I have a few oral thermometers at home, but they aren’t always consistent with each other. I have two oral and one touchless, but sometimes they can be off by a degree from each other. Regular thermometers can easily be tested against things with known temperatures like ice baths and boiling water, but oral thermometers may only register within a small range around body temp. Is there any way for me to create something at home with a consistent temperature in the 97-100 degree range that I could use to test my body thermometers?
All I can think of that’s easy is first to check the calibration of a regular thermometer as you say against ice slurry and boiling water (adjusting for elevation). Even if it’s off, if you know the error at freezing and boiling you are probably safe to assume that the error is close to linear for intermediate temperatures. So that thermometer (with adjustment for error) gives you a reference for setting up a water bath of known temperature for checking the oral thermometers.
I went through this recently with a sick child who would cycle temperatures between an indicated 99 and 107F. Two different oral thermometers and a forehead thermometer could differ by 3-4 degrees within minutes of each other.
I tested the oral thermometers by comparing them to a Thermapen cooking thermometer, which was verified as accurate in an ice bath and a boiling water bath. I’m not sure what you’ll do if you don’t have something which is known to be accurate to use as a reference thermometer.
I filled a small glass with hot tap water, and measured the temperature simultaneously with the the Thermapen and the two oral thermometers. It was about 102F. I found one of the oral thermometers was within .2 or so degrees of the Thermapen, and the other was 1+ degrees off.
I would trust the old style glass thermometers more than an electrical readout. Nothing to calibrate since they are based upon the size of the tube capillary and the composition of the liquid. Unless its been broken or allowed to freeze or boil there isn’t much that can go wrong with them if the glass hasn’t cracked
A Thermapen or something similar may be the least expensive thermometer you can rely on for calibrating other thermometers. Measuring body temperature using a simple thermometer leaves a lot of room for error though no matter what device you use, you’re just taking a quick sample at a convenient location, at least your body is trying to maintain a fairly consistent temperature all around though.
You can’t easily test their accuracy though in the range of human body temperature. You can easily tell if it’s in the ballpark for freezing and boiling temps, but what would you use for comparison around 100F?
As I said above - I think you can be reasonably certain if you calibrate to freezing and boiling that any error will be linear for intermediate temperatures.
But if we’re looking for accuracy to a degree or less, for boiling point you need to adjust for air pressure at your local elevation. Weather conditions make a difference too, but it’s not usually significant compared to elevation. These websites take you through it:
(1) Check weather conditions, your local barometric pressure (always stated at sea level equivalent):
https://origin.wpc.ncep.noaa.gov/basicwx/basicwx_ndfd.php
(2) Reduce that sea level pressure according to your local elevation, table here:
(3) Check the boiling point at that pressure, e.g.
https://www.omnicalculator.com/chemistry/Boliling-point
Or, of course, instead of steps (1) and (2) you can check the pressure directly with a barometer. A lot of watches have them.
If you play around with it, you can see that a change in elevation makes a much bigger difference in pressure compared to the normal range of pressure variation with the weather.
For glass bulb thermometer I wouldn’t be concerned about calibrating it because there is no way to adjust it. The design itself is based on physical properties of the bulb, the liquid and the inner diameter of the tube. Nothing a user could adjust if they wanted to. It may have some inherit inaccuracies within manufacturing tolerances but it will at least be consistent. If you are sick the most important thing is the delta between your normal temperature and what it is reading. That is whether your temp is rising or falling.
Another idea, although it won’t work immediately. Bring your home thermometer with you to your next in-person medical appointment and compare against their (presumably calibrated) thermometer. (They can also tell you if you’re not using an oral thermometer correctly.)
I do think that is the important part, consistency instead of great accuracy.
But if you want accuracy, and the OP asked for the correct temperature, then it should be better than you’ll find in some household thermometers. You could take several of these, find them all consistent, and within 1F degree of each other, but in the middle of their range not know which one is closest to the mark. That’s not very useful to information to determine if you’re having symptoms but your inconclusive information will be slightly more exact. After accounting for all factors it might even be imperceptibly more exact. That’s why precision, in the sense of consistency, is often more important than accuracy.
This is a long thread to study, and plenty of good things have been said, but let me just add:
Please don’t put a mercury-filled, glass-bodied, thermometer into water that is even barely simmering. I did that as a kid, the glass instantly shattered, and liquid mercury went everywhere.
If you have some other style, then go ahead and play. But work within the other caveats listed in posts above. And go ahead and tack on “metallic mercury isn’t THAT toxic” or “I did it digital so maybe …” in your nest post if you must. However, boiling water is only a method for checking the calibration of devices designed to handle that.
