why we are preferring 3pin plugs in some devices?

Anything with a metal case probably should have a ground pin, although it is not always true.

i think tri-pin is used in the case of heavy loads and double pin in the case of light loads as in heavy loads the earth fault current will be more when compared to light loads..

I would suggest that it is more to do with "heavy loads" having to get rid of a lot of energy - which makes complete double-insulation of any parts that could become live problematic. They usually have something that needs to come out of the innards, such as a (metal) rotating shaft, stream of fluid (water/air/refrigerant), wires, etc that could, potentially, either provide a conductive path or provide a hole through which a conductive path could be established. And/or, of course, "heavy loads" are far more likely to have metal parts, because of the temperature of operation or the amount of energy needing to be transferred.
Whereas light loads can get rid of the very small amount of energy that they consume through the layers of electrical insulation.
There is no automatic "earth fault current" with a two pin device. Such a current may not flow until a user touches it. At which point a fatal current could easily result. Whilst protection devices, such as GFCI, are certainly life-savers - the key thing about the design of two pin devices *should* be that any failure modes that present a shock hazard are extremely unlikely.
Yes, a "heavy load" is likely to have a much higher earth fault current capability. But it only needs a few hundred mA of earth fault current to kill - and that can come from a "light load" as easily as a "heavy load".
-- Sue

Sue, fault current is determined by branch circuit size and the type of fault, not the attached equipment. I think "light loads" use 2 pin plugs, simply because they are cheaper to make. Here in the US there are plenty of desk and floor lamps with 2 pin plugs and metal construction. I don't know why they are not considered a hazard. We don't even have fuses in the plugs like you folks.

Where have I said otherwise? "Heavy loads" generally need higher rating infrastructure. Which, in turn is likely to result in higher earth fault current capability. A "heavy load" is likely to have a much higher earth fault current capability - because it is connected to a higher rated branch circuit. But my point was that this is irrelevant to whether the circuit needs a third wire, or not. There is more than enough capability for electrocution from a "light load" appliance.
I consider them a hazard. Hence the *should* in, "the key thing about the design of two pin devices *should* be that any failure modes that present a shock hazard are extremely unlikely."
The UK used to allow similarly dangerous practices. I remember well plugging a smoothing iron into a two pin socket on a two way adapter plugged into a (ceiling) lamp fitting.
Never mind, I expect that the US will catch up with Europe, one day ;)
-- Sue

The normal US NEMA 5-15 receptacle will either be on a 15a or 20a circuit.As you pointed out the potential for electrocution is the same. Any difference in fire danger is minimal. I suppose if the tombstones were there to justify a change in U/L listing practices we would see 3 pin plugs on everything. The 3 pin receptacle has been standard in new construction here for 40 years.

We still allow unearthed appliances (we don't have a 2-pin mains outlet in common use, but plenty of appliances only have a 2-core flex). Such appliances have to be double insulated.
The concept of double insulated seems to exist in the US too, but the requirements for it don't seem to be the same, specifically don't seem to include the requirement for a double insulation barrier between live parts and a person, based on appliances I see marked as double insulated in the US which don't come close to being double insulated for Europe.

The present USA sockets have two flat blades and a round ground pin. In fact the blades are of different widths with the "hot" lead on the narrow blade. Two blade plugs were meant to be fitted so that any possibly exposed parts were connected to the "neutral" though the wider pin. This is most often the base of an Edison socket. The socket center contact was supposed to be connected to the narrow pin through the switch.
Many cheaper products, including lamps with Edison sockets, have two blade plugs with equal width blades.
I worked as a "broadcast engineer" many years ago, before three pin plugs and sockets were common. The first thing we did when a remote broadcast set-up put hum on the telephone line back to the studio, was to: "reverse the plug". Parts of "Europe" still have two-pin sockets with no earth pin. In some places they are not even "polarized".
I believe modern USA electrical codes have "caught up", unless you consider the 120/240 split phase and lack of loop wiring to be archaic!
The UK have done a splendid job of "catching up" since I traveled extensively there in the late 1960's. I plugged my two blade, (equal width), electric shaver plug into every conceivable kind of outlet using various "adapters". I remembered to set the switch to 240V and didn't destroy the shaver or myself. I changed rooms in a London hotel once, where they had the usual 240/120 isolation transformer, which I forgot to re-set to 120V after I'd set the razor to 120V. I "pulled the plug" just in time!
On a more serious note: I would like to see statistics comparing fatalities from electrocution versus those from electrical fires in both the USA and the UK.

They really ought to be grounded too. Years ago I had a halogen torchier in my office. I could swear I got a tingle several times when touching it, then one day I pulled the monitor cable off one of my PCs and in doing so the metal connector shell brushed the lamp and POW! It tripped the circuit, burned a chunk out of the connector, and left a small blob of melted metal on the lamp. I did some investigating and found that a wire had chafed where it runs through the threaded tube up at the top and the entire lamp was hot.

Can you give any examples? I'm in Canada, which has the same double-insulated appliances as the USA, and all the ones I've looked at really do have two insulation barriers. Tool housings are typically plastic. The output shaft or chuck is metal, but there's a non-conductive coupler between the motor armature and output gearsbox or shaft, or the armature is assembled with an insulating spacer between the armature shaft and the laminated armature magnetics.
Dave

They've been mentioned here in the past. It's difficult to get directly compariative data as the same incident may be very differently classified in different places. Classifications are also not always reliable or what you might expect. Around 5 years ago, I took a detailed look at UK figures for incidents in the home (not elsewhere). Fires are often classed as electrical just because no other cause can be found in the remains, and one fire was classed as electrical which was an open chip pan fire which happened to be on an electric stove when it caught fire. Although chip pan fires are not normally classed as electrical in the UK - they have their own dedicated catagory here, being the number 2 cause of house fires after smoking. So we probably over-count electrical fires in the UK.
The UK average figures per year for the 10 year period 1992-2002 are:
Fatal Non-fatal Installations (electric shock) 5 576 Appliances (electric shock) 14 1700 Fires (electrical) 25 590
although they were dropping over that period, so they're higher at the start than at the end of the 10 year period.
I don't have the US figures any more, but they were much higher per capita. Actually, the US per capita figures for just home incidents was well higher than our total figures including public and industrial incients. Electrical fires was very much higher in US homes.
The reason I investigated the UK figures was to respond to the proposed change in UK legislation to bring in what's now known as Part P -- the requirement for inspection of home wiring, which previously didn't exist in the UK. Prior to this, the incidents in the UK had been steadily declining over the previous 30 years. A number of us thought the legislation was unnecessary, and would be counter-productive. However, I don't think any of us expected the extent to which we would be proved to be correct -- electrical fatalities in England and Wales instantly doubled with the introduction of Part P. So, from being the second lowest in Europe and hence one of the safest in the world, we are now up nearer to (but still less than) the US figures per capita.

I really think most electrical fies in the US involve misuse of extension cords and cube taps or simply defective equipment plugged in but like you said, a lot of fires get pegged as electrical because they don't find any other obvious source. Unless arson is suspected most fire investigations are pretty superficial from what I have seen. Some guy at the fire department, trained investigator or not, checks a box on a form based on his first impression and that is "the cause". I do understand that when foul play is suspected and they call out the big guns, they can pinpoint the fire cause pretty well. That just doesn't happen that often, particularly if nobody got hurt.

I also wonder if it has to do with the fact that a lot of wiring in the US is much older, it's not uncommon to find original 1920s knob & tube wiring still in use, and often it has been added onto multiple times over the years. I still come across old fuse panels as well, replaced one last year in a friend's 1950s house with a modern breaker panel, I found a 30A fuse screwed into a socket connected to a 14 AWG circuit that should have had a 15. The wire was so hot the insulation was starting to melt in the panel.
I'm not sure why, but someone mentioned at one point that you don't tend to find such old stuff still being used in the UK.

Precisely why I'd like to see some reliable statistics.
Strange, because they're more electrical appliances in use at the end, especially portable ones.
That's surprising. I wouldn't be surprised at higher electrocution figures since the US has more swimming pools per capita, and the code for electrical outlets in bathrooms anf kitchens has only recently been strengthened.
Was it the increase in "do it yourself" wiring? I hope it wasn't the inspectors getting electrocuted! :-)
There is a proposed US code change to require "arc detecting" circuit breakers in many areas. This sounds like a good idea, until one sees the problem with implementing such a device. GFCIs, (RCDs), are now common, and and are required in dangerous areas by the new codes.
See: "Use and regulations differ widely from country to country. In Europe, the UK is the only country that does not mandate the use of RCDs (however this is due to change in July 2008 for most new installations). In contrast, Germany requres the use of RCDs on ALL sockets up to 20A which are for general use. This rule was introduced in June 2007 (DIN VDE 0100-410 Nr. 411.3.3). In the U.S., the National Electrical Code requires GFCIs in bathrooms, kitchens, garages, outdoor areas, crawl spaces, unfinished basements, near wet bars, swimming pools, and spas in residential construction."
I don't mean to re-ignite, (no pun intended), the 240V vs 120V argument!

One that amazed me was a two slot toaster (for toasting sliced bread, just in case it's known as something else somewhere;-)
The mains lead was a 2-core zip cord (figure-of-8 as it's more commonly known here) with a molded 2-prong reversable plug. OK, that's common in the US, but being only a single insulation layer with no sheath, isn't allowed in Europe for mains voltage. I assume double insulated doesn't apply to mains cords in the US, even for appliances which claim to be double insulated?
The toaster case was metal. The element was a bare wire threaded over an insulating backing. There are a few stiff metal guard wires in front of the element to prevent the bread touching it. There's nothing to prevent the element breaking and contacting the guard wires (e.g. burning out, or more likely damaged by a piece of toast jamming and being pried out with a fork or similar). Out of curiosity, I turn the toaster over, and it's marked double insulated on the bottom. Well, it never could be in Europe, as there just aren't 2 layers of insulation between live parts and you. I don't think we have any double insulated toasters here. The designs are identical to this US one, but would always have a 3-core sheathed flex and 3 pin plug, so in the event of the case contacting a live part, the case remains safe.
Looking at the steam iron next, again, same single insulated zip cord and molded on 2-prong plug. It is conceivable that the element really is double insulated from the sole plate, but I can't check. I wonder how reliable that insulation might be when someone has filled the thing under the tap (fawcet) and got water spilled into all the wrong places? Again, I'm really dubious this would meet EU standards -- irons here are normally 3-core earthed, but I do have a small double insulated travel iron, so that is possible.

If the initial data capture is inaccurate, there are none to be had no matter how you process the data subsequently.
There was a noticable effect around 1990, which was put down to a significant increase in HiFi/TV/Video appliances. However, the general trend would appear to be safety improving faster than the rate of increase of appliances (although that's a gross over- simplifiction).
No. DIY wiring has always been very popular here. The change in the law made it much more beurocratic, and expensive, to make safe DIY changes legally, and encourages temporary and less well designed changes. There was no evidence that DIY wiring was responsible for any of the incidents. Indeed there was much evidence that the incidents resulted from _not_ making changes and improvements to wiring when they were required. So a law to encourage temporary and less well designed changes, and to dissaude you making improvements properly, had exactly the effect many of us predicted it would, but much more drammatically than even we imagined beforehand.
I was reading that recently New Zealand reversed similar legislation they have had, and saw safety improve. I'm not familiar with local issues with New Zealand wiring, but given what we predicted and have subsequently seen here, that comes as no surprise to me.
I have to say that seems to me like the wrong solution to your arcing problem. I think you should look at the design and quality of your wiring accessories compared with other countries at similar economic levels. This just isn't a problem in most countries. An arc fault detector is a sticky plaster, and you would do better to solve the actual arcing problems, IMHO.
Do bare in mind that over half of Europe pays no attention to their wiring codes at all, so it doesn't really matter what they say. Our wiring regs in the UK have been very good for years (and have been adopted by some other countries too). However, the newest (17th) edition you refer to above has gone rather commercial now with companies managing to force their products into it without the considered risk and benefit analysis which made it a highly respected set of regs in the past. The 17th Edition regs claim to replace the 16th Edition in July 2008, but their legal force is via building regulations, which still explicitly require use of the 16th Edition for homes. So currently this is all in a mess, because the processes have been taken over by politicians and commercial concerns with their own agendas, whereas it used to be handled by real engineers who understood wiring and what safety and risk analysis is.

