You "missed the boat" with my 'safety device.' piglet's semiconductor switch only makes sense to me when its part of the electrical wiring and not plugged into either an outlet or a daisy chain of extension cords. (BTW, isn't a daisy chain of extension cords against the regs?)
The tripping characteristics of circuit breakers listed for use in US low voltage power distribution systems is well defined. A quick search on "Circuit Breaker Characteristic Trip Curves and Coordination" found:
formatting link
and much more, which I did not pursue.
Power distribution system reliability philosophy requires interrupting faults as close to the source as possible, in order to minimize impact to other loads. For example, the power distribution to a hospital OR might be fully coordinated against a single fault; designed so that when the distribution system is fully loaded no single fault (overload or short circuit) can trip any but the closest breaker. Other systems may be partially coordinated or coordinated against multiple faults.
Applying this basic philosophy to plug connected loads, they should never be able to trip the main panel breaker except in the case of a cord fault.
I haven't done any coordinated power dist work in this century, but IIRC all low voltage (600 volts or less) molded case circuit breakers rated for 600 Amps or less which are listed for use in the US are guaranteed not to trip at less than 80% of current rating continuous duty, guaranteed to trip within two cycles for short circuit up to rated Amps Interrupting Current (10kA for almost all 120/240) but not guaranteed to remain usable after clearing a short circuit (must fail open).
If doing a coordination study one uses trip curves for the exact breakers used, but for general consideration of breaker performance they are all very similar and the Schneider curves are representative.
On Thursday, December 28, 2017 at 10:04:44 PM UTC-5, snipped-for-privacy@gmail.com wrot e:
our 'safety
rotection.
hat's a well-known way of burning down buildings, even without deliberately shorting the mains to ground.
Phil, (or someone) can you explain about the daisy chained power strips. We got yelled at, at a trade show, but never quite understood the explanati on.
There's a short in something on the end of the chain... Then.?
Thanks, George H.
ilities of a hypothetical breaker at an unknown upstream point, not so much .
r might very well not clear the short before the thyristor melted. Or the w iring might be marginal, but good enough for many more years unless somebod y did something idiotic like shorting the mains.
ing the mains to ground, you'd get sued right down to your socks, and right ly so. If you did it in France or Italy and someone died as a result, you m ight very well go to jail.
There are a couple of issues. One is the multiple connections. The chances of a fire increase drastically, the more plugs inline. Extension cords tend to be used often and the outlets tend to wear (relax) causing high-resistance connections. Multiple in series isn't a good idea. Another issue is the concept of using the right tool for the job.
In the UK daisy chained extension leads suffer from earth conductor resistance that rises above permitted spec, reducing short circuit to earth fault current, which has 3 effects.
Earthed metalwork is then at a dangerous voltage until the fault is cleared
Fault clearance by the MCB is very slow
Fire may break out as a result - it's a lot of power to dissipate for too long.
However it has nothing whatever to do with what I was talking about, where the device in question a) limits fault current to a safe value, b) if the breaker supplying it does not clear it, then it clears itself by built in overheat or fuse protection
Best bet is to incorporate your own fusible link into the device. Coordinate with a 20 Amp breaker trip curve so that goes first. But if the upstream device fails (or is bypassed), your device's internal fuse blows.
Your device is not repairable, but at leat you don't get sued for burning down some moron's hose.
--
Paul Hovnanian mailto:Paul@Hovnanian.com
------------------------------------------------------------------
An ordinary fuse needs a short circuit current that is several times the fuse nominal current in order to blow in a few seconds. With only twice the nominal current, it takes many minutes before the fuse blows.
The short circuit current depends on the distribution transformer secondary output voltage and the _total_ loop resistance. The worst case loop resistance consist of the transformer to house, internal wiring in the house to the mains socket, the extension cord(s) resistance to the load connected to last extension cord. Multiply this resistance by two, to include both the Live and Neutral wire. With a short at the furthest point, this is the total loop resistance as seen from the distribution transformer.
Assuming that a 5x nominal current will blow the fuse fast enough, in
230 V land with typically 10 A fuses, the _total_ loop resistance must be less than 5 ohms.
In some countries it is assumed that there is a single maximum length extension cord in the longest in-house wire to a socket for a specific cross section (both in-house and extension cable). Plugging multiple extension cords in series, especially with smaller cross section, will increase the loop resistances too much.
