why is the US not Metric

[tooki]

You might as well ask when North America will join the 230 V world :-D Lots of advantages there (~3 kW available for common domestic appliances, lower final circuit currents leading to fewer fires and higher energy efficiency, eventual death of split-phase and weird stuff like high leg delta).

I would rather see the world adopting low-voltage DC residential electric circuits for lighting and electronics. At home I have about a hundred devices ranging from LED lamps to phone charges that don’t really need 120V, each with its own mains-operated LED driver or isolated SMPS (what a waste!), and only a handful appliances that genuinely benefit from 120VAC and two-phase 240VAC supply.

Though I understand the thinking behind that, that that would simply shift the waste elsewhere in the system. Suppose we used the 5V standard from regular USB. A desktop computer might need 300W, which is 60A at 5V. That's some seriously thick wiring.

A standard US residential circuit can provide 1800W (120Vx15A). At 5V, that's 360A, which would require insanely thick wire in the walls (over 100x as thick as what we can use at 120V, according to the national electrical code). Even if we used 24V, that'd still be 75A, still needing wire many times thicker than what we use now. With copper prices higher than ever, that makes absolutely no sense.

And the connectors needed for that would be crazy, as would the circuit breakers and switches, all of which would need to be able to handle massive arcing.

Rather than reducing the voltage of the power in the house, what we need to do is to raise the voltage of the devices themselves. This is precisely what USB Power Delivery does, allowing devices to request up to 20V, so that they can carry far greater power over thinner cable.

[vad]

You might as well ask when North America will join the 230 V world :-D Lots of advantages there (~3 kW available for common domestic appliances, lower final circuit currents leading to fewer fires and higher energy efficiency, eventual death of split-phase and weird stuff like high leg delta).

I would rather see the world adopting low-voltage DC residential electric circuits for lighting and electronics. At home I have about a hundred devices ranging from LED lamps to phone charges that don’t really need 120V, each with its own mains-operated LED driver or isolated SMPS (what a waste!), and only a handful appliances that genuinely benefit from 120VAC and two-phase 240VAC supply.

So instead you want the greater expense and complexity of switching and protecting DC circuits and, translating to actual standards, the losses involved in ELV distribution.

Yes, this makes much sense.

Hell yes, if that would prevent LED lights from flickering and LED driver popping because of overheating. Switching and protecting 12VDC / 15A branch circuit is not that much expensive than switching and protecting 120VAC / 15A circuit.

[vad]

You might as well ask when North America will join the 230 V world :-D Lots of advantages there (~3 kW available for common domestic appliances, lower final circuit currents leading to fewer fires and higher energy efficiency, eventual death of split-phase and weird stuff like high leg delta).

I would rather see the world adopting low-voltage DC residential electric circuits for lighting and electronics. At home I have about a hundred devices ranging from LED lamps to phone charges that don’t really need 120V, each with its own mains-operated LED driver or isolated SMPS (what a waste!), and only a handful appliances that genuinely benefit from 120VAC and two-phase 240VAC supply.

Though I understand the thinking behind that, that that would simply shift the waste elsewhere in the system. Suppose we used the 5V standard from regular USB. A desktop computer might need 300W, which is 60A at 5V. That's some seriously thick wiring.

A standard US residential circuit can provide 1800W (120Vx15A). At 5V, that's 360A, which would require insanely thick wire in the walls (over 100x as thick as what we can use at 120V, according to the national electrical code). Even if we used 24V, that'd still be 75A, still needing wire many times thicker than what we use now. With copper prices higher than ever, that makes absolutely no sense.

And the connectors needed for that would be crazy, as would the circuit breakers and switches, all of which would need to be able to handle massive arcing.

Rather than reducing the voltage of the power in the house, what we need to do is to raise the voltage of the devices themselves. This is precisely what USB Power Delivery does, allowing devices to request up to 20V, so that they can carry far greater power over thinner cable.

My point is - majority of the devices in a typical modern home are low-power low-voltage devices. Having, let say 12VDC / 10A branches in my home, in addition to existing 120VAC and two-phase 240 VAC circuits makes a lot of sense for powering such devices.

[Zero999]

I tend to use Fahrenheit for higher weather temperatures and Celsius for the lower ones.

In the UK we still have road distances and speed limits in miles despite having most other things in SI units... strange really.
... and when it is really hot the tabloid papers say... "Oooo what a scorcher <big number>F"... I seriously doubt many people understand F in the UK now... not me anyway. 

That's true, especially with higher temperatures. The BBC gave the temperature in both Fahrenheit and Celsius, when the heatwave in the summer just gone, broke the UK temperature, by reaching  38.7°C (101.7°F), yet the record breaking temperature in winter, at 21.2°C was only given in Celsius.
https://www.bbc.co.uk/news/uk-49157898
https://www.bbc.co.uk/news/uk-47374936

I wonder if it's because 101.7°F sounds more impressive than 70.2°F?

I think we should all use Kelvin for temperature. Celsius is no better than Fahrenheit because both scales have negative numbers which are totally unnecessary.

[tooki]

My point is - majority of the devices in a typical modern home are low-power low-voltage devices. Having, let say 12VDC / 10A branches in my home, in addition to existing 120VAC and two-phase 240 VAC circuits makes a lot of sense for powering such devices.

I understand the appeal, but a 12V/10A circuit is just 120W, and you would easily bog that down with just a few lights and things. (For context, a mere 3 foot segment of LED tape is 1A.) I think people don't realize how much current many of their low-voltage devices draw! A cellphone charging is around 12W (which can be 2.4A at 5V, or sometimes less current at higher voltage), another amp at 12V. Little LED desk or night table lamp? Another half an amp. Next thing you know, your "low power" circuits are overloaded.

What does make some sense is to have low-voltage AC->DC power supplies that can handle multiple devices. I've been doing this, by replacing power strips full of individual USB chargers with 6- and 10-port chargers.

P.S. US 120/240V circuits aren't two-phase, it's a split-phase system. The 120V circuits are created from a single 240V phase, by using a center-tapped transformer.

[CatalinaWOW]

You can blame pride for the US attachment to traditional units, and you would be at least partly right, but why do you care one way or another.  The main reason that we haven't changed is that the primary intrinsic benefit of the metric system is some convenience in the relationships of quantities to each other.  That wasn't sufficient to overcome the inconveniences and real costs of changing.  The other benefit was interchangeability with other market's, and has nothing to do with the merits of either system.  At one point that benefit was nonexistent for the US.  Being isolated from the dominance of Imperial units had benefit to some and influenced their change to metric.  As metric has become dominant worldwide that has changed and currently is a strong benefit for metric in the US and we are changing, slowly.  Again, the only reason anyone outside the US should care is that you want to buy from or sell to us, but find the reasons for those desires don't overwhelm your distaste for an unfamiliar measurement system.  Or are intellectually incapable with the aid of Google of dealing with scale factors other than ten.

[Monkeh]

You might as well ask when North America will join the 230 V world :-D Lots of advantages there (~3 kW available for common domestic appliances, lower final circuit currents leading to fewer fires and higher energy efficiency, eventual death of split-phase and weird stuff like high leg delta).

I would rather see the world adopting low-voltage DC residential electric circuits for lighting and electronics. At home I have about a hundred devices ranging from LED lamps to phone charges that don’t really need 120V, each with its own mains-operated LED driver or isolated SMPS (what a waste!), and only a handful appliances that genuinely benefit from 120VAC and two-phase 240VAC supply.

So instead you want the greater expense and complexity of switching and protecting DC circuits and, translating to actual standards, the losses involved in ELV distribution.

Yes, this makes much sense.

Hell yes, if that would prevent LED lights from flickering and LED driver popping because of overheating. Switching and protecting 12VDC / 15A branch circuit is not that much expensive than switching and protecting 120VAC / 15A circuit.

LED lights flicker because they're cheap.

