Well they look really effective. Maybe it will trip a week after a shovel goes through your extension lead with a 3kw heater on the end of it in the rain. :-/ Junk
You have the Li/A version which has a delayed trigger mechanism so you need 30mA of high frequency current component while using 500mA for a longer time, up to a second. The higher the load current, the faster it trips. If you put ~5A in parallel it should trip faster and more reliable. These things are tricky to do (and to trigger) since there is a certain overlap in characteristics of ordinary switches and those faults. The spec sheets and curves are somewhat hard to interpret too for certain manufacturers.
I'm not fully up to speed with your points Dennis, although I think I get the gist - a load is required of a certain characteristic which corresponds to the characteristics of this AFDD. However, shouldn't we be testing these things under actual/'normal' operating scenarios? If we had a domestic lighting circuit with no load, but damage occurs resulting in serial/parallel arcing then shouldn't we expect this to operate? Or is your point that John is testing a type of AFDD not typically installed in a domestic lighting circuit? Cheers.
I sort of get your point on the arc detection though I do agree with Coubertin67s comment. However that doesn't excuse the over voltage fail. If we look at the evidence from the US these things are a nightmare; either failing to trip when they should or continually nuisance tripping. Also the price of them is ridicules. I'm about to put a new RCBO board in my own house at a materials cost of just shy of £200. If I were to use AFDDs I'd be putting another 0 on the end. The argument that the price will come down with increase in unit sales is bogus, again I point to the US where they are still very expensive indeed and have been code for a while now. (And I'm expected to pay for the juice they use!)
@@Raysnature I have never bothered too much about the characteristics of its over voltage protection abilities, but I know that certain manufacturers don't use AC RMS but rather the maximum voltage peaks as a trip criterion, so in case of using his transformer setup I would wonder what the waveform looks like and if it maybe doesn't go all the way up to 380V
Dennis Lubert - transformers don’t change the waveform shape that much. The output will still be an approximate sine wave with the peek voltage being 1.41 times the RMS voltage.
I would like to see a test done on a hair dryer. These are the devices that the cords end up fraying on the inside of the insulation and spontaneously flashing into a hot white plasma fire on me. If an Arc Fault Circuit Interrupter can't protect you against that, then these devices are just a load of snake oil.
Yes we have managed to live without them for quite a while. Snake oil really. Create a scare and provide a magical device to stop it all, scare the public and we have this magical £200 thing that will save you from burning alive. Yeah right. AFDDs seem to trip with harmless brush sparks but will ignore a flashing light fixture with arcing between the millimetre thick aluminium strip that consists of "Crabtree" pendant light contacts. Those Craptree luminaires lasted 10 years, all replaced because all arced thanks to the crap 1mm thick aluminium terminals. New ones have heavy duty copper terminals with a strong spring instead of a coke can's tab doubled back on itself for a bulb contactor. Arced with the room lit up all flashing like a rave until the room stank of ozone and the lampholder was brown and smoking. What do you think the AFDD did? Ignored it. Tried it with all 3 failed Craptree luminaires and also a demo arc I set up. Brush arc in an antique Singer (perfectly normal, these old Singers really spark) and it tripped instantly. The arcing luminaires are a fire hazard. The Singer isn't. Oil of adder, eye of python.
I would like to hear what the manufacturer of these AFDD's has to say. Could the high voltage trip need to be a spike within milliseconds? In order for it to trip, and the arcing need to be a heavier load? Is there a spec sheet with these AFDD's? Seem pointless for domestic installations. Looking forward to your next report captain JW great video.
A constant overvoltage condition is what *should* trip it. A spike should be irrelevant - that's why MOVs exist. Forget the spec sheet, there's no point if all it does is detect some arbitrarily defined arc test that never occurs in the real world. And we all know angry pixies can't read spec sheets anyway. If it can't detect a realistic fault as tested here, then what's the point?
1,000% they do. Ooh evil arcing is happening, superhero AFDD will come and save you! Except it ignores a lot of arcs and will trip out with brush sparks on electric motors (sewing machines, drills, etc) that won't do any harm.
That or a space heater. The only time I've ever had a device that produced noticeable sparking and caused me concern, it was a faulty 1800 watt space heater.
So, in short ( no pun intended ) these devices are useless and offer even less protection than a standard RCBO ( the tripping times are way higher than most). Well done Eaton.
I wonder if the device is rejecting the sparks and arcs you're causing because they're infrequent and are similar to what might be expected when switching devices on and off, particularly with a worn switch? Could a test be devised where arcing is more or less constant? In real life, I'm reminded of the fizzing socket that's got a loose wire behind it. Thanks for interesting and informative videos!
I wonder if the AFDD is calibrated to detect a much smaller amount of arcing over a longer period, such as might accompany a high-resistance fault that generates a lot of heat. The kind of arcing shown in the test seemed unstable; if left untouched, I'd expect it to either go open circuit (problem solved!) or revert to a high-resistance fault with less arcing. It would be interesting to just leave it as a loose connection, let it arc or not arc by itself and see how hot it gets before the AFDD trips. (If indeed it does trip.)
I'm thinking they are designed to not support nuisance tripping as people would just get annoyed and not use them. I would also expect the drill to produce more arcing than the bad connection for the light and again more when the drill is under load related to the amount of current going through it. I'm thinking they will only trip under high current arc's with some thinking being that if you can disperse the heat faster than the arc can generate it then there wont be any fire and these are are sold as 'fire protection?' Although you can argue that a socket full of dust will go up on a smaller arc than one that isn't full of dust. So it might be a careful balance of how annoying is annoying. Thanks for the videos, I have been keeping an eye on these for a little while now.