As such, plugging multiple short (say 3 m) extension cords in series with sufficient cross section is not a big problem.
One way to indirectly measure th total loop resistance is to plug an incandescent lamp at the end of the last extension cord and then plug in and out some big load and observe how the lamp flickers during the transition. Incandescent lamps are very sensitive to small voltage variations. A strong variation will indicate a relatively high loop resistance.
For a typical residential fuse, it takes a few seconds to blow at 5x nominal current. In order to blow the fuse in less than a second (a few mains cycles) a 10x short circuit is required.
For a 230 V system, less than the following loop resistances are required.
10 A fuse 5x: 4.6 ohms
16 A fuse 5 x: 2.9 ohms
10 A fuse 10x: 2.3 ohms
16 A fuse 10x: 1.4 ohms
of 14 mOhms. Thus a 100 m long extension cord will have a loop resistance of 2 x 100 m x 14 mOhms/m or 2.8 ohms.
Such extension cord requires several seconds to blow the 10 or 16 A nominal fuse.
There seems to be some disagreement about terms used.
To my understanding "daisy chaining" refers to something like 10Base2 (thin wire) ethernet in which there is a RG58 cable between each station T-connector.
If the OP really asked about a tree structure with multiple branches, i.e a singe multiple socket extension cable and a short multisocket extension cables connected to each socket in the primary extension cord so with 5 socket extension cords you get 5 x 4 = 25 sockets. Some standard organisations seem to think that if the end user sees a socket into which you can mechanically plug a 10 or 16 A load that the customer will connect 25 such 10/16 A loads into the network causing a
250/400 A current drawn from the primary extension cord and hence causing a fire hazard. However, even if the end user would make such stupid thing, the fuse feeding the original extension cord should blow, protecting the house for sny fire hazards.
5 amp loads plugged into power strips 2,3,4 and 5 = 20 amps total, breaker does not trip. Power strip 1, rated at 13 amps, "sees" a 20 amp load, overheats.
Same for daisy chained extensions:
formatting link
20 amp ==ext1==ext2==ext3==ext4==etc The cumulative current flows through extension cord 1, rated for 13A (see link above) overheating it.
Here's what daisy chaining can look like:
formatting link
Here's a good read:
formatting link
And another good read:
formatting link
And here's some images that can stand your hair on end:
formatting link
It's especially chilling when you've actually seen that crap.
** Australia does not use fuses in residential installations - it's either a thermal/magnetic breaker OR a piece of "16 A fuse wire" in a ceramic hold er.
16A fuse wire holds at 35A for a long time, but opens at 45A in a second.
So 5 ohms max loop impedance is realistic.
Same number YOU posted - d*****ad.
ce
** The correct number is 17.5mohms.
** The correct answer is 3.5 ohms at room temp, 4.2 ohms at rated temp of 7
5C.
** Wrong again, it will be much quicker.
to get many outlets at an exhibition booth - not some great long run.
** Not when you include the context - f*****ad.
** Utterly irrelevant.
** NO - a simple "daisy chain" where one outlet on each strip is used to po wer another strip. Each strip having it own thermal breaker or HRC fuse if in the UK.
The mains supply breaker is NOT intended nor designed to protect the power using equipment. That breaker is there to protect the permanent wiring from the panel to the outlet to which the power using equipment is connected. Relying on the mains breaker to make the power using equipment safe in the event of a fault in that equipment is wrong, unsafe and stupid everywhere, and is illegal in the USA.
No. The overheat cutout, or any other equipment power safety device, must be designed to interrupt the current before the mains supply over-current device does. You must NOT rely on the mains panel breaker to protect the equipment, or personnel using the equipment. The panel breaker's job is to protect the building wiring up to the point where the external equipment draws power from it.
No. If the equipment is listed by a certified testing laboratory such as the UL (Underwriters Laboratory) (or other accepted testing lab) the lab is saying, in effect, it was manufactured in accordance with the rules and meets electrical (and other) safety requirements. If a fuse (or breaker or whatever) is required and the device doesn't have it, it won't be listed, nor will it be legal for the manufacturer to sell it in the US.
You could spend months reading all the applicable rules and regulations concerning product electrical safety, but I believe the precis above captures the essence. If you want to do some wading through it, it's probably best to start with the NEC and the UL White Book. (links below)
Ed
You can wade through the wording of our National Electrical Code article 90-7 (et al) if you like:
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.