Are there ways I can test my regular thermometer at various temperatures so I can know how accurate and consistent it is for the intermediate temps? I have several cooking thermometers that I can test with freezing and boiling water, but what can I use for in between temps? Are there other common and safe substances which have melting/boiling points in between which I can use to calibrate my cooking thermometers? If I find my cooking thermometer is accurate at a variety of temps, then I’ll feel more confident it will be accurate when testing the body thermometers.
Even if the thermometer is properly calibrated, there could easily be more than a degree difference if you measure the temperature in the armpit, below the tongue or in the ear (all of which can be done). Hospitals often use thermal pens and ear temperature - which are fine but use disposable covers.
I agree not much can go wrong with a glass type thermometer. You could buy a thermal pen to check. Mixing equal amounts of ice and boiling water seems excessive (especially if you consider latent heats). Better to consider the temperature you get approximate and remember doctors are almost always much more concerned with symptoms than temperature (cough, vomiting, behaviour, etc.)
For cooking, easiest way to calibrate is to make bonbons, which have definable stages.
How stable are the melting points of fats? I see that coconut oil has a melting point of 78 degrees. Would that mean that it would be stable at 78 until it all melts?
A degree different one way or another from different body temperature thermometers isn’t significant enough difference in average household use to go through testing or calibration procedures, especially as elaborate as ones described. Also, outside of a professionally equipped lab with carefully controlled conditions, you’ll still get variances, just different ones.
Unless a physician has told you otherwise, choose the thermometer that is the easiest for you to use and use that one without comparing it to another one. The important thing to learn is my temp is X degrees different now than my normal temp. Take your temperature when you are feeling well. Note what it is. If it is 97.0 when you are feeling well and it is 99.0 when you are wondering if you are ill, then you need only to know that YOUR temperature is now 2 degrees higher than YOUR temp usually is.
I strongly caution against using any glass thermometer with mercury-too easy to break and then the mercury becomes an environmental health risk. That is why doctors and hospitals no longer use them. When discarding glass/mercury thermometers find out from local health dept or municipal recycling agencies the safe way to dispose of it.
Even with the same thermometer you will usually get different readings from the typical body areas, contactless skin (forehead or temple), ear, oral, armpit or rectal. If you are reporting a temperature to a health provider tell them the type (or method) of thermometer and the site used on the body.
Always read the the instruction pamphlet with the thermometer because doing it correctly each time will give the most accurate results.
If the temperature is being taken for ovulation prediction reasons a specific ovulation or basal thermometer should be used. Of course, these are available on Amazon. If your temperature concerns are relating to screening for Covid, temperature elevations are by no means a certainty with contagion-the absence of a temperature does not mean necessarily the absence of contagion.
You aren’t picking up the message here, Your thermometers are good enough, you just need to know they are consistent. You aren’t going to be able to calibrate a thermometer to any great accuracy based on the freezing and boiling points of any substances because you can’t control the temperature of a reasonable sized sample well enough. Your thermometers are fine, your doctor’s thermometer is not going to be significantly more accurate, and your body temperature isn’t that consistent or readily measurable with great accuracy.
@BippityBoppityBoo the professional nurse nailed it.
The other thing you can do is simply get a coffee cup, fill it with what feels to your hand like lukewarm water, and put all your body thermometers in it simultaneously. If the thermometers all agree the water’s between 95F & 105F AND one thermometer is more than 2 degrees off from the others, put the oddball in the nearest trash can. Repeat as necessary buying new one(s) if you want to have lots, or just culling your herd if that leaves you enough for your purposes.
Once you have a batch of thermometers that are each close enough to the others, then you can reliably use Bippity’s method. e.g.
I own two electronic thermometers. When I’m normal / healthy they read 97.7 & 97.6. That’s close enough that they’re interchangeable. If I grab either one and read 100 I know I’m running 2-and-a-smidgen degrees high.
That’s all the precise this gets, and that’s more than precise enough for health tracking.
Agreed, but I would still find it interesting from a home scientist perspective to be able to figure out how accurate and consistent my thermometers are. Whether or not they are good enough for simple temperature checks is not necessarily the only reason I’m asking. If there is some at home way to consistently achieve some temperature in the typical body range, I would find that useful and interesting even if it is silly and unnecessary.
Liquid in glass thermometers can drift for a variety of reasons. You could effectively calibrate an oral thermometer if you knew an offset for it, or better still if you had a calibration curve that showed slightly different offsets at different readings. It doesn’t have to have an adjustment capability built in.
The closest reference I can think of would be the melting point of gallium, 29.7646 °C or 85.5763 °F. It’s a reference standard for high precision thermometry. IIRC there are pure alkanes with melts closer to body temp but I think they make weak standards because they change phase slowly and have small latent heats.
Yes, here’s one: Icosane, C20H42, has a melt of 36 to 38 °C or 97 to 100 °F. Not a very easy melt to determine crisply, though.