This article is another part of the retro-engineering series. Where we take a look at engineering efforts of the past to find insights into the generalities of good engineering practice. This then lets us become better global engineers. This time we will explore the time a group of General Motors (GM) engineers were engaged to develop what many consider to be the first true heart pump.

If you want to read more about this project, then take a look at this article in the New York Times.

In summary though, a team of engineers from GM developed a heart pump that, on July 3, 1952, was used in the first successful surgery where the patient survived while a mechanical heart maintained blood supply. This was a major step forward in open heart surgery.

The first thing that I note about this project was that it was an automotive company that was engaged. I understand that GM was and is a large company with many capabilities, but I would also think that a company that specialises in pumps would be better suited.

So why was GM the company of choice for this project?

It was because, in 1949, Charles E. Wilson, a GM president, was also the chairman of the newly formed Michigan Heart Association. And this was because Wilson had a major interest in heart disease. If he did not have such an interest, then he would not have been a member of the association and unlikely to have been approached. Who knows who would have then been engaged or if the project would have even progressed at all.

The lesson here for the global engineer, have other interests and network so that you can be a part of more opportunities.

So that explains the first step – finding out about the project and being invited to join – but what about the decision to actually get involved, and take on such a challenge?

This is an example of understanding the transferable nature of engineering fundamentals.

“We have pumped oil, gasoline, water and other fluids one way or another in our business,” wrote Edward V. Rippingille Sr., the leader of the team of engineers and researchers who developed the heart pump. He wrote further, “It seems only logical we should try to pump blood.”

There are indeed differences to how blood and Newtonian fluids behave, but, at the same time, lessons from pumping the latter can be applied to pumping the former. Rippingille showed the ideal global engineering perspective that one should not assume a new field is completely foreign and off limits to them. Instead, it was realised that there are fundamental commonalities, and that these can be used when moving from one field to a new one.

But still, it was different, and those involved were wise enough to understand this. The General Motors Research team started by reviewing almost everything that had been written anywhere on the subject. Further, Rippingille travelled extensively, examining various pumps that had already been made for such use but had failed for one reason or another. This shows an at least implicit understanding of the importance of systemic thinking and first principles. By studying the prior art, an engineer can quickly acquire the respective domain knowledge needed for systemic thinking and use of first principles. And that’s what would have been achieved by reading prior art and looking at what others had done.

But even then, with all that research done prior, it was not a linear product development process.

Over a period of 30 months, 6 to 10 concepts were built and tried – and 84 dogs were lost through testing. For this project to work there was funding from The Michigan Heart Association and GM had provided support as a public service. This commitment and support did not seem to waver after the first prototype. Thus, there was an implicit understanding of the need to implement solutions to better understand the challenge – many call this iteration, but, in contexts like this where the challenge is new, the global engineer knows it to be co-evolution. And initial failures are not just one more step to success, they are an essential part of defining both the challenge and the solution.

Despite the example of engineering expertise noted above, there was some hint of cognitive laziness or automatic association.

Sometime later, Dr. John W. Kirklin from the Mayo Clinic, who was conducting research into heart-lung machines, reviewed The Michigan Heart Pump and one that had been developed by IBM. He noted the former looked like a car engine and the latter looked like a large computer. This led to some thinking the engineers actually miniaturised a V12 for the job. In reality, this was more likely a case of fixation – where a designer has ideas in their head that they do not realise they have and that they and can’t shake. These are not always bad – and in this instance it might have allowed the engineers to focus more of their engineering efforts on the real challenge – moving blood.
​
In summary, the case of The Michigan Heart Pump is an excellent example of global engineering expertise: engineers networked, found new opportunities, understood the fundamentals, leveraged existing knowledge, understood what it really takes to tackle such a challenge, and let their fixation from prior experience reduce the cognitive effort. It is something that you and I can use as a reference for engineering best practice.

In this article, we're going to delve into the design of "clickers," or old TV remote controls from the 1950s. This marks the first installment in our retro-engineering series, where we examine past engineering designs to gain insight into the methodologies engineers employed. With the benefit of hindsight, we can discern which designs succeeded and which did not. By reverse-engineering the processes behind successful designs, we can glean valuable lessons in sound engineering practice.

Let's begin by exploring how these clickers functioned.

You might be aware—or perhaps not—that these clickers operated using sound. Inside each clicker were several metal bars. When tapped, these bars resonated at specific frequencies, each corresponding to a particular television function: changing channels, adjusting volume, powering the TV on or off, and so forth.

Pressing a button on the clicker activated a spring-loaded toggle mechanism, causing a small hammer to strike the appropriate bar. This produced a clear ringing sound at an ultrasonic frequency, inaudible to the human ear. From the user's perspective, the experience was akin to using modern remote controls.

Understanding the mechanics of these clickers allows us to critically assess the development process behind them.

Firstly, it's noteworthy that the designers accurately identified user needs. From an outcome-driven innovation perspective, this was spot on. Users preferred the convenience of controlling the television without the need to physically approach it – it made the job to be done much easier.
This necessity raised the question: how to transmit the user's command to the television?

This framing led to two critical considerations: the nature of the signal and the medium through which it would travel.

Potential options at the time included:

• Infrared signals, akin to those used in many modern remotes.
• Radio waves, similar to those employed in remote-controlled toys.
• Wired connections, as seen in some early remote controls.

However, each of these options had drawbacks. Infrared and radio wave technologies were prohibitively expensive due to the cost of components (recall Moore’s Law where electronics have been regularly halving in size and cost while doubling in power for decades now) and the limitations of battery technology at the time. Wired connections, while feasible, were cumbersome and detracted from user convenience.

Consequently, engineers had to explore alternative solutions. Sound, particularly ultrasonic frequencies, emerged as a viable option. By applying fundamental principles, they realized that the natural frequencies of metal bars could be harnessed to generate distinct signals.

This realization reframed the design challenge: creating a handheld ultrasonic transmitter.

Because TVs were already established, along with the internal technoilogy, it was possible to create a circuit to process the signal from the microphone using vacuum tubes. They were not transistor type electronic devices that are the subject of Moore's law. Therefore, it likely would have seemed quite obvious that the microphone should be attached to a circuit to activate the respective switch depending upon the frequency of the signal reaching the microphone and being converted into the electrical signal. I don't want to be dismissive of this though, it still would have been a design challenge and an engineering challenge, it's just that it probably would have flowed a lot more after the previous frame, designing an ultrasonic generator that can be held in one’s hand, was formed.

For a more detailed history of TV remote controls, you can refer to Zenith's heritage page: Zenith Remote Background.

Now here’s a question for you: would you have conceived the same solution under similar constraints? Contemplating this can provide deeper insight into your own engineering skills and how to further develop them.
​
Additionally, if you have alternative ideas for creating a remote control using 1950s technology, feel free to share them in the comments. Personally, I pondered the use of inductors and capacitors to filter the microphone's signal into the respective circuits as an alternative to vacuum tubes. That’s because I recall making a filter for a speaker box I made some time back, and I have a fixation on such filters. So also think about what made you come up with the your idea.