A 12V 15A circuit over a practical length, say the 15m or so to get all round my office, would require something like 10mm² (that's north of 8AWG to you..) to keep losses practical. This is clearly ridiculous. And that's one supply per room. The total run of my (32A..) circuit here is probably approaching 35m. That's a lot of copper to do it in 12VDC - and I'd need as many as 12 at 12V 15A to supply this room!

48V is rather more practical in terms of distribution but still has a number of complexities - a PoE switch alone costs about the same as having a 240V 20A circuit installed - one gets you about 350W for very specific devices, the other, 4.8kW for, uh, anything you like.

Slightly more practical may be a standardised connector for supplying say, 12V or 24V at around 2A (and maybe a 5A version) coupled with power supplies which mount in place of outlets.

[BravoV]

You might as well ask when North America will join the 230 V world :-D Lots of advantages there (~3 kW available for common domestic appliances, lower final circuit currents leading to fewer fires and higher energy efficiency, eventual death of split-phase and weird stuff like high leg delta).

I would rather see the world adopting low-voltage DC residential electric circuits for lighting and electronics. At home I have about a hundred devices ranging from LED lamps to phone charges that don’t really need 120V, each with its own mains-operated LED driver or isolated SMPS (what a waste!), and only a handful appliances that genuinely benefit from 120VAC and two-phase 240VAC supply.

So instead you want the greater expense and complexity of switching and protecting DC circuits and, translating to actual standards, the losses involved in ELV distribution.

Yes, this makes much sense.

Hell yes, if that would prevent LED lights from flickering and LED driver popping because of overheating. Switching and protecting 12VDC / 15A branch circuit is not that much expensive than switching and protecting 120VAC / 15A circuit.

LED lights flicker because they're cheap.

A 12V 15A circuit over a practical length, say the 15m or so to get all round my office, would require something like 10mm² (that's north of 8AWG to you..) to keep losses practical. This is clearly ridiculous. And that's one supply per room. The total run of my (32A..) circuit here is probably approaching 35m. That's a lot of copper to do it in 12VDC - and I'd need as many as 12 at 12V 15A to supply this room!

48V is rather more practical in terms of distribution but still has a number of complexities - a PoE switch alone costs about the same as having a 240V 20A circuit installed - one gets you about 350W for very specific devices, the other, 4.8kW for, uh, anything you like.

Slightly more practical may be a standardised connector for supplying say, 12V or 24V at around 2A (and maybe a 5A version) coupled with power supplies which mount in place of outlets.

Wait till you meet a clueless one, that insist USB power delivery that can put 100W , is the future that will replace this so called annoying/unsafe hundreds of volt AC mains power, yeah, I met one in real life. 

[SparkyFX]

Virtually every Japanese camera made in the past 40 years has a tripod socket with a 1/4-20 thread.  That's 1/4 inch, 20 threads per inch.

Yep, only reason for me to get an imperial tap and die. But actually they were not that much expensive or anything, just another one for the set.

Thread pitches and forms are somewhat separate from SI-units, as there are plenty of definitions and technical uses (you need self-sealing tapered threads, threads to transmit higher forces (ACME, buttress), fine pitches and so on). The point is: there is a definition and that's what counts. It might give some people headaches to convert or manufacture it when needed, but can't really blame that to the existence of a second or third system, you need to be able to calculate things anyway.

NIST however says that their definitions are rooted in SI-units anyway. So it is about the common, everyday use.

[Monkeh]

Wait till you meet a clueless one, that insist USB power delivery that can put 100W , is the future that will replace this so called annoying/unsafe hundreds of volt AC mains power, yeah, I met one in real life. 

I knew I was forgetting something.

There's my 'slightly more practical' idea already on the market. I'll never be rich at this rate.

[Kleinstein]

The US are at least using metric inches already for quite some time.  That is an inch is exactly 25.4 mm and not the measured widths of the thumb or whatever body part of a local chief.

[joeqsmith]

Because 402.336 meters just sounds stupid.   So does 304.8 meters. 

I was taught the metric system in grade school.  Throughout my professional career, I have mostly used metric.   I suspect like most countries we use a mix of different units.   

