When Electronics Age Faster than My Product

Time to Read 6 min

Most electronic components are available for three to five, sometimes ten years. This is because, driven by consumer electronics, it makes no sense for manufacturers to produce components for longer. Unfortunately, key components such as processors and memory are particularly susceptible to these cycles.

This is a problem especially for networked systems and the IoT (Internet of Things), as capital goods and buildings are often digitalized, i.e., networked. It is also worthwhile for other embedded systems and mechatronics to address this issue at an early stage.

There are various aging mechanisms, or ways in which this aging affects components. Let's take a look at them:

Aging of the Components Themselves: Limiting the Service Life of the Product

A machine tool is controlled by a microprocessor which, partly due to the harsh environment (heat, vibrations), fails and puts a six-figure piece of equipment out of action. If the chip is then no longer available (obsolescence, see next paragraph), this can become a real problem for the user and then also for the manufacturer.

Measures to Mitigate Risk

In a sense, this is the simplest type of aging, as it is well known in electronics and therefore established measures exist to reduce risk, all of which are taken during the development phase:

• Component selection: there are components (e.g., from the automotive sector: AEC-Q (Automotive Electronics Council) qualification) that should have a longer service life and components for which, cycle counts and service life are specified, for example
• Calculation and optimization of service life: there are basic principles for calculating the service life of electronics; unfortunately, the data on which the calculations are based date from the last millennium and the results should therefore be treated with caution
• Derating of components: a method used in the telecommunications industry, for example; the manufacturer's important maximum limits (e.g., voltages, currents, clock rates, temperatures) are “derated,” i.e., reduced by a percentage (e.g., the processor is only operated at 50% of the maximum specified clock rate)

Obsolescence of Components: Limiting the Market Life of the Product

The production of the control system for a measuring device fails because the key components (microcontroller and CAN bus control) are no longer available. In order to continue selling the device on the market, the control system, which „actually still works”, must be completely redesigned, including a significant portion of the software, as new integrated circuits for the processor architecture of the existing device have not been available for a long time.

Measures to Mitigate Risk

This type of aging is often not on the radar of developers and product managers. However, there are some measures that can be taken to reduce the risk of having to develop a new product:

• Use of “longevity” components: components that are sold by manufacturers specifically for such requirements, with a more or less (usually less) specific product life guarantee of typically ten years
• Established manufacturers: use components from established manufacturers and those known for not discontinuing products quickly; note that these are often not the cheapest suppliers
• Good software structure and documentation: if the software (including operating system and drivers) is well structured, e.g., has a sensible HAL (hardware abstraction layer) and is well documented, then the effort required to port it to new hardware can at least be reduced if the worst comes to the worst
• Store components: it is also possible to have the assembler store sufficient components under inert gas so that they remain ready for assembly over a longer period of time

Software Obsolescence: Limiting the Service Life of the Product

This is perhaps the most brutal way a product can become obsolete: A manufacturer of networked building installations opted for a computer platform with an operating system some time ago. After fifteen years of operation, customers and their compliance auditors are now demanding that only those who keep their software up to date are allowed to connect to the network. The processor has been discontinued for twelve years and has not been supported by the operating system for ten years. Unfortunately, the rather exotic but inexpensive form factor of the processor module (COM: Computer On Module) has also been unavailable on the market for some time.

What now? Both the software and hardware would have to be completely redesigned just to avoid having to dismantle the devices at the customers' premises.

Measures to Mitigate Risk

Also this risk can be mitigated, with measures that are becoming increasingly relevant anyway due to cybersecurity regulations:

• Standard form factors: if standard form factors are used for the processor modules, the effort required to switch to newer processor generations can be massively reduced, as no or only minor adjustments to the module with the latest available processor are necessary
• Standard platforms: use processors, operating systems, libraries, and programming languages that are not exotic, as this will increase the likelihood that they will still be supported in many years' time
• Good software structure and documentation: if the software (including the operating system and drivers) is well structured, e.g., has a sensible HAL (hardware abstraction layer) and is well documented, then the effort required to port it to new hardware can at least be reduced if the worst comes to the worst.

Aging and Sustainability

Of course, this topic also has a sustainability aspect beyond the examples above. Does it make sense to throw away entire devices just because a component is no longer available?

Aging and Business Model

All risk mitigation measures have one thing in common: the solutions are never the cheapest solutions at first, i.e., during development. Here, too, it is necessary to consider commercial sustainability (TCO: Total Cost of Ownership), which is unfortunately often lacking.

Another important point is that maintenance (e.g., updates, see also cybersecurity below) costs money throughout the entire product life cycle. How should this be reflected in the business model?

One solution is „on demand” models, such as „power by the hour.” But what happens when there is no longer any „demand”? Is the entire product removed from the building or does it end up in the trash?

Impact of Cybersecurity Regulations (CRA, NIS-2, RED, etc.)

The new cyber regulations require the tracking and correction (i.e., software updates) of vulnerabilities throughout the entire product life cycle. This forces manufacturers to maintain their products. These costs must be factored into the price or the overall business model.

Impact of safety regulations (Machinery Directive, etc.)

As soon as a “significant change” has to be made, there is a risk of a new conformity assessment in accordance with the current state of the art, which may then entail additional changes. With the associated costs...

Risk Mitigation and Cost

As mentioned above, long-term considerations are required here, because once the risk has materialized, once the damage has been done, it usually becomes very expensive with subsequent developments and replacements in the field. The manufacturer's reputation can also be damaged; just think of the publicity surrounding car recalls.

Sustainability in terms of long product life is one of our values, and quality is how we achieve this. Talk to us about your challenges in an initial, free workshop. We have also solved obsolescence problems in a „minimally invasive” way on several occasions. Contact us for an initial assessment.

Andreas Stucki

Do you have additional questions? Do you have a different opinion? If so, email me  or comment your thoughts below!