Reflow Toaster Oven - Part 7 - Taking it up a notch

Top to bottom: clunky thermocouple in oven (181.2), thermocouple attached to pcb in oven(207.9), thermocouple in air in oven (210.1), thermocouple on back of SSR (51.8, yikes!).

I was trying to keep the heater at 300 deg C, check out how fast the temperature drops when the heater is turned off! And how long it takes for the element to heat up again. Let's have a closer look:

So what do we see here? The thermocouple registers 300+ deg C (blue line) and the Arduino turns the SSR off (red line). The temperature still rises a little bit to 301 degrees, but then after a second or two the temperature starts to drop. Two seconds after that the temperature drops below 300 and the SSR is turned back on. For the next 14 seconds, the temperature keeps dropping to 273, wow! Then it takes another 10 or so seconds until 300 deg has been reached again.
So what happened? Why does it drop so fast? I guess the main reason is the difference in temperature between the heater and the air around it. The air is almost 100 degrees C cooler than the heater. So even though the heating element has been turned off for just a few seconds, it takes quite some time compensate for the energy that has been lost.

So what do these tests tell us? I'm just guessing here, I didn't study physics or thermodynamics so my guess is as good as yours :) But I'm guessing that if we run this test for a longer period then the ratio between on/off (duty cycle) can also be used with a much shorter period (say, 1 second). So if on average the heater is turned on for 24 seconds and off for 4 seconds, then we have a duty cycle of 85% (24 out of 28 seconds on). Then my guess is that the heater can be kept at the same temperature by turning the heater on for 0.85 seconds and off for 0.15 seconds. But the result would probably be a much smoother temperature curve.

My second guess is that the duty cycle will need to be higher when the difference between the heater and the air around it is greater. So that means that in order to keep the temperature of the heater constant, the duty cycle needs to be slowly decreased as the temperature in the oven rises.

I really need to order more MAX6675's and some thermocouple connectors, being able to log only one thermocouple is annoying. I noted a few more numbers on paper during testing.
At heater=200, the air was 140 and the pcb 120.
At heater=250, the air was 170 and the pcb 150.
At heater=300, the air was 200 and the pcb 180.
After some time, the air was 205 and the pcb 190
After some more time, the air was 208 and the pcb 200
After some more time, the air was 209 and the pcb 205
Finally, the air was 210 and the pcb 208. By this time the SSR was well above 50 deg C so I turned everything off.

But as you can see, the air seems to reach a maximum temperature of about 210 deg C, and the pcb slowly catches up. The air will most likely never reach the same temperature as the heater, since it's in contact with the walls of the oven and transfers quite some of its heat to that. The PCB on the other hand is only in contact with the air (and the wire rack, but let's ignore that for now) so it will eventually reach the temperature of the air.

This brings us to the topic of insulation. If we can decrease the amount of heat that is lost, then the temperature of the air will follow that of the heater more closely. Now we can't prevent the air from heating up the walls of the oven, unless we add some sort of heat reflective coating to it (not a bad idea actually), but we can prevent the walls from transfering heat outside of the oven. By padding the walls with insulation we should be able to reduce losses quite a bit. I've ordered some thermal fiber glass insulation designed specifically for insulating commercial and industrial ovens. This stuff is different from normal glass wool that you'd use to insulate your house. Although they're both made of glass fibers, the bonding material used in normal glass wool will start to melt/burn at high temperatures. This oven rated insulation has a different type of bonding material which should be able to withstand temperatures of up to 500 degrees C. It'll be interesting to see the difference this stuff will make :)