With this instructable, we will take a cheap toaster oven and turn it into an accurate, temperature controlled tempering oven that will be able to achieve a stable and accurate temperature controlled by a microprocessor.

We will combine some common and off-the-shelf components to easily and safely achieve this goal.

Why do we want to do this? After heat treating steel, it's in a very hard and brittle state. So we can use it with less risk of it snapping, we want to temper it. Tempering will slightly soften steel, but more importantly will restore flexibility to it. Most steel manufacturers specify particular temperatures to temper steel at so you can achieve a given hardness rating.

Having the temperature accurate to a fraction of a degree isn't critical for tempering, but I want more accuracy and repeatability than the simple bimetallic strip thermostat that most inexpensive ovens use. ±10°C could easily be a difference of 1-2 Rockwell C points in hardness.

Step 1: Requirements

Here's what you'll need

  1. Toaster Oven
  2. PID Temperature Controller
  3. Solid-state Relay
  4. Temperature Sensor
  5. Insulation
<p>The chinglish instructions with these controllers is very hard to decipher!</p><p>Nice 'ible. Very well written. It helped a lot. :-D</p>
<p>Awesome work! That looks mint. Good to see that the temperature is spot on according to three different sources too.</p><p>What temperature probe are you using with the PID? The Type K? What temperature offset did you need to program in for it?</p><p>Glad my instructions helped in getting the PID set up correctly - the instructions are pretty confusing.</p>
In these pics i was using the K thermocouple that came with the controller. I think this thermocouple isn't great though because as you see in these pics it's spot on, but at higher temps there is an offset. I can't fathom why that would happen though since thermocouples are supposed to have linear response.<br><br>Anyway,<br>I swapped in a Thermoworks thermocouple and it worked perfectly. So I'll probably ditch the one that came with it.
<p>Thanks for the update. I've got a PT100 sensor on order, I figure that if it's made to spec so it is 100&Omega; at 0&deg;C then it should be much more of a known quantity than the thermocouple I have. I'm concerned that with the thermocouple I have that something isn't right having to put in such a huge offset (more than -70&deg;C)</p>
<p>Hey Kai, I was using the PT100 and i had a offset of 2&deg;C</p>
<p>&quot;...is classed as hazardous. It may cause cancer by inhalation and is irritating to the skin...&quot; sounds bad, there was no better option?</p>
<p>all insulation is classed as hazardous due either the inhalation problems and even the skin reaction to a lot of fiberglass. the best insulation for this type of job is stone wool which is trill an inhalation hazard, but this stuff works for the temp ranges he did, excellent structabe BTW thanks</p>
<p>Pretty much this. Insulation has to resist heat and trap a large amount of air within it's mass. Most, if not all, materials that have these properties are not good to breathe as they're fine mineral fibres - Asbestos is a naturally occurring insulation fibre. Synthetic insulation is not as bad for you as asbestos, but you still don't want to get it into your lungs as the fibres can become lodged in there, build up over time and cause health issues much later down the track.</p>
Most things aren't good to inhale, water for example, but I'm not sure water will cause cancer. ;-)<br>I understand why it was easier to skip the wool for the manufacturer in the first place. <br>I have been looking differently an those mini-ovens now... I've got a few used spare regulators at work and I'm tempted.
<p>Yeah, there are probably a few other alternatives, but I had some of this stuff on hand, it's totally non-flammable and is rated at temperatures up to at least 1000&deg;C, so it doesn't matter that it's directly in contact with the ends of the heating elements. If I was buying some materials directly to do this project, I probably would have got some rockwool or something similar to that. Isowool is easer to work with for this application as it's in thinner sheets (~20mm thick or so) whereas rockwool is designed for home insulation so comes in much thicker batts.</p>
Seems to be also perfect for reflowing of solder on PCBs.
<p>If you want to make a super-deluxe reflow oven, you'll want a slightly more advanced PID controller - one that adds timing to it.</p><p>It's my understanding that for reflow, you need to heat up at a certain rate, hold it at a particular temperature and then let it cool at a certain rate as well. A more advanced PID will allow you to program in all these ramps automatically, whereas with my version you'd need to monitor it yourself.</p>
