Used a couple of US recipes recently and most of the ingredients are in cups, or spoons, not by weight. This is a nightmare to convert. Do Americans not own scales or something? What’s the reason for measuring everything by volume?
I keep using this example: In the wood shop, I’m going to cut a bridle joint. Requires cutting boards into thirds of their thickness. Metric lumber is often milled to 19mm thick. What’s a third of 19mm? You want to show me which line means 6.3333mm on a metric tape measure? US Customary lumber is milled to 3/4" thick. What’s a third of 3/4"? You want to show me which line means 1/4" on an inch tape measure? Now let’s cut a half-lap joint in that same lumber. In metric that works out to 9.5mm, there’s also no line on a typical metric tape measure for that. But there is a line for 3/8".
I’d much rather build furniture in inches than millimeters because in the wood shop I have to divide or multiply by powers of 2, 3 and 4 way more often than powers of 10. It is in this context that the inch standard which is subdivided by powers of two rather than ten arose, and it still works very well.
Metric users often correctly accuse Imperial or US Customary (though the two share names of units they are not identical) users of making excuses or relying on workarounds, in the context of woodworking joinery I find it’s the reverse. “Of course we don’t use 6.3333mm, you just know to cut the cheeks 6mm and the tongue 7mm. 6+6+7 is 19.”
I’ll grant you, doing stoichiometry in ounces and pounds would be a fucking nightmare. But woodworking joinery? Nah I’m doing that in fractional inches.
While neat, I believe your lumber example is misleading in the context of metric vs imperial. Woodworking is extremely imprecise compared to many other types of engineering and using that system for those problems may be ideal.
Deliberately using 1/3rd of 19mm to get 6.33333mm is not as a complex problem as it may look at first glance. 6.333mm IS 1/3rd of 19mm just with more precision. The nature of woodworking requires fairly large tolerances and .3333mm is likely within any tolerance range you would work with. Hell, even +/-3.333mm (10x) is probably within spec in many cases.
Your example conversion from 1/4in to 9.5mm is irrelevant unless you are working a project that is deliberately converting imperial to metric. If a project is designed in metric the measurements and reference points are going to be rounded to metric. The same goes for designs that are in imperial. While it’s possible to design identical pieces in each measuring system, it’s not ideal. Tolerance can compensate for most small differences and you will get two extremely similar pieces.
From your standpoint, everything has been imperial and you make design choices around how imperial works. It just makes sense to you. Design conversions from imperial to metric won’t make any sense and the “natural math” of each system is lost. If you were raised on metric, the same situation would apply I suppose.
You explained the biggest complaint of imperial as a positive: fractions. Pure math is just easier then fractions when working up and down ranges of precision. Divide 10cm by 2? 5mm. 5mm by 4? 1.25mm, etc… Problems like 19mm/3 are irrelevant because of allowable tolerance. Every exact measurement is not abstracted by a 16th or 8th or 32nd or 64th…
Admittedly, I am no woodworker. However, I am curious if someone from the EU could chime in on this problem from their perspective.
I didn’t convert a quarter inch to 9.5mm. First of all a quarter inch is somewhere around 6.5mm. I divided 19mm by 2. Just like I divided 3/4" by two to get 3/8".
I’m talking woodworking here, not carpentry. .3333mm is ~1/64", which is the maximum error I would allow in making a joint. I usually work well within that. Missing by 3.3333mm on a feature specified to be 6.3333mm is an abject failure. That’s an error of over 50%.
PVA wood glue contracts as it dries, so it doesn’t fill gaps well. The looser the fit, the weaker the joint. I aim for a friction fit. Most of the time with glue on the mating surfaces the joint must be tapped together with a mallet.
Your reading comprehension is what I’ve come to expect from Lemmy. I repeat myself for the slow kids at the back: because I often have to divide my work into halves, thirds or quarters and rarely into tenths, fractional inches in powers of two are more convenient in this application.
Remember folks, in metric, 5/2=2.5, 5/3=doesn’t matter because “tolerances.” That’s one of those excuses I was talking about.
Do keep trying to lecture me about something you don’t even slightly understand. It’s adorable.
Your reading comprehension is what I’ve come to expect from Lemmy.
Chill. We both wrote walls of text and there are going to be misunderstood details. If we want to talk about details, I called out my ignorance of woodworking and why imperial is likely good for what you are talking about.
My overall points, and I’ll summarize this time, is that:
-
Wood working (carpentry? Whatever.) is not exact.
-
Dividing 19mm by 3 is a weird example. Your example did a better job of highlighting a math peculiarity, TBH. (My first thought is that the cut was was going to account for any minor errors.)
-
Fractions suck. You are comfortable with them, but I see them as a useless layer of an outdated measuring system. We made our points, for and against. Cool.
-
A key point that I didn’t call out specifically is that imperial does not work in high degrees of precision easily without eliminating fractions. It’s possible, and vocalized, but not generally written. 1/1000" as a good example.
While I was awaiting your reply, I also thought of the abuse the imperial system has suffered over the years. A 2x4 is not a 2x4. In reloading (another hobby of mine), .300 actually means .308. .223 could mean .222, .223 or even .224. However, .222 always means .222. I am forced into imperial for safety and consistency reasons. (Don’t even get me started on ‘grains’, wherever the fuck that came from.) For some reason, the metric system is now mixed up in that field as well and it’s a mess.
The word “misleading” was chosen with purpose and doesn’t mean that you writing with malice. It seemed, true or not, that conversions got mixed up in this which would even confuse an MIT graduate.
I have to admit: That sounds pretty nice. Next time I build something with wood I will try to use inches. Thanks!
This is why I think the metric system should be redone in base 12. Or maybe base 36.
I mean, you can make the exact same argument the other way round.
My bed is made with boards of 27mm thickness. One third of that would be 9mm. Easy.
Also if you need precision, calipers go down to 50um (micrometer), 1/20th of a mm.
What’s half of 27mm?
We can keep doing that dance, it’s possible to find similar inconveniences in the fractional inch system, like “what’s a third of a whole inch?” but I find that within each system’s conventions (like using 19 and 27mm stock versus 3/4" or 1 1/2" stock) you’re less likely to run into them working in fractional inches. I think because the wood shop is just a fractional kind of place, I divide by two and three out there a lot.
The machine shop isn’t so much, which is why we tend to work in either thousandths of an inch or increasingly in metric. Most CNC machines will gladly accept both.
As for calipers: In the wood shop, I frequently use a set of dial calipers calibrated in 64ths of an inch. Especially with my thickness planer on which one full turn of the handwheel moves the cutter head 1/16", so the major, medium and minor marks on the caliper dial work out to a full, half and quarter turn on the handwheel. The analog display makes the relationship between the calipers and the tool very intuitive in a way that improves accuracy and repeatability largely by decreasing error.
I don’t really need precision beyond 1/64", but I do need to be able to tell if it’s a thin 64th or a fat 64th or a dead nuts 64th.
