7 ms·

No, it's _precise_ down to 0.1mm. There's no way I'm trusting it's _accurate_ down to 0.1mm. Just the mere thermal expansion coefficient of abs, 100 * 10^-6 m/

No, it's _precise_ down to 0.1mm. There's no way I'm trusting it's _accurate_ down to 0.1mm. Just the mere thermal expansion coefficient of abs, 100 * 10^-6 m/(m C), is enough to tell you that across a very conservative range of 50 degrees the full scale measurement of 10 cm will expand by L0 * alpha * DeltaT = 0.1m * 100 * 10^(-6) * (50 C) = 0.5mm. (Units don't check out, I just added them here and there so you know what's what). Even if you use fiber-glass reinforced ABS that's still 31% of that, 0.155mm > 0.1mm. And who knows if the digital scale is even printed so it's accurate at any point in time - for all you know they just took a random other inaccurate tool to check against. And this does not account for any mechanical slop or for deformations from using what's essentially a slightly colder bar of chocolate to clamp down on hard metal. Accuracy: if you measure something with this tool, and then with a standard, it's going to come up with the same number, "down to 0.1mm of accuracy". As we've shown above, this thing is not accurate. Precision: if you measure something with this tool, and then measure the same thing with this tool again later on, then it's going to shoe you the same number. And even here, this tool isn't really "precise", but we can assume that for all intents and purposes you won't be using it to measure things years apart (wear and tear can change its shape easily) or during massive temperature swings. Moral of the story: Accuracy /= Precision Edit: at least tools made out of chocolate can be eaten https://www.thisiswhyimbroke.com/edible-chocolate-tools/ https://www.thisiswhyimbroke.com/edible-chocolate-tools/

stdbrouw 4y ago
In your eagerness to show the world how well you know the difference between accuracy and precision, the point the OP was trying to make seems to have flown right over your head. For DIY, you often need to make measurements that are a little bit more accurate than 1 mm, but not much, as well as measurements where 1 mm is plenty accurate but you need a reliable way to measure it, provided by the inner and outer jaws of the calipers. Even cheap calipers are excellent for this purpose, and the point is that people should not be scared off from them because they are not calibrated. Also, no, 50C is not a very conservative range for that purpose, that would be ±5C around the steady 20C temperature inside the average house.
- boltzmann-brain 4y ago
  I agree that there are times when you don't need absolute accuracy. I have plastic calipers myself for crappy jobs. But I don't say that they're accurate to 0.1mm because they're not. > Also, no, 50C is not a very conservative range for that purpose, that would be ±5C around the steady 20C temperature inside the average house. That is true until you realize many people keep their tools in the garage or in a cellar or an attic, where the temperature can go anywhere between -20C and +45C. Then you measure a small nut or screw as often used in, exactly, home appliances, and between the thermal coefficient, slop, bad geometry, and an inaccurate digital potentiometer you arrive at 4.8mm instead of 4.1mm and you order a 5mm replacement for your 4mm screw and have to wait another week or two after you figure that out. I've had this exact story happen to me in the past, and my tools were stored in, as you say, "±5C around the steady 20C temperature inside the average house" and not out of plastic - just still not accurate enough.
  - zihotki 4y ago
    For measurement of small nuts you apply an error calculated for 10cm. If the nut is small, the error will be also small, not 0.5mm, but something like 0.02-0.03. So if you measure 4.0-4.03 you won't make a mistake which nut it is. You can live with that as a DIY-er unless you need to measure 10cm+ nuts with high precision, I doubt that.
    - boltzmann-brain 4y ago
      No, that's incorrect, because for small measurements the slop as well as the influence of bad geometry is going to be larger. Lack of accuracy is a problem that multiple issues contribute to. I don't know where you got your numbers from. I doubt it's anything other than you thought the numbers just looked fine and illustrated your point.
walrus01 4y ago
If I was getting into the level of measuring things where the temperature coefficient mattered at all, other than a standard 20C room temperature, I sure as hell wouldn't be using a $9 plastic caliper from Amazon. My point was that the ultra cheap caliper is suitable for measuring basic things where you can't get a good reading using just a steel ruler with 1mm increment measurements on it by eyeball, and maybe you might want to measure something a bit more precisely than 1mm increments. And it might be suitable for basic home DIY purposes. As I said, if I wanted a serious caliper I'd go spend $250 on one. A cheap caliper like this, for example, might be suitable to tell me if a stripped copper wire's diameter really matches 23AWG or if it is 24AWG which is hard to see by eyeball.

stdbrouw 4y ago

In your eagerness to show the world how well you know the difference between accuracy and precision, the point the OP was trying to make seems to have flown right over your head. For DIY, you often need to make measurements that are a little bit more accurate than 1 mm, but not much, as well as measurements where 1 mm is plenty accurate but you need a reliable way to measure it, provided by the inner and outer jaws of the calipers. Even cheap calipers are excellent for this purpose, and the point is that people should not be scared off from them because they are not calibrated. Also, no, 50C is not a very conservative range for that purpose, that would be ±5C around the steady 20C temperature inside the average house.

boltzmann-brain 4y ago
I agree that there are times when you don't need absolute accuracy. I have plastic calipers myself for crappy jobs. But I don't say that they're accurate to 0.1mm because they're not. > Also, no, 50C is not a very conservative range for that purpose, that would be ±5C around the steady 20C temperature inside the average house. That is true until you realize many people keep their tools in the garage or in a cellar or an attic, where the temperature can go anywhere between -20C and +45C. Then you measure a small nut or screw as often used in, exactly, home appliances, and between the thermal coefficient, slop, bad geometry, and an inaccurate digital potentiometer you arrive at 4.8mm instead of 4.1mm and you order a 5mm replacement for your 4mm screw and have to wait another week or two after you figure that out. I've had this exact story happen to me in the past, and my tools were stored in, as you say, "±5C around the steady 20C temperature inside the average house" and not out of plastic - just still not accurate enough.
- zihotki 4y ago
  For measurement of small nuts you apply an error calculated for 10cm. If the nut is small, the error will be also small, not 0.5mm, but something like 0.02-0.03. So if you measure 4.0-4.03 you won't make a mistake which nut it is. You can live with that as a DIY-er unless you need to measure 10cm+ nuts with high precision, I doubt that.
  - boltzmann-brain 4y ago
    No, that's incorrect, because for small measurements the slop as well as the influence of bad geometry is going to be larger. Lack of accuracy is a problem that multiple issues contribute to. I don't know where you got your numbers from. I doubt it's anything other than you thought the numbers just looked fine and illustrated your point.

walrus01 4y ago

If I was getting into the level of measuring things where the temperature coefficient mattered at all, other than a standard 20C room temperature, I sure as hell wouldn't be using a $9 plastic caliper from Amazon. My point was that the ultra cheap caliper is suitable for measuring basic things where you can't get a good reading using just a steel ruler with 1mm increment measurements on it by eyeball, and maybe you might want to measure something a bit more precisely than 1mm increments. And it might be suitable for basic home DIY purposes. As I said, if I wanted a serious caliper I'd go spend $250 on one. A cheap caliper like this, for example, might be suitable to tell me if a stripped copper wire's diameter really matches 23AWG or if it is 24AWG which is hard to see by eyeball.