Here at HiETA we’re developing some cutting-edge products through metal additive manufacturing (AM). We’re a growing company that is constantly developing our capabilities to meet customer and internal requirements. One of these developments has been the introduction of a heat treatment furnace into our manufacturing facility at the Bristol and Bath Science Park.
By heat treating our products we can alter their material properties which, among other things, effects the strength of the material; hence it’s important that we get this correct to ensure our products can withstand the application for which they have been designed. A heat treatment process consists of heating a component up to a high temperature and holding it at this temperature for a set time (typically around 730°C and 2 hours for Titanium); followed by a controlled cooling procedure.
The problem
To be sure we are correctly treating our components, we need to log the temperatures to which our components are subjected, which is a feature that our new furnace did not provide as standard. The furnace manufacturer would be able to provide us with this feature, but at a large additional cost, which would then need further integration into our IT infrastructure to be able to capture the data on our internal servers.
The solution
After a quick assessment with the engineering team, we decided to develop a solution ourselves as we could control the whole process, make changes as and when we needed, we would end up with a solution that is better integrated with our IT infrastructure and finally, it would be significantly cheaper than the furnace manufacturer’s solution. The better integration with our IT infrastructure would allow the temperature logging to be an automated system that reduces operator workloads and errors of capturing and storing temperature logs for each individual heat treatment process. The solution we decided upon was based around a Raspberry Pi.
We decided upon using a Raspberry Pi for many reasons:
- Firstly, it’s an inexpensive computer, at around £50 once all the additional components are purchased with the Pi itself, that can be easily run without a display, keyboard or mouse, but has the benefits of having a complete on-board Linux operating system.
- It is a low powered device which avoids the need for cooling fans and hence it can be mounted within a sealed box, minimising dust ingress. It’s low power also allows the device to be permanently on without incurring large energy costs.
- The full Linux operating system means interfacing between other devices and systems required to measure the furnace temperature and capture the data on our server would be easy.
- The on-board RJ45 Ethernet port would allow the Pi to be easily connected to our internal wired network.
- The on-board USB ports mean we can connect to simple temperature measurement devices with ease.
- The native Python programming language is easy to work with, has a large number of available libraries that make it a very capable language and has a large, active online support community.
How it was done
We decided to use a K-type thermocouple to measure the furnace temperature as these have a suitable range and accuracy for the temperatures we’re planning to log, and they can be made in the lengths and thicknesses that we need to place the measurement tip at the correct location. On top of this, simple, cost-effective data acquisition equipment is available that needs only a USB connection.
At HiETA we have developed our own internal MySQL database for the capture of production information. As the Pi is permanently on, it is constantly monitoring the furnace temperature and providing a real-time reading onto our database. Once this temperature is greater than a set minimum temperature (around 100°C) the Pi starts to log the measured temperatures on the database. We set this minimum log temperature as there is no point in logging the temperature whilst the furnace is not in use, and below this minimum temperature, none of the heat treatment cycles we use have any effect. This reduces the data we capture, and hence required storage space; whilst being permanently on and monitoring eliminates operator error of forgetting to start a manual logging system.
Pi-powered temperature logging: clever engineering by clever engineers
By capturing the heat treatment temperatures directly onto our production database, we have a single centralised location for all of the data. This reduces the chances of losing data. Also, the database is backed up automatically, hence the security and stability of the captured data is ensured. It is also available to the whole company, as opposed to the chance of it being kept locally on a single operator’s computer.
Within our production facility we have tablets available such that operators can log and monitor activities. We developed a simple webpage that interrogates our database and displays the real-time temperature as well as the last 20 logged readings. This real-time read-out gives a quick indication to the operator that the Pi is working correctly, as well as showing that crucial heat treatment temperatures are being logged. If the real-time read-out is in the past (by more than its refresh rate) it indicates that there is an issue with the Pi and this should prompt investigations.
Our IT infrastructure has Power Over Ethernet (POE) capability such that we can transfer power to devices as well as data to and from them using a single Ethernet connection. We decided to make use of this, as the fewer the connections to our logging box, the easier it is to maintain and install. As such, we installed a 5v POE splitter that separates out the power supply and allows the Pi to be powered from just a single Ethernet port in the production facility.
Putting it in place
To be able to mount the Pi, it needs a case, of which many are available. We decided to design and manufacture our own bespoke case as we would not be using the majority of the connections on the Pi. As such we were able to create a simple case onto which we could place our own branding keeping a simple design with the bespoke functionality that we need.
“HiETA has an automated furnace data capture system with minimal operator interaction, reducing their workload and chance of errors occurring due to incorrect storage of data”
Finally, after installation and testing, HiETA has an automated furnace data capture system with minimal operator interaction, reducing their workload and chance of errors occurring due to incorrect storage of data.
Though a lot of what we do is pushing the possibilities of available additive manufacturing technologies, all of our work involves the innovative use of a wide range of technologies whilst constantly assessing the impact, benefit and cost of all available solutions. The development of this automated temperature logging system shows that HiETA has a close interdisciplinary team that is not solely additive manufacturing focussed but includes instrumentation, software development, and data and production management expertise.
