Feedback

September 7, 2011
 by Paul McGowan

Funny how daily life can many times mirror the circuitry that runs our high-end systems.

There appears to be an increasing customer frustration level in all sectors of business towards errors in products and services, reported through customer feedback, and the company’s response time to fix those errors. Customer expectation levels for faster error correction, perhaps even elimination of errors, appear to be rising exponentially.

Here’s the interesting thing: all feedback systems require an error to occur before a correction can be applied. An audio amplifier first commits an error, then a correction is applied, and it waits for the next error to occur. So, short term, errors are made and tolerated.

Consider the significance of what that statement means; errors are not only acceptable and inevitable, they are required for the system to function. An example might be the speed control in your car, which is also a feedback loop. The car accelerates or decelerates above or below the desired speed first, the corrective action happens second. Run, make error, correct error, run.

So if errors are inevitable and acceptable in a system, what has changed to cause increased expectations for quick responses? I think people are getting comfortable with the idea of instant access to information on a 24/7 basis because of the internet. It’s fundamentally changing our expectations as a society.

If I have a question or need an answer, I jump on the net and Google or email till I get an answer. The process takes minutes (sometimes seconds) where even a couple of years ago, we were talking days. We’ve designed in a faster device and are having trouble compensating and keeping up with it.

In a circuit we would be required to shorten the loop and response time to meet the increased speed demands of the circuit. Failure to do so results in an unstable situation, possibly fatal.

I am not sure companies are any different than the circuit. We must shorten our loops and response time or risk instability in the marketplace.

Ain’t life challenging sometimes? 🙂

Tomorrow: Break in

Subscribe to Paul's Posts

3 comments on “Feedback”

  1. It’s also funny how many terms we use on a daily basis can have disparate and directly-contradictory meanings. “Feedback” is one of those terms.

    In audio alone, feedback can refer to a stabilizing force (feedback within an electrical or electronic circuit) or a destabilizing force (think of acoustic feedback). It’s not unlike that confusing term, “oversight”, which can refer to both a power which reviews or acts as a watchguard, or to an error or lapse in that same power. Weird.

    Even more confusing is the variety of connotations the term “negative feedback” possesses. It’s amazing we’re able to understand one another at all!

  2. I disagree somewhat with your statement that “…errors are inevitable and acceptable in a system”. The intent of Statistical Process Control (SPC) espoused by W. Edwards Deming and Walter Shewhart of Bell Laboratories was to narrow the variation of any process to the point where failures were extraordinarily rare. Only in the event of these rare failures was a “feedback” system, in the form of Root Cause Analysis, utilized. It is interesting to note that Deming learned these techniques from the Japanese but could not “sell” them to American manufacturing companies until after they kicked our manufacturing butts, e.g., Toyota. Only then did Chrysler and Ford get religion. Their are numerous instances documented of Japanese manufacturing processes so tightly controlled that there was no need for a Quality Control function…they just packaged them for shipment! In reality Quality Control does not control quality, only the errors accepted in a process that is out of control. Part of Japanese success in dominating manufacture and pricing of resistors was the fact that their tolerances were so tight that even their “errors” were better than those from other sources.

    1. Indeed, this is a good point but I still stick to what I said: in any feedback system, it must make a decision and take an action based on that decision before it can make a recognition of success or a correction to compensate for an error. In most cases, overshoot or undershoot is the result of the action and then a further refinement gets it closer to the ideal.

      A good human example is throwing a ball. If you throw a ball to hit a target, the possible outcomes are perfect hit, too far one way or too far the other way. Feedback informs us of the results of the action, we make a correction and try again. Even if we make a perfect score, we then use feedback to continue doing exactly the same thing until the circumstances (the target or our action) change and then we have to correct.

      Action, error, correction, evaluate, action. True for electronic systems, humans, the universe.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301
1-800-PSAUDIO

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram