These examples are NOT intended to show our capabilities, but instead our passion to not let anyone fail. We approach the unknown with a sense of confidence that we can always find a way to move forward in a positive direction. Most examples involved research to explore the work of the many creative people that we have learned so much from. A basic understanding of physics helps.
Researched and delivered a complete market survey of how facsimile equipment was designed, manufactured, marketed, and serviced in the US market for Nightingale & Associates and New Japan Monitors, for the Japanese Ministry of International Trade. After interviewing a former Exxon executive, I predicted the exit of Exxon Corporation from the US IT market, because of style of management issues.
This survey facilitated Japanese office equipment manufacturers in establishing a strong position in the US office equipment market, leading to a dominant share of the computer printer market.
A small desktop computer integrator was struggling to survive as the desktop market transitioned from low volume high margin to high volume lower margin. A larger partner company, Aquarius Systems, died. After researching the market and looking at the various niche markets, I transitioned the company into selling integrated CAD/CAM systems, which tend to be highly custom and therefore had higher profit margins. A further competitive advantage was achieved by sourcing an HP equivalent plotter at 40% less than HP's entry level model, but which still used readily available HP media.
I designed the enclosure for a passive backplane set top box for Jenair. Working with the sheetmetal fabricator, the prototype was competed in only two halves, reducing cost to below the clients target.
A $35,000.00 semiconductor test handler was failing acceptance at a defense contractor, Raytheon, because a design change was vectoring hot gas from the thermal soak rail into the tube holder. At that time, twenty-two inch long plastic tubes were used to transport the integrated circuits.
Based on earlier discussions with an architect on the flow dynamics of air curtains, a miniature air curtain was designed and installed, splitting the gas stream, such that it vectored to the sides of the tube holder, deflecting the hot gas away from the plastic tubes. The cost was $22.00 per machine for the new part.
Because of signal quality issues, MCT designed a controlled impediance contactor, for use in testing higher speed integrated circuits at AMD. The individual contact fingers broke off at the 120 degree bend, ruining the entire expensive assembly.
All of the stress concentrated on the single 120 degree sharp bend, resulting very early failures at this bend.
Replacement fingers were designed with two softer bends, allowing a more gentle flex. Replacement fingers were lap soldered over the broken ends of the original fingers, avoiding the need to replace the flexible circuit and connector pin assembly.
Because under certain conditions, an LN2 vaporizor would allow raw liquid to reach the power lead seals of an air heater, this resulted in LN2 leaking from the danaged lead seals, which failed after several cycles of thermal contraction and expansion. This created a serious potential safety hazard, spilling raw liquid nitrogen out of the front of the instrument.
A conical TFE Teflon seal, which looked like a two hole laboratory test tube stopper was put into a brass tube fitting, replacing the ferrel. This became the new compression lead seal.
A prototype table top diffusion chamber was custom designed for Dynamit Nobel to for their now closed Sunnyvale wafer sales office. The thin glass chamber failed. As I had sold them the infrared lamps, I redesigned a much thicker quartz chamber and added a stainless steel safety cage, saving the project.
Stanford Department of Chemistry designed a custom sputtering chamber that used quartz lamps to fabricate custom sensors. The infrared lamps that I was representing would become coated during the process. I advised them to put the lamps inside of quartz tubes that could be removed and cleaned, and made maintaining the vaccum seal easier, as well.
Decades ago, when integrating a semiconductor robotic device handler with semiconductor tester, the tester was failing acceptance at a defense contractor, Litton Data Systems, because of calibration issues.
The test head was mounted in a desk type cabinet originally designed for manual socket testing, which provided very poor air flow. Heat from the power supply was cooking the test head.
Once the test head was mounted on the adjustable positioner that I designed, the power supply was relocated to the base of the positioner, adding weight to the positioner for additional stability. Eight four inch box fans were added to provide cross flow cooling, resulting in the tester passing and holding calibration with no further problems.
AMD was attempting to test higher speed memory devices, and was having a major issue with ground bounce. We worked with the Test Engineer in determining that the ground bounce could be reduced, if we connected a capacitor across the power and ground pins, of the device under test, during the test cycle.
A chip capacitor with the lowest possible inductance was sourced, an FR4 circuit board was fabricated with the same flexible fingers used in the contactor, and installed into slits in the circuit board. The test site floor plate was thinned to allow for Teflon insulators and the circuit board, with contact fingers and the capacitor.
When the device to be tested enterred the test site,the contact fingers closed on the device leads and onto the fingers on our circuit board.
A small KOH type electrolyzer was installed on several vehicles, feeding Brown's gas into the air filter housing. It was powered off the Run signal through a relay. The gas was bubbled through a bubbler preventing any migration of KOH from the electrolyzer.
As the unit sucked back water from the bubbler, because of thermal contraction of the gas, when not running, there was no need to add water to the electrolyzer. The user just tops off the water in the bubbler.
On ICE engines with fuel mixtures controlled by oxygen sensors, there was a noticable improvement in power and efficiency. As predicted, the 2H2 - O2 gas mixture sped up the flame front, resulting in more thorough combustion over a wider range of engine speeds.
Our market research determined that the technology had been previously published, and we determined that if we were to manufacture these units that we would soon see copies made offshore at lower prices.
We continue to be puzzled that Hythane did not catch on, where they added H2 directly to natural gas powering transit buses, improving efficiency and reducing maintenance costs at Sunline Transit. GE is finally mixing hydrogen wth natural gas in power plant turbines.
After building a complete library of vendor design information, catalogs, it became apparent that an indexing system was needed. The four numeric character SIC system from the Department of Commerce was grossly inadequate, and thereofre we expanded it from four to ten numeric characters.
The system provided an ability to index a product under multiple categories, which could then result in rapid vendor identification.
As this database was never migrated to the newer operating systems,it would require some work to bring it back to life.
The Fairchild Mask Shop had purchased a heated chromic acid etching tank, but found a 7.5 degree C vertical gradient in temperature. A magntic stir bar that came with the tank was unable to stir the thick solution. We fabricated a sparging system, bubbling nitrogen throught the bath, which resulted in no detectable variation.
We have advised an investor to not put money into a startup, who was planning to use wind turbines to power refrigerated shipping containers. In many cases, the containers are stacked!
Instead we suggested using a hydrogen fuel cell system currently used to power fork trucks.
In testing 80286 chips in side brazzed packages at Intel, there was impact damage as the chips slid down into accumulator tracks. The first part sat 24" from the top of the rail with gravity accelerating the second and following chips, resulting in extreme impact.
Stainless steel washers were spaced every 6 inches down each of the tracks, such that the chip would slightly lift the washer, as it passed under, absorbing just enough inertial energy to stop the impact caused chipping.
When IBM bagen to require the integrated circuits be packaged in PLCC packages the "J" bend became an issue.
IBM reported that package leadframe planarity must be within 0.004 of an inch, or reflowed solder joints would lack adequate strength.
Consulting with a specialist in material science, it was determined that the solder dip process was crossing the eutectic point of the copper leadframe, resulting in the sharp bend coming out of the package unfolding slightly, more in one grain direction than the other. We advised the packaging engineers to switch from solder dip to solder plating.
The MCT 4600 PLCC handler's contactor assembly was expensive and fragile. The radius on the contactor fingers tended to impress radial dents in the early lead frame solder layer, where the floor plate would fail to eject the part, when the contactor plunger retracted. The fingers were redesigned,using berylium nickel,to both scrub clean and to aid in part ejection.
The contact finger pocket combs, with very narrow slots that located the contact fingers, were made of Vespel.
When the drill rods came unblued from the floor plae and several floor plates broke. We redesigned the floor plate out of tool steel. The pocket comb was remade out using G-10, much less expensive than Vespel and much stronger, as it is a fiber glass reinforced composite.
Advised founding team of medical patient support group, to pivot from plans to sell directly to patients and family members to more realistic plans to sell to the large healthcare providers, such as Kaiser. Anything that improves patients peace of mind results in a placebo like effect, which improves the efficacy of other therapies, thereby reducing patient treatment costs.
We pointed out to the founders that once patients were charged extra to have televison in their rooms, but now every room usually has one. Their support group fee was estimated at $10.00 per patient per month, which we pointed out is less than the TV cost.