[vad]

My point is - majority of the devices in a typical modern home are low-power low-voltage devices. Having, let say 12VDC / 10A branches in my home, in addition to existing 120VAC and two-phase 240 VAC circuits makes a lot of sense for powering such devices.

I understand the appeal, but a 12V/10A circuit is just 120W, and you would easily bog that down with just a few lights and things. (For context, a mere 3 foot segment of LED tape is 1A.) I think people don't realize how much current many of their low-voltage devices draw! A cellphone charging is around 12W (which can be 2.4A at 5V, or sometimes less current at higher voltage), another amp at 12V. Little LED desk or night table lamp? Another half an amp. Next thing you know, your "low power" circuits are overloaded.

What does make some sense is to have low-voltage AC->DC power supplies that can handle multiple devices. I've been doing this, by replacing power strips full of individual USB chargers with 6- and 10-port chargers.

P.S. US 120/240V circuits aren't two-phase, it's a split-phase system. The 120V circuits are created from a single 240V phase, by using a center-tapped transformer.

In the US we typically have a separate 15A branch circuit per bedroom. So 120W low-voltage circuit per bedroom is plenty enough for LED lights - with 100 lpw devices you can light up bedroom with 12,000 lumens - equivalent to 15 x 60W incandescent bulbs - far more than anyone would need.

240V is a two-phase system with a neutral wire and two hot wires running at 180 degrees phase shift.

[Monkeh]

My point is - majority of the devices in a typical modern home are low-power low-voltage devices. Having, let say 12VDC / 10A branches in my home, in addition to existing 120VAC and two-phase 240 VAC circuits makes a lot of sense for powering such devices.

I understand the appeal, but a 12V/10A circuit is just 120W, and you would easily bog that down with just a few lights and things. (For context, a mere 3 foot segment of LED tape is 1A.) I think people don't realize how much current many of their low-voltage devices draw! A cellphone charging is around 12W (which can be 2.4A at 5V, or sometimes less current at higher voltage), another amp at 12V. Little LED desk or night table lamp? Another half an amp. Next thing you know, your "low power" circuits are overloaded.

What does make some sense is to have low-voltage AC->DC power supplies that can handle multiple devices. I've been doing this, by replacing power strips full of individual USB chargers with 6- and 10-port chargers.

P.S. US 120/240V circuits aren't two-phase, it's a split-phase system. The 120V circuits are created from a single 240V phase, by using a center-tapped transformer.

In the US we typically have a separate 15A branch circuit per bedroom. So 120W low-voltage circuit per bedroom is plenty enough for LED lights - with 100 lpw devices you can light up bedroom with 12,000 lumens - equivalent to 15 x 60W incandescent bulbs - far more than anyone would need.

And once again, 15A at 12V requires a stupidly large cable - unless you put the 12V supply at the load. Which has already been done for decades - you too can go out and buy 12V constant-voltage drivers and lamps.

240V is a two-phase system with a neutral wire and two hot wires running at 180 degrees phase shift.

No.. it's definitely single-phase centre-tapped.

[wraper]

240V is a two-phase system with a neutral wire and two hot wires running at 180 degrees phase shift.

No.. it's definitely single-phase centre-tapped.

What makes "phases" must be 120 degrees apart from each other?

It does not need to be necessarily 120^o as you can use phase count other than 3. But 180^o makes little sense because it would be not better than single phase with double voltage.

[CatalinaWOW]

I think what he is trying to say is that there is no benefit to motors from having a 180 degree phase.  Which is the only area I am aware of that benefits from 120 spacing in any meaningful way.  That one is however a big benefit.

[james_s]

Sigh... Until American children are taught metric the Imperial system will persist, simple as that. It's not about what is better, it is about what is familiar, and in the US the Imperial system is by far the most familiar one.

We were taught metric in school, and contrary to what many think, quite a few Americans are familiar with metric units, anyone in engineering, science, firearms, automotive, etc is going to encounter metric measurements.

