Hardware control of a long lifetime system

[tobiasg]

Hello there,
I'm working in a laboratory and currently planning a project to renew the control of a measuring station. The station is equipped with 4 motors (PWM for drivers is needed) and 5 sensors (digital), which need to be controlled via a PC (GUI). It's important to note that the project have a lifespan of >10 years and the components should still available in the case of failure.
I'm considering to implement the I/O stuff either via Arduino or via a LabJack T7. Whereby, Arduino is already out, because of its short lifetime, but I'm interested about your thoughts about it. Maybe there is a similar, longer lasting solution.
In addition, the development speed with the LabJack should be faster, because of a provided library which can control the device, no need to program a µC. LabJack has currently no plan to discontinue the T7 and have designed it for 10+ years lifetime. But even in the case that LabJack discontinues the T7 it supports Modbus. If I control theT7 not with the provided library, but via Modbus directly (slower development speed, but still faster than programming a µC), I should be able to replace the T7 even with a device from another vendor.
Also, the price difference is heavy, and I don't really know if a LabJack T7 is an overkill since I will only use a small subset of the functionalities provided. Possibly, there is a cheaper solution for this task.
Since I have not much experienced in tasks like this (purely Software Engineer), I would be happy about some feedback about my plans.
Thanks,
Tobias

 

[Bluejets]

Arduino is already out, because of its short lifetime,

News to me........

 

[tobiasg]

News to me........

Maybe I worded it wrong, can I be sure that a certain Arduino board will still be available in 10 years?

 

[Bluejets]

If anyone knew anything remotely like that, they'd be very rich indeed.
Unlikely though as technology moves on.
Your idea may well be old hat by then as well.

 

[tobiasg]

Of course, they will be old at some point. LabJack uses especially parts from manufacturers which have a history of supporting their parts for a long time to ensure that they can continue their devices and are explicitly committed to do so. I couldn't find anything like that at Arduino and because I know models from them that aren't continued, that's why I had the presumption that an Arduino isn't the right decision.

 

[Harald Kapp]

It's important to note that the project have a lifespan of >10 years and the components should still available in the case of failure.

Get comfortable with the idea that during a lifespan of > 10 years you will have to deal with discontinuation of components. Happens all the time in the business I'm in.
What is more important than the availability of specific components is the availability of the architecture underlying your project. A stable architecture allows you to continue with the project even in the absence of original components by using updated/upgraded components.
I am not going to promote the Arduino architecture in particular - there are others as well suited - but just as an example:

• Standard module pinouts allow replacement of the CPU by another, more modern type. Not only in case of discontinuation of the original CPU but also in case you need more power or additional functionality.
• The Arduino IDE is quite stable (although a complete rework is in the pipeline). Alternatives exist, too (e.g. Visual Studio Code with PlatformIO). SO chances are good that you can work in a familiar environment.
• Arduino has a huge community where you can find support of all kind, be it electronic circuit theory, libraries for using other components, help with coding issues etc.

Of course you can find similar support for other platforms, to. E.g. ESP32, STM32, Raspberry Pi etc.. Just don't get stuck on the idea of > 10 years support. It won't work. Unless you are going to pay a premium for stocking up in parts.