I'm using the term "bought solution" because I couldn't think of something better to refer to those PID temperature controller devices for sale.
In the realm of small projects, can I make an Arduino based temperature controller better than any of the pre-packaged closed-source solutions out there?
My thinking is that, while coding and building myself a solution, I can tweak anything, auto-tune in different ways with different algorithms, and so forth. While if I just go out and buy a temperature controller, I'll be tied to the hardware and firmware in whatever form, and with whatever bugs, that seller was able to deliver.
Am I right, or are the premade solutions good enough that I can only, at most, match their perfomance?
Your thinking is correct. Certainly there are excellent premade temperature controllers that will do their job to any level required. However, by building your own you will achieve the highest level of control over the firmware and hardware. I've seen some temperature controllers with open source firmware that can be programmed with the Arduino IDE so that could be a middle ground if you want to customize the firmware but don't need any specialized hardware configuration. Another benefit may be that it's just a fun project that you can learn a lot from and end up with something actually really useful.
I've been using temperature controllers on my glass annealing kilns for the last 15 years. I always used the low end models and went through quite a few different models over the years as I upgraded, added new kilns, and replaced dead ones. As soon as I moved past the most basic setpoint controllers I was always annoyed with the user interface. I also ran into some firmware bugs that I had no way of fixing. By the time I was experienced enough with Arduino to make my own I knew exactly what I wanted in a temperature controller. It's possible that a very high end controller could meet most of these needs but I value understanding and having complete control over every part of this device that I use daily and rely on to work correctly. My decision was also helped by the fact that I use multiple temperature controllers. Making each subsequent controller is much less work than the first since the design and firmware are reused. Even so it's a lot of work and the parts add up to around the cost of a cheap premade controller.
For me it worth doing and I've certainly been glad I did over the last few years of using it but I'm sure your situation is not identical to mine so in the end only you can decide if it's worth the work.
If you're controlling anything (whether using a premade or DIY controller) that could end up in dangerous or costly damage if you experience a runaway temperature event then I recommend adding a completely separate "limit controller". In my case this is a cheap setpoint controller set to a temperature a bit above the highest temperature my system should ever reach under normal operating conditions. This means if there is a firmware bug, relay contacts weld, thermocouple knocked loose, etc. which would cause the kiln to over heat, the limit controller will prevent this.
Thank you for your answer. I'll be looking into a setpoint controller. In my case this isn't for work. My plan, for now, is melting and molding aluminium, and controlling reactions, ranging from 100°C (212 F) to 900°C (1652 F). After that, if I ever manage to develop a version 2.0, then it will be awesome to reach those kinds of temperatures required to work with glass and iron.
As long as you're not using borosilicate glass, you can do glass fusing or casting at ~1500 F. Borosilicate glass requires a higher temperature but it's rarely used for kiln forming processes anyway.