The USA has a mix of metric and imperial units, just like Canada and the UK which are the two other countries I've visited. Of course which things are metric and which are imperial is different in each of them. UK even uses some esoteric old units like stones when referring to a person's weight. Doesn't really matter, you learn what the units are and it's easy enough to convert, especially with modern technology.

If I were starting now, metric would be an easy choice. With >200 years of using imperial units there's a lot of expense and challenges in changing now, so a good deal of inertia. Eventually I suspect metric will get more and more common as everything gets globalized.

[james_s]

I think what he is trying to say is that there is no benefit to motors from having a 180 degree phase.  Which is the only area I am aware of that benefits from 120 spacing in any meaningful way.  That one is however a big benefit.

Another very big benefit was iron transformer xray machines. With a single phase machine, the anode voltage follows the half wave rectified sine and is only near the desired kVp for a brief moment at each peak. Outside of that time the anode voltage is lower resulting in lower energy photons that contribute more to radiation absorbed dose and less to creating an image on film.

With a 3 phase machine the sine peaks overlap and the voltage never drops near zero so you can use a much shorter exposure and much lower absorbed dose for the same image.

The modern stuff is all going to inverter drives that give the tube clean DC from a high frequency SMPS, so with these the only real benefit of 3 phase is the much smaller bulk filter capacitors required.

[vad]

My point is - majority of the devices in a typical modern home are low-power low-voltage devices. Having, let say 12VDC / 10A branches in my home, in addition to existing 120VAC and two-phase 240 VAC circuits makes a lot of sense for powering such devices.

I understand the appeal, but a 12V/10A circuit is just 120W, and you would easily bog that down with just a few lights and things. (For context, a mere 3 foot segment of LED tape is 1A.) I think people don't realize how much current many of their low-voltage devices draw! A cellphone charging is around 12W (which can be 2.4A at 5V, or sometimes less current at higher voltage), another amp at 12V. Little LED desk or night table lamp? Another half an amp. Next thing you know, your "low power" circuits are overloaded.

What does make some sense is to have low-voltage AC->DC power supplies that can handle multiple devices. I've been doing this, by replacing power strips full of individual USB chargers with 6- and 10-port chargers.

P.S. US 120/240V circuits aren't two-phase, it's a split-phase system. The 120V circuits are created from a single 240V phase, by using a center-tapped transformer.

In the US we typically have a separate 15A branch circuit per bedroom. So 120W low-voltage circuit per bedroom is plenty enough for LED lights - with 100 lpw devices you can light up bedroom with 12,000 lumens - equivalent to 15 x 60W incandescent bulbs - far more than anyone would need.

And once again, 15A at 12V requires a stupidly large cable - unless you put the 12V supply at the load. Which has already been done for decades - you too can go out and buy 12V constant-voltage drivers and lamps.

No, it does not. 10 AWG would be sufficient for 10A for residential wiring.
10 AWG copper wire has resistance of 0.9989 Ohm per 1000 feet at room temperature. 10A through 75-feet long circuit branch gives 1.5V voltage drop both ways, which is acceptable.

[Monkeh]

No, it does not. 10 AWG would be sufficient for 10A for residential wiring.
10 AWG copper wire has resistance of 0.9989 Ohm per 1000 feet at room temperature. 10A through 75-feet long circuit branch gives 1.5V voltage drop both ways, which is acceptable.

And 10AWG is ridiculous for a 120W circuit! Nor is >10% drop acceptable by any reasonable standard. How on earth is this better than putting a 12V supply at the point of load, as you can already do? It's more expensive, harder to work with, and less efficient!

[VK3DRB]

here in Australia everything is Metric,  like most of the world.

WRONG.

Most engineers in Australia, including me, still use the term "0402", "0603", "0805" etc for surface mount resistor sized. There is some crossover in name sizes between imperial and metric which just adds confusion. But I can live with this.
 
But what is really annoying is engineers in Australia who use an imperial grid for schematic symbol grid lines in Altium. So when my schematic symbols have a 2.50mm grid pattern and I have to use them to modify some imperialist's 0.1" grid, the bloody pins don't line up.

