When I was an apps engineer for Supertex, I focused on flat panel displays and power supplies. The companies I supported were essentially starting up in their business, and looking for the best display drivers. These companies worked with displays that required high voltage drivers integrated with low voltage logic.
They were big customers to support, but I enjoyed working with even the small customers; the ones that brought the interesting problems that make engineering fun.
There were two problems that really showed the creative and investigative approach to applications:
The first situation was with a customer that was working with over a dozen EL lamps for a new system design. At the time, the company assumed it would need more than 12 EL lamp drivers — one for each lamp. My focus was displays and not lamps, yet I came up with a solution that was display related rather than lamp related.
With the display approach, I showed the customer a two-chip solution instead of many. Though it did not provide the larger business volume, it did make the customer very happy that we were able to come up with a solution that would save it money.
The second problem started with a customer returning a part that had failed. A semiconductor company has the ability to take a failed chip and identify the failure mode — or so I thought.
The product engineer stripped the plastic and visually inspected the IC. The report was “EOS failure” (Electrical Over-Stress) — failure due to over-voltage or over-current. To me, this was meaningless. I had to explain to the customer the meaning of this in the best way I could. The customer continued to have intermittent failures and sent back the failed ICs, which we kept reporting as “EOS.”
I started to get concerned and questioned the customer about the application, voltage, and everything else that I could think of to give me a picture of why the problem was occurring. Now, essentially, the part was a 450V part in a 400V application. Because of my development background, I started to question the voltage being applied.
On a whim, I asked our product engineer to test the part beyond the 450V spec. The basis was to see where the failure of the parts actually occurred. The product engineer found that the failure rate increased dramatically above 460V — to me, this was not enough margin. I pushed to have the part returned to engineering, and they found an error in the design. They quickly corrected it and the company flushed out the old parts with the new. The customer that was frequently reporting problems never came back with any additional failures.
In both of these cases, the customer purchased at low volumes, but the ROI, in my opinion, was well worth the time spent. In the first case, the customer was very satisfied that it left with a cost-reduced solution for its product. I am sure the engineer would look to our company for future solutions. In the second case, spending the extra time on the customer's problem solved a bigger problem that would have probably been bad for our business in the future.
Now that I'm back on the design side of things and out of the semiconductor industry, I recently had a problem with an analog switch that was not following the data sheet. It was operating under all the right conditions, yet under a certain condition, would not switch. I contacted applications numerous times and offered the apps engineer a board that continuously exhibited the problem, but they never said “send it.”
I posted on their discussion board and had people look at the report, but no solutions. To this day, a year later and after several attempts to get the problem in view, no one at the company has taken this problem on. Now, I'm not a large-volume customer, but I can say that when I look for an analog switch, I know what company I would not select first. I am now doing some contract work for potential large volume designs. What is the potential ROI loss for this company with the bad reputation?
Note that I am not naming companies only out of respect — they do have other products of high value.
Please share your thoughts and experiences.