving a system, even in dry environments, if the system is cold and the room is warm. Gradually bringing the system to room temperature before applying power should avoid condensation problems. Reducing the test area humidity to the specified operating range will result in fewer problems.

The Temperature Factor
Semiconductors in electronic assemblies exhibit predictable characteristics based on temperature. Operation in the specified temperature range is the best way to ensure that test systems run smoothly.

Very high or very low temperatures affect the reliability of computer systems within the tester. High temperatures also increase leakage currents, reducing accuracy somewhat.

Summary
The test engineer must understand the need to maintain consistency in temperature and humidity in the test area. Dramatic seasonal changes may require dedicated environmental control systems. Air conditioning extracts moisture from the air and can cause an increase in ESD activity. A humidifier may be required.

Whatever it takes to maintain a consistent and controlled test area environment is well worth the cost when you consider the losses you can incur through excessive test failures or system downtime.

Fixturing's Secret Formula

High-end fixturing is a complicated business. But the foundations for understanding the nuances of testing ball grid arrays and ,010" to .016" ultra-fine pitch centers are found in a simple formula:

HSMIN = SD + PT(TanO)

Where HSmin is the minimum hold diameter, SD is the shaft diameter of the pin, PT is the plate thickness, and 0 is the angle of deflection. 0 can be determined by:

0= SIN^-1 (PD/PL)

Where PD is the pin deflection and PL is the pin length. Don't start digging out your high school trigonometry books yet. There is an approximation:

HSMIN = SD + PT * PD/PL

Minimum Hole Size = Shaft Diameter + (Plate Thickness x Pin Deflection/Pin Length).

With the serious math out of the way, we can draw some conclusions. Note that the angle of deflection affects the minimum hold size. This is especially important for those responsible for fixturing software. These packages allow for dynamic hold sizing -- the hole for a given pin diameter can be programmed to vary with the angle of deflection. It's important to take advantage of this feature.

The more a pin is displaced, the larger the hold needs to be to accommodate the pin without binding. If the largest hold diameter were used for all pins, a pin that was not displaced would be very loose in its hole, producing poor test results in both repeatability and false opens.

When testing ultra-fine pitch devices (.010 to .016), all tolerances must be held to near zero levels. Should the tolerance of the UUT and the tolerance of the test fixture exceed 1/2 of the minimum pad width, the test will not be reliable. In the case of a data-driven program, false opens may occur.

However, the worst case is in a self-learned program where PCB shorts may be undetected.

The ball grid array (BGA) poses a different problem. Pointing accuracy and tight hold sizing are not so important, since .050" centers are typical. What is important is the ability to meet the high density requirements of BGA testing. Pins may be displaced at the conservative limits of the fixture (.600" for ValuGrid). With this large displacement, dynamic hold sizing may have to be adjusted. This is particularly important for fixturing software packages that adjust hold size based on a percentage of the maximum pin deflection

There are many considerations when fixturing for high-technology board designs. These formulas will simplify the fixture design process.