Abstract
This post describes a problem (and a partial solution to it) when using cheap IR thermometers to measure high temperatures on ultralight tent stoves. These stoves are made of low emissivity stainless steel (or titanium) which causes a ~50% underestimation of the true temperature. The application of a DIY high emissivity render to the stove can greatly improve the measurement.
At the same time, the stove render improves the heat radiation from an ultralight tent stove surface. This is good news about improving the cooking power and radiant heating comfort for campers in a winter tent.
However, it makes the issue of temperature measurement more complex. Because the render causes more heat radiation it lowers the very temperature that we are trying to measure, as demonstrated in this post.
“If all this is to head-banging for you, please do not continue. I would not like you to lose any more hair or sleep! Otherwise, stick with me and see if what I say makes sense to you. Alternatively, you may like to kindly explain to me what I have got wrong.”
Introduction
I consider my tiny tent stoves to be ‘heat radiators’ rather than tent air heaters, although they do both. Consequently, this post focuses mainly on factors related to the optimization of radiant heat rather than conducted or convected heat. “In my experience, with tent stoves, the air in the top of a small tent can get unpleasantly hot. This is possibly good for drying gloves and socks and keeping the tent free of condensation but not much else.”
Importance of stove surface temperature
Radiant heat increases as Temperature4. This comes from the Stefan–Boltzmann law. It describes the power radiated from a black body in terms of its temperature in degrees K.
This means that higher stove temperatures will cause enormous increases in the amount of radiant heat that will be provided to campers from their stove.
This increases according to the absolute temperature T4 (degrees K multiplied by itself four times). If I have the maths correct, this means that compared with a stove at 300C the stove at 400C would radiate twice as much. It would be three times as much at 500C and 5.3 times at 600C. “These are big and important differences that are largely invisible to us if we cannot measure stove temperature.”
To manage and optimise this important heating factor, I need to have crudely accurate temperature measurements that I currently do not have. “As the saying goes, if you can’t measure it, you can’t manage it.”
What is the true temperature of my stove?
Colour temperature estimation. Using eyeballing to estimate temperature has been used by Blacksmiths forever. Mick Wests informative analysis verifies this robust method of estimating the temperature of very hot metal by its colour.
An example of this colour/temperature relationship in the chart below is from Reddit Inc
However, even in darkness, this temperature estimation is rather subjective even with good colour vision. It becomes almost impossible in daylight, and even worse if you are to any extent red/green colour-blind. Consequently, cheap IR thermometers could provide good supplementary temperature measurements.
An ode to the colour challenged person:
What is this vague sense that many call red,
The range of colours is beautiful, it’s said,
Imaginations figments or real colour pigments,
Wavelengths and frequencies are just sensed in the head.
Another crude but effective temperature ‘digital’ temperature measurement
For years I have used a simple temperature ‘measurement’ to do a safety check on my flue pipe temperature where it passes close to the nylon tent canopy where its exits. I use a ‘dab of saliva’ on my fingertip. I repeatedly test down the pipe until I find the zone on the flue pipe that spits back. I assume that this is at ~100C if it can make steam.
An ode to a primitive digital thermometer that you can really trust:
Digital thermometers on both hands did sit,
Alas, for only one temperature, were they fit,
For 100 C to display or absolutely 373 K,
Reached when steam from fingertip did spit.
Caution: On a more serious note, this test is only for ultra-thin SS or titanium flue pipes, nothing thicker. When the steam point is cautiously found by moving down the pipe, a slight ‘zzzt’ sound can be heard and ‘bubbly feeling’ can be felt. However, it does not hurt or feel particularly hot.
I think this is because the thin metal foil is a very bad heat conductor and has negligible thermal mass. This means that only a tiny amount of spit turns to steam that you can hear and feel. It simply collapses in the remaining spit and does no harm. “It has a kick with no heft!”
Infrared thermometer temperature measurement
Cheap IR thermometers would be good if they worked correctly, particularly in situations where there is no visible radiation to make colour /temperature estimates.