I thought there were standards for what the transients had to meet in order to trip. You wouldn't want them tripping on every single like you would have in a light switch or turning on the vacuum sweeper. You could put a scope across your two contacts when you make the arcs and see how they compare with the standards. Odd about the over voltage not tripping it though. Make me wonder how high you would have had to push it before it tripped. Interesting video. I thought these were smart enough to look at changes in the waveforms over time, almost like making a rolling signature analyzer. It would be nice to see you dive into the details on how they work and if your series arc should have caused the unit to trip.
Our's was increased from 11ppu to ----->> 38ppu !!! When the company owning the properties, changed ESP from British Gas to N-Power. The contract was changed, without consultation with the owner-occupiers or tenants, from a 'domestic' to 'business' tariff, which allowed N-Power to levy a higher charge, but giving the owners a tax loop-hole, that charge was then subsequently passed on down by the owners to us, the consumer. . The 'owner-occupiers' (this being a 'Park Home' and caravan estate) kicked-up, because they were paying by post-use billing and noticed the £200 quarterly charges... They refused to pay. The owners caved in and agreed to reduce the charge to 18ppu. Those of us on 'pre-payment' energy card meters - mostly the caravan tenants - have not been given any reduction. We still have to pay (what I feel sure must be illegal) inflated charges. But then I'm one of only a handful who knows that these charges were increased, most tenants on sight are oblivious, because the owners never communicate to the tenants. . If only we could find some legal team to look into this. My winter electricity bills are around £80 per month and that is with being tight on what is switched on, not forgetting ever more milder temperatures. In high winds (such as now), the caravan is draughty, the walls are about 35mm thick, no double-glazing (which would be pointless anyway). No gas heating or hot water... So my sole heating devices are two portable electric fires, one static oil, the other a small fan heater and an electric blanket. For cooking, it's the kettle, microwave, mini-oven/grill and crock-pot.
@@Deebz270 I am real careful with my gas. And this year I did not use it very much after February. I topped meter up a few weeks ago and wo. £89 of debt for not using any gas. And when I top up it takes £7 out of £10 for debt or £14 out of £20 for debt. This is insanely disgusting...The daily standing charge is annoying but to claw it back at 70% of a top up is really bad. Im with EDF.
The mind set of manufacturers is an AFDD in each circuit, so fitting in with (enlarged) conventional consumer units. Given that arcs emit an RF signature, there must be a better way. A single centralised receiver/analyser scanning the circuits would seem less costly and space hungry. Various developments on this theme are possible.
Maybe try a capacitor as the load, or a switching power supply (with some load). Those often end up with loose/broken connections. Also would be interesting to see if that will kill the input rectifier diodes in it. I don´t have high hopes for it to trip though.
Great video going with the 18th edition I think they're going to start making AFDS compulsory lot of us don't really know too much about them at the moment so thank you for the video
Hi John, you know I expected to see the results you got, these things must be set to a compromise otherwise they would probably trip everytime a worn switch is actuated! It's a good idea and it does provide an extra layer of protection well at least in theory! I think another way is to have a device that can detect extreme electrical noise generated within the installation, maybe with an adjustment facility for setting up against oversensitive tripping! A kind of realtime monitoring of the health of an installation! Kind regards Warren
I would suggest that the reason that the AFDD is not tripping is that you are not subjecting it to an arc. You are only generating a series of short sparks which the device is specifically designed to ignore. How about trying again using something inductive such as your Variac as the load. The drill doesn't trip it because it has an internal RFI suppression filter that prevents the RF from being conducted back into the mains as per the EMC regulations.
The Eaton AFDD is obviously not working properly and therefore should be withdrawn by Eaton. At the very least it should trip on overvoltage! Maybe you should report it to them and the local trading standards body to trigger a product recall.
I have the same exact model and I already had to swap out one faulty one, was tripping constantly and without any load. Having 2 adiacent ones, i can confirm also the "empathic" tripping, now the second one is also tripping falsly sometimes. Quite a dissapointment, I might switch back to the RCD's.
Question: Is it worth testing a heavy duty brushed power tool (drill/angle grinder) under heavy load. Assuming the tool is otherwise safe and maybe the noise suppression capacitor has seen better days to gauge how much 'nuisance' tripping would occur? However, based on your findings so far nuisance tripping would be the least of our worries! 😊
Well performed tests. A bit of arcing has under normal circumstances hurt no-one, except when combustible substances were near. An other thing where these devices can't help is bad contacts. I had once a situation where someone had not tighten the screws in a wall socket. There was just no fire thanks to my quick reaction when I smelled the burning PVC. I had to replace the wall socket and 3 m of wiring. And repaint the (wooden) wall.
Dear Mr John, Thank you for making the videos. I have chanced other AFDD videos. In which, the person had used higher ampere load like you did in first video. His AFDD tripped after he disturbed for a while. Another video has had cable fire before AFDD tripped. Could you share your opinion on those?
Tim Ballam - what rubbish are you talking? First, ALL wires ARE inductors (although in the wire that John used, the inductive value is so small, it can be ignored) . Second, inductance is likely to make an arc worse with a resulting bigger voltage spike. The greