Hello, my name is Josh Leworthy, I’m a Mechanical Engineering student at Heriot Watt University, Edinburgh, and I have just completed my 4th of 5 years there. Since the beginning of July I’ve been working at Dukosi as an engineering intern, designing and building a thermal test chamber as well as working on a few other smaller projects.
The Thermal Test Chamber Project
When I arrived at Dukosi, they were making plans for an Arbin cell tester, a piece of kit used to characterise Li-ion cells. The Arbin will be used to run the cells through what is essentially a life cycle, charging and discharging them at different current rates. As well as that, we need to be able to vary the temperature of the cells during tests, as with most electrical products, temperature massively effects how Li-ion cells behave, so this is where the thermal chamber comes in. The chamber is essentially a box, that houses the cells under test and controls its temperature, in this case, at any given point between -20°C and 50°C. Joel (CTO) challenged me to build him one for less than £1000, a tenth of the cost of a conventional chamber. So here’s a very brief summary of what I managed to do and how I got there.
First of all, I identified the three major functions of a thermal chamber, heating, cooling and temperature control, and set about designing each component. For the cooling and base of the thermal chamber, I used a humble household chest freezer. An ideal starting block as it is air tight, well insulated and more than capable of achieving temperatures down to -20°C. I could have constructed my own chamber and used peltier devices to remove heat, but that would never have done a better job than the freezer for the same cost.
Once we received the freezer, I could do a few experiments and calculations to find out how much heat power the freezer could remove, and then put together a heater with a similar power output. The heater assembly uses a combination of resistors fixed to two large heat sinks and a fan. The fan draws air across the heat sinks and circulates air around the chamber, giving a more even temperature distribution inside the chamber. Another variable that needs control is the humidity of the air within the chamber, humid air can lead to condensation and ice – not good! We planned to tackle this with some desiccant or CO2 flushing.
The last component is control; I used a PID temperature controller, which is essentially a clever thermostat that uses a probe to monitor the temperature inside the chamber, and then controls the heater and freezers compressor in order to achieve the desired temperature.
Now we had all of the components, it was time to cut out all of the freezers original wiring and thermostat, tentatively drill some holes and wire it all together. Hey presto, you have a thermal chamber. Well, not quite…
Building the thermal chamber was the easy bit; the next stage involved fiddling with many pages of parameters on the controller in order to achieve steady control of the chambers temperature. A long and tedious task but worth it for the result! I also used these parameters to implement some safe guards against potential damage to the chambers components and the cells by adding high and low temperature alarms and output cut offs in the event of serious alarms.
So that’s a short and sweet summary of what it is, but how well does it work? Well, after a few weeks putting it all together, and another few weeks of testing a tweaking, here’s what we have.
I am happy to report that we have a thermal test chamber, with a temperature range of -20°C to 50°C to an accuracy of ±1°C. The graph above shows the chamber temperature over time as it controls it through various set points. The controller has managed to achieve a nice, steady control of temperature throughout the range with minimal overshoots. Overall I am happy with the result, and all under budget, in fact less than £500!
I’ve got a couple of days left here before I go back to uni, so I’m adding a few final features. I have added an email alarm system using Lab View, this means we can leave the thermal test chamber running without having to baby sit it, and it will notify us if there are any unexpected temperatures spikes. And last of all is to connect the controller to the PC, so as we can have full control over the Arbin and chamber from one point, making the process leaner and easier to manage.
And that brings me to the end of my internship, I’ve had a cracking time at Dukosi, a great team and a great BMS! I’ve been able to apply a lot of what I’ve learnt and also learn a great deal more, I couldn’t have asked for a better opportunity really.
Thank you to the team, Adrian, Carys and Gordon. Also a special thanks to Joel for hiring me in the first place and being a great mentor throughout.
Onwards and upwards!