Electronic Design

Bob's Mailbox

Dear Mr. Pease:

We are a garage operation startup that has just received a federal SBIR (Small Business Innovation Research) phase I grant to develop and evaluate several nontraditional tactile imaging sensors. Our literature search turned up two old magazine articles authored by Dudley B. Clark, Clark Electronic Labs, Palo Alto, Calif. In them, Mr. Clark talks about his proprietary piezoresistive pressure sensors that have a world of desirable features, e.g. large dynamic range, good resistance to overpressure damage, little hysteresis, low (and linear) thermal drift, little aging drift, good sensitivity and repeatability, lots of output ranging options, etc.

The implication is that the materials employed change their fundamental resistivity when loaded (stressed). If true, this is quite different from those sensing technologies which depend on changes in internal connections, path length, and/or cross-section. Such sensors seem almost too good to be true.

Clark said that his sensors were made of intermetallic "plastic" materials composed of zirconium tetrachloride and rare earths (presumably scandium, yttrium, and the lanthanide series elements or their compounds). I would like to include evaluation of Clark's technology as part of our SBIR if more information can be obtained from Mr. Clark, his associates, his firm, its successor, or any other source. Unfortunately, time is short.

So far, I cannot find hide nor hair of D.B. Clark or his firm. As you might expect, the publishers of the magazines have no records dating back to the '50s and '60s.

If you or any of your readers can shed any light on this subject, I can be reached at (612) 881-2824.


The Oxboro Technical Group Inc.

Bloomington, Minn.

Does anybody know the whereabouts of Mr. Clark? Or where his technology may reside?----RAP

Dear Bob:

I was pleased to see your column in the May 2 issue about the oversimplification that is so often seen in the JIT process. Let me tell you my experience.

I started with a major instrument manufacturer in Beaverton, Ore., in 1977. During that time, it was difficult to get parts. They either couldn't get the inventory they needed, or couldn't efficiently distribute the inventory to where it was needed. So, in the early '80s, they built an "automated warehouse" with a vast capacity for components in computer-picked storage bins. They got the process up and running, and eventually perfected their component delivery system. Then they got the JIT religion. A few years later, they held an auction to try to sell off a large number of assets (reducing staff from about 25K to 8K left them with a lot of extra stuff). As I recall, nobody even put in a bid for the storage system.

It is possible that the high-level people understand the complex factors involved in optimizing inventory. But the low-level people who actually buy the parts do not. A plastics manufacturer was saying that he sells them a small part, a year's supply of which would fit in the palm of your hand and cost 5 cents each. Instead they insist on buying these parts a week's supply at a time at 50 cents each. This JIT is not costing them "a little bit more," but "A LOT MORE," at least on some items. It has been suggested they buy through distributors, who would hold stock for them at little additional cost. But that is contrary to the JIT religion, so they won't do it.

The problem with the JIT method is that it compromises profit margins forever, or at least from the moment JIT is implemented until it is abandoned. Every year, those higher JIT materials costs cut into the gross margin. Tightening inventory through JIT releases cash from the balance sheet only once, as a one-time event occurring at implementation of the JIT program. From then on, JIT impairs profitability every month of every year. Only a very shortsighted management team could expect to prefer a one-time cash-flow boost over long-term income statement profits.

Since then, I've started a small company, and our inventory decisions are not based on religious zeal. We consider our usage rate, how long it will take to get additional stock, physical size of the part, quantity pricing discounts, availability from alternate sources, item cost and margin, and whether this is a critical part of a larger system versus a minor accessory.

If sales shoot up, we can build up and ship from stock until replacement materials arrive. If sales are slow, well, we have some money tied up in inventory. But I would rather we make continuing profits from sales of product (which leads to repeat sales, larger market share, and a more stable income stream), than from manipulating inventory to extract cash once (which leads to inability to meet demand, and loss of sales).

When we introduced our first product nationally, we were overwhelmed with demand. On many parts, we found that we had bought up all of the available inventory in the whole country! It is disheartening to call the largest North American distributors and find that their total nationwide stock of a critical part is fifty pieces. Since then, we have set up a relationship with a couple of smaller distributors to carry extra stock of critical components, so that even during a surge in demand we will not run out of stock. They get most of our semiconductor business.

Bob, I think that part of our frustration is that as engineers, we see the world as a bunch of tinker-toys that can be put together in various ways to achieve the results we want. Buyers can be clerical types who are used to strictly adhering to policies and rules, leaving little room for personal decision making. And senior managers are sometimes so driven to achieve quarterly goals that they make decisions which sacrifice long-term success.


Chief Engineer

Fascinating Electronics Inc.

Beaverton, Ore.

I agree, statements that JIT saves money are SUSPECT. Seems to be a case of "proof by repeated assertion." If the world were a LINEAR place, the apples-and-oranges types would be heroes. But the world is NOT a linear place. Life ain't fair. (*!&*@% the bean counters.)----RAP

Dear Mr. Pease:

Your column in the May 30 issue, "What's All This Teledeltos Stuff, Anyhow?," was interesting. You mentioned having difficulty finding conductive paint.

A recent problem with my telephone led me to find a conductive paint pen that works very well. The flexible circuit board that is used behind the keypad developed some open circuits. The plastic obviously doesn't take to soldering without melting, so the only way to repair the traces is with a conductive paint. A few weeks before this incident, I had noticed a conductive paint pen available at Halted Specialties, 3500 Ryder St., Santa Clara, Calif. This pen makes it very easy to apply the paint just where you need it.

The pen is called the "Circuit Works" conductive pen model 2200 and costs about $11.00. The company that manufactures the pen is Planned Products, located at 303 Potrero St., Suite 53, Santa Cruz, Calif.


Sr. Controls Engineer

Matheson Electronic Products Group

San Jose, Calif.

Exactly! That's another good way to repair conductors that don't like to be soldered----and to contact onto Teledeltos paper.----RAP

All for now. / Comments invited!  RAP / Robert A. Pease / Engineer


Mail Stop D2597A

National Semiconductor

P.O. Box 58090

Santa Clara, CA 95052-8090


Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.