I have made many observations, over many years with cheap IR thermometers. I have had a growing suspicion that they grossly underestimate hot stainless steel stove temperatures (~50%). I confirmed these suspicions when I calibrated the IR temperatures against the above crude but robust colour/temperature estimation under ideal conditions.
I am certain that the problem stems from the low emissivity of my shiny stove surfaces coupled with the cheap IR thermometers inability to adjust for a particular surface emissivity.
Improved IR temperature measurement by increasing surface emissivity
“I should have done these simple tests years ago, but anyway you know how it goes with procrastination……..” To prepare for this post, I put refractory render on various surfaces. I did this to the whole surface or just a part, in order to have a control surface under similar temperature conditions.
Bean can emissivity/temperature test
I put a refractory render spot on one bean can and coated the whole surface of another. I heated the cans filled with water until the water boiled. When removed from the stove I rapidly measured the temperature of the spotted can above the spot, on the spot and to the left then right of the spot. I did the same measurements for the other can in similar places. I repeated these measurements three times. I used the laser pointer spot to judge where the temperature measurement was being made.
As a side issue, I also monitored the surface temperatures of the cans while they were heating on the stovetop. The IR temperatures were erroneously influenced by what I assumed to be infrared radiation from the stovetop. I have noticed this effect when monitoring temperatures of stove components that are adjacent to other much hotter surfaces.
Results of such experiments in my experience never work out as I may imagine. The temperatures were recorded according to the above positions above, left, on, right (degree C).
Sensing position above, left, on, right
Spot rendered can, 80, 42, 81, 35
Fully rendered can, 89, 90, 90, 89
The shiny bean can gave lower temperatures measurement when taken at the side off the refractory spot. The ‘above’ reading was high and this may be because of errors in aim with the laser pointer?
Clearly the refractory coating increased the measured temperature and probably did this by improving the surface emissivity.
Stove flue pipe emissivity/temperature test
I put refractory render spots along the length of a roll up flue pipe that is made of 0.1mm thick stainless steel foil.
As may be expected, the measurements were variable, but on average the temperature from the spot was about 90-95C and from beside the spot it was about 60C.
Again at a temperature of approximately 100C the refractory coating increased the measured temperature to become much closer to the actual temperature. It probably did this by having an improved surface emissivity.
Dome Stove cooktop emissivity/temperature test
I put a band of refractory render across my beautiful new and shiny Simple Dome Stove cooktop.
The big surprise. Now if you have not already peeked at the photos, what would you expect the colour of the refractory coated band to be during the test firing? Have a look below.
‘The temperatures on the top of my stoves are a flame-dance. Nothing stays the same for long.” I ran the stove at full power and it became visibly quite red (except for the banded area). I measured the stovetop temperature at its highest point within the band and then measured it and adjacent points beside the band. I observe 500C in the band while only 300C on the adjacent surface. “A striking result, but as the famous TV professor Julius said; Why is it so? And I would add what does it mean?”
Again improving the surface emissivity appears to make the temperature measurement by IR thermometer more accurate. The 500C IR thermometer measurement was in general agreement with the barely visible red colour, within the band, being about 500C.
The emissive coatings conundrum
A high emissivity surface on a stove is supposed to make it radiate much more heat.
However, we have quite a conundrum here. If you trust my eyeballing of the shiny stovetop as being quite visibly red. Then it must be 700C according to the above chart ,although only 300C or so by the inaccurate IR thermometer. The refractory render band reduced the red colour significantly.
Unlikely as it would seem, the reduced redness of the band could be because it higher emissivity is radiating so much more heat and it becomes 150-200C cooler than the surrounding metal?
conclusion
It seems to me, that a thin emissive surface coating can greatly improve the accuracy of a cheap IR thermometer temperature measurement.
However, the coating at the same time appears to lower the surface temperature that I am trying to measure.
The test also will inspire me to to make an even higher emissivity coating by adding some more black refractory minerals into the already complex mix. This may be able to make the IR temperature better and also make the stove abetter radiator.
“In the meantime, I would appreciate any constructive comment on my conundrum by anyone who could confirm my conjecture or help me find a better explanation.”
Tim