As an outsider looking in, and having managed computers in US computer rooms, I have several theories...
Older wiring in the US maybe an issue, but isn't one I've personally experienced. I don't think we ever used knob and tube in the UK. Wiring of that age in the UK included rubber insulation, and that will have long since disintegrated. Some of the last rubber cables installed in the 1950's are still found occasionally, but are very fragile -- the rubber exposed to air (and even more so, ozone near switch contacts) is brittle and turns to dust if flexed. This means nearly all of it has had to be replaced.
Another factor is that in the 1930's and early 1940's, we decided to design a new wiring scheme for power outlets, which became the 13A socket. This was phased in from 1946-1956 on new installations. I think that makes the 13A socket the newest mains socket design in widespread use in the world; certainly all the others I know of are older and date back to eras where safety was not such an issue and not so well understood. This means we have a plug which has learned the lessons from its predecessors, e.g. it doesn't wobble in the socket, generate poor contacts, and overheat. By around 1970, just about all premises which predated the 13A socket had been rewired to use it. That means we just don't have any premises with 1920's wiring -- it would be useless as you can't obtain plugs to fit in many of the various socket styles we used back then.
A number of the connection methods you still seem to use were phased out here over 60 years ago in favour of more reliable methods (or in the case of Wirenuts/Screwits, even longer ago). Generally, our connections are all some form of high pressure contacts to form gas-tight connection surfaces which thus can't corrode at the surface contact. You can't get that sort of connection from something like a wirenut/screwit or a backstub. Also, wrap-around terminals pretty much vanished ~40 years ago.
The irony is that with half the mains voltage and hence double the current for the same mains load, you generate 4 times the heating effect we would at a poor contact, so it's actually much more important for you to have reliable low contact resistances than it is for us. My guess is that this mismatch of relative requirements with reality is one of the most significant differences to be found between the wiring systems, and likely responsible for much of the difference in fire incident figures.
Safety of our wiring accessories and ease of assembly has improved many fold over the years. It's a selling point the manufacturer's use to differentiate and compete with each other. It's probably also a factor in DIY installation being quite a safe thing to do with our accessories. Whilst I don't suggest trying this, a well wired UK 13A socket manufactured in the last 20 years and installed well could be unscrewed from the wall with its circuit still live, pulled out on its wires, and you should be able to reach fingers around all sides and back without getting an electric shock. In comparison, when we see current US wiring accessories, they look like the pictures of wiring accessories we used to use back in the 1930's.
These are the points which appear most obviously different to me in ways likely to relate to safety.

The Germans blew it all up ?
I have always said Europe and Japan got a kick start after the war because they had to build new infrastructure. We still had all of our old "plant". (particularly the "big iron" tank factories) How else can you explain why cars in the US didn't advance much from the 30's until the rest of the world kicked our ass into changing.
I suppose you are right, we do have a lot of homes wired for a "60a" world and trying to live the American "200a" lifestyle. That does tend to get back to my "extension cords and cube tap" theory.

| | |>why we are preferring 3pin plugs in some devices (eg.in ups,extention |>boxes...) and why we are preffering 2pin plugs in some devices(eg.in |>ordinary night lamps)? | | Anything with a metal case probably should have a ground pin, although | it is not always true.
The way a device is used might also factor in. Aside from those using a wall wart power supply with a definite low voltage barrier, computers are devices that humans regularly operate (even though the keyboards tend to be make of plastic).
The way the wiring is done internally might also factor in. A device that has a greater risk of creating a path from the hot wire to the case, or is more subject to potential damage internally, or can get wet, might need a ground wire.
But I see many exceptions to a lot of these. A friend has an electronic organ that directly connects to AC power with a 2-conductor cord that is a lot like a shaver cord. Many kitchen appliances still use ungrounded plugs.
I think a lot of it is legacy, too.
A neighboor of mine, when I was in grade school, was wiring up an antenna to a new 2nd TV. It kept blowing fuses when it was turned on. He had the TV repair guy come check it out. After blowing yet another fuse he was trying thingswith it and notice it was OK with the antenna disconnected. So he went to put the antenna back on while it was on and got to see a nice bright arc and blew another fuse. Turns out the 1st TV was the culprit and the TV antenna was "hot" and sufficiently isolated from ground that it didn't draw enough to blow a fuse or burn out the twin line. The 2nd TV had the antenna line grounded. Reversing the plug on the 1st TV was the workaround. As far as I know, they left it like that. This was in the 1960's.
I had an old shortwave radio my grandmother gave me that would have its antenna lead "hot" when the plug was reversed. My grandfather gave me an old capacitor tester when I was getting interested in electronics. I got to feel what 120V was like one time when I touched it while barefoot in the basement.