<p>Very well done!</p><p>Your Instructable is very well thought-out, clear, flows well, has good photos (with helpful text overlays), and is easy to follow. I like the extra care you took to elaborate on each of steps. I particularly like the explanation regarding the thermocouples.</p><p>Thank you for sharing this with us and for the preparation you took to make it clear.</p>
<p>Thanks for the feedback!</p>
<p>Good job! A few questions though: 1) what temperature are you trying to achieve and is your oven/equipment rated for such a temperature, 2) have you tuned the controller (determining the coefficients of the algorithm)?</p><p>SSRs do indeed fail closed. As a safety feature, a thermal cutoff (fuse) can be added and/or a regular fuse rated at 10A or 16A.</p>
<p>The oven had a thermostat that went up to 240&deg;C.</p><p>For tempering metals, this is about perfect. 230&deg;C is &asymp; 450&deg;F. If you're tempering any higher than this, you're losing a lot of hardness.</p><p>I've got a piezo buzzer on the way that I'll wire up to the alarm outputs of the controller and set the high temperature alarm at something like 250&deg;C </p>
<p>A 74C initial error in the temperature sensor is enormous! Anything over +/-5C and I would send it back to the manufacturer. Are you REALLY sure you have the right type of sensor for the way it is programmed? Red and Blue wires usually means it is a type T (copper/nickel) thermocouple. Type K wires are usually Red and Yellow.</p>
<p>I wasn't sure about that - whether or not a large initial error away from 0&deg;C was an issue or not.</p><p>Despite this large offset, it does seem to be pretty much spot-on for 0&deg;C and 100&deg;C.</p><p>The eBay listing definitely said it was a Type K thermocouple. According to Wikipedia, red and blue are sometimes used for Type K, whereas it's red and purple for Type T.</p><p>This is the first (and so far only) thermocouple I've had to calibrate an offset for, I have no prior experience with these devices.</p><p>I've ordered a PT100 sensor and am waiting for it to come in from overseas, the only thing is that taking the oven apart and putting it back together again is a right royal PITA. What I now plan to do now is run a few spot checks at different temperatures and check with an IR thermometer or something else to make sure that I'm getting a relatively accurate reading.</p>
<p>Very nice instructable, I was considering getting an oven for SMD soldering, now I have gotten a few good ideas. </p>
Thanks for this cool instructible.<br>I loved reading it. did hope to see pics of the insulation process.
<p>I got over to the workshop today and installed the insulation. It was pretty fiddly, but on the right side, you can see how I cut some slots for the heating elements to fit into and it all holds in place very well.</p><p>https://www.instructables.com/id/PID-Temperature-Controlled-Oven/step12/Add-Insulation/</p>
<p>I will be installing the insulation on the weekend and will take more photos to update this instructable.</p>
Nice job. The as built pictures look really good. Our maker group is doing a similar project where the oven will be used to reflow solder SMS circuit boards.<br>Our local pottery suppliers sells a ceramic blanket that isn't carcinogenic. Apparently the fibres are water soluble so they won't stay in your lungs.<br>I've been told that the failure mode of an SSR is in a closed position. As such the oven would be energized without any temperature control. This can be a problem with unattended devices. Th PID control I have has an over temperature set point that can be used to initiate an alarm if things go wrong.
Yes, you're right (according to Wikipedia) - SSRs tend to fail closed whereas mechanical relays tend to fail open.<br>The Inkbird PID does have a high temp alarm (and a low temp alarm too) - I might have to look into hooking it up to a buzzer to the alarm outputs.<br>Temperature regulation with a failed SSR will be more critical when the oven is insulated as the temperature will be retained more than with the factory configuration of just an air gap for insulation.<br>
<p>I thought SSRs failed open and electromechanical relays failed closed ... Could be wrong though.</p>
<p>According to Wikipedia, SSRs tend to fail closed whereas relays tend to fail open.</p><p><a href="https://en.m.wikipedia.org/wiki/Solid-state_relay">https://en.m.wikipedia.org/wiki/Solid-state_relay</a></p><p>I'll have to keep it in mind when operating the oven</p>
<p>You got my votes!</p>
5 stars. Nice writing. Thank you.
<p>This is not a hack, this is an <strong>Arc de Triomphe</strong> of engineering, what a great job!<strong><br></strong></p>
<p>Great project and a really nice write up!</p>
<p>Thanks for the feedback!</p>

About This Instructable




Bio: I like making things.
More by kai.h:PID Temperature Controlled Oven 30 Second Zipper Repair Make a knife from high-carbon steel 
Add instructable to: