Re:It's time for the US to use the metric system
by ddingus on Saturday January 09, 2016 @10:17AM
(#10871493)
Getting closer, but we aren't there yet.
Truth is, there is a TON of industrial machinery, tooling, fasteners, and more centered on Imperial units. Increasingly, new projects are happening in SI units too. This may start to snowball in the near future. Some of this is driven by global networking effects. Common units are a must, and global engineering efforts are less efficient with Imperial units.
Most of the new work in the major vertical industries is now Metric (SI). That's been a long, painful transition. They still carry a lot of ongoing, sustaining engineering and a supply chain supporting it and it's Imperial units however. I'm not sure the author really appreciates the weight of this legacy. It grows a little lighter every year, but it's still very significant and very costly.
In the US today, there are ~200 to 250K small to mid sized manufacturers. These remain all over the map, and they typically are the last to adopt new units and technology. It's not that they are lazy, or unable, though unable is a factor. It is all about that investment making sense for them. Often, it just doesn't. If what they have works and they can compete they will do that. Machinery investments are often significant and multi-decade long service life is expected.
Older machinery often doesn't work in dual units, and when it doesn't, conversions are error prone and confusing to everyone in manufacturing, which leads to the problems of dual units, duplicate drawings, etc... It's often easier to either continue with Imperial units, or target some switch over date to minimize the problem, and when older machinery remains capable, incentives to phase it out are low. Incentives to rework / upgrade it can be attractive though, depending.
It's a lot better than it was in the 80's and 90's, when I began to have some direct familiarity with manufacturing. Back then, I was a prototype mechanic for a while, and maybe 1 out of 100 jobs was metric. Back then, the first thing most often done was a conversion and check. It was rare to work in the Metric units directly. Most test and measure equipment in use, as well as the machines, were Imperial units and often didn't even have the capability to switch.
I just toured a nice mid sized shop recently. Almost all the equipment in the building was dual unit capable, with a nice mix of units seen in the drawings and data on the shop floor. Some older machines were running new, reasonably advanced controllers too. (Yes, I looked some of that over out of interest as I will be sending that shop some work in Metric.)
The company I manage has just started it's first couple of Metric projects. Reason: comparability with robotics, where Metric dominates.
If you ask me, my hunch is that we will see a more significant incentive to switch formally when robotic automation has matured in the small to mid-sized manufacturing market, and the last of the old guard machines has gone past it's useful service life and is phased out.
This may be accelerated by new, small, lean, localized manufacturing coming on line about now. It's growing more practical to setup shop and serve niche ecosystems and still make money. This is a ripe target for new business, new machinery, and that's all very highly likely to at least support metric (SI) units.
All of the useful software is dual unit now. That has been a non issue for well over a decade. Legacy software can take forever to cycle out of manufacturing, and it's finally happening en-masse. Really old stuff, DOS and older, is almost never seen now, meaning the tools from the 90's is in aggressive phase out and with that, the last of the single unit software solutions goes out the door.
Until then, there will be more than enough inertia backing Imperial units.
Maybe 10, 20 years at the earliest? We might be able to take a step, like preferred units, depending on how we want to handle our huge Imperial unit based legacy.
All that said, if we want a jobs program, just pass a mandate and fund the crap out it. We will pay hard for a decade, but we will have largely switched too. I don't see that happening, though it would be good for us to do, largely because we will carve out lots of stuff for cost reasons, and we will do that, because investment spending is poorly differentiated from other kinds of spending in the US right now, taking meaningful legislation and the funding needed for it off the table hard. A well thought out plan has the potential for real economic stimulus as well as some gains in terms of overall manufacturing capacity and efficiency. From a pure competitive standpoint, this kind of investment could be a great idea.
Barring some real change in how we think about spending overall (fat chance!), ongoing and incremental industry improvement driven by global network effects will nudge us there, bit by bit.
Truth is, there is a TON of industrial machinery, tooling, fasteners, and more centered on Imperial units. Increasingly, new projects are happening in SI units too. This may start to snowball in the near future. Some of this is driven by global networking effects. Common units are a must, and global engineering efforts are less efficient with Imperial units.
Most of the new work in the major vertical industries is now Metric (SI). That's been a long, painful transition. They still carry a lot of ongoing, sustaining engineering and a supply chain supporting it and it's Imperial units however. I'm not sure the author really appreciates the weight of this legacy. It grows a little lighter every year, but it's still very significant and very costly.
In the US today, there are ~200 to 250K small to mid sized manufacturers. These remain all over the map, and they typically are the last to adopt new units and technology. It's not that they are lazy, or unable, though unable is a factor. It is all about that investment making sense for them. Often, it just doesn't. If what they have works and they can compete they will do that. Machinery investments are often significant and multi-decade long service life is expected.
Older machinery often doesn't work in dual units, and when it doesn't, conversions are error prone and confusing to everyone in manufacturing, which leads to the problems of dual units, duplicate drawings, etc... It's often easier to either continue with Imperial units, or target some switch over date to minimize the problem, and when older machinery remains capable, incentives to phase it out are low. Incentives to rework / upgrade it can be attractive though, depending.
It's a lot better than it was in the 80's and 90's, when I began to have some direct familiarity with manufacturing. Back then, I was a prototype mechanic for a while, and maybe 1 out of 100 jobs was metric. Back then, the first thing most often done was a conversion and check. It was rare to work in the Metric units directly. Most test and measure equipment in use, as well as the machines, were Imperial units and often didn't even have the capability to switch.
I just toured a nice mid sized shop recently. Almost all the equipment in the building was dual unit capable, with a nice mix of units seen in the drawings and data on the shop floor. Some older machines were running new, reasonably advanced controllers too. (Yes, I looked some of that over out of interest as I will be sending that shop some work in Metric.)
The company I manage has just started it's first couple of Metric projects. Reason: comparability with robotics, where Metric dominates.
If you ask me, my hunch is that we will see a more significant incentive to switch formally when robotic automation has matured in the small to mid-sized manufacturing market, and the last of the old guard machines has gone past it's useful service life and is phased out.
This may be accelerated by new, small, lean, localized manufacturing coming on line about now. It's growing more practical to setup shop and serve niche ecosystems and still make money. This is a ripe target for new business, new machinery, and that's all very highly likely to at least support metric (SI) units.
All of the useful software is dual unit now. That has been a non issue for well over a decade. Legacy software can take forever to cycle out of manufacturing, and it's finally happening en-masse. Really old stuff, DOS and older, is almost never seen now, meaning the tools from the 90's is in aggressive phase out and with that, the last of the single unit software solutions goes out the door.
Until then, there will be more than enough inertia backing Imperial units.
Maybe 10, 20 years at the earliest? We might be able to take a step, like preferred units, depending on how we want to handle our huge Imperial unit based legacy.
All that said, if we want a jobs program, just pass a mandate and fund the crap out it. We will pay hard for a decade, but we will have largely switched too. I don't see that happening, though it would be good for us to do, largely because we will carve out lots of stuff for cost reasons, and we will do that, because investment spending is poorly differentiated from other kinds of spending in the US right now, taking meaningful legislation and the funding needed for it off the table hard. A well thought out plan has the potential for real economic stimulus as well as some gains in terms of overall manufacturing capacity and efficiency. From a pure competitive standpoint, this kind of investment could be a great idea.
Barring some real change in how we think about spending overall (fat chance!), ongoing and incremental industry improvement driven by global network effects will nudge us there, bit by bit.
Good as it gets right now kids.