Australians should abandon the antiquated imperial measurements once and for all. And use kPa rather that PSI.

I think the only countries that have not moved to metric are Liberia, Burma and, you guessed it... the USA

[vad]

No, it does not. 10 AWG would be sufficient for 10A for residential wiring.
10 AWG copper wire has resistance of 0.9989 Ohm per 1000 feet at room temperature. 10A through 75-feet long circuit branch gives 1.5V voltage drop both ways, which is acceptable.

And 10AWG is ridiculous for a 120W circuit! Nor is >10% drop acceptable by any reasonable standard. How on earth is this better than putting a 12V supply at the point of load, as you can already do? It's more expensive, harder to work with, and less efficient!

10% power loss in the worst case scenario (the maximum distance from distribution box) is far better than typical SMPS losses at the point of load.

Yes, it would be more expensive. Let me guess... Wiring 10 low-voltage branches for lighting receptacles would be $200-300 more expensive in material costs than wiring ten 14AWG lighting branches. That’s a whooping $10 per year, when amortized over 27.5 years period. Jeez...

[vad]

here in Australia everything is Metric,  like most of the world.

WRONG.

Most engineers in Australia, including me, still use the term "0402", "0603", "0805" etc for surface mount resistor sized. There is some crossover in name sizes between imperial and metric which just adds confusion. But I can live with this.
 
But what is really annoying is engineers in Australia who use an imperial grid for schematic symbol grid lines in Altium. So when my schematic symbols have a 2.50mm grid pattern and I have to use them to modify some imperialist's 0.1" grid, the bloody pins don't line up.

Australians should abandon the antiquated imperial measurements once and for all. And use kPa rather that PSI.

I think the only countries that have not moved to metric are Liberia, Burma and, you guessed it... the USA

At least we drive on the right side of the road, unlike people of Botswana and Bangladesh.

[Monkeh]

No, it does not. 10 AWG would be sufficient for 10A for residential wiring.
10 AWG copper wire has resistance of 0.9989 Ohm per 1000 feet at room temperature. 10A through 75-feet long circuit branch gives 1.5V voltage drop both ways, which is acceptable.

And 10AWG is ridiculous for a 120W circuit! Nor is >10% drop acceptable by any reasonable standard. How on earth is this better than putting a 12V supply at the point of load, as you can already do? It's more expensive, harder to work with, and less efficient!

10% power loss in the worst case scenario (the maximum distance from distribution box) is far better than typical SMPS losses at the point of load.

And yet you've still got those losses back at the point of distribution - only a little lower in practice, with much higher transmission losses to account for.

Yes, it would be more expensive. Let me guess... Wiring 10 low-voltage branches for lighting receptacles would be $200-300 more expensive in material costs than wiring ten 14AWG lighting branches. That’s a whooping $10 per year, when amortized over 27.5 years period. Jeez...

And it's not going to save you $10 a year, and is going to waste materials which didn't have to be used. 14AWG x 10 circuits is already a stupid waste.

[Rick Law]

Years ago I remember being told that the reason the USA hadn't converted to metric was the cost of retooling, imagine the cost of replacing all those lathes and milling machines, plus the drill bits and everything else.

However............................

Lots of US companies now have their stuff made in Asia, who are quite comfortable working with either system, and the cost of conversion has now shrunk.

re: "Years ago I remember being told that the reason the USA hadn't converted to metric was the cost of retooling..."

I remember the same.  Pride or familiarity was not the main topic of discussion, but the cost vs benefit of having the same units.  The fight was between the importers, exporters, manufacturers, so on.  Manufacturers will be the ones particularly hard-hit in the change because their manufacturing machinery is expensive.

One must also consider, metric is as abstract as any other measurement and there is no intrinsic advantage in using it other than international trade.   Manufacturers/suppliers now adopted a different route to internationalization - with their own units that is even more abstract.  Now a "size 10" shirt from one manufacturer isn't the same size as a "size 10" from another.  Heck, it is not even the same "size 10" a year later from the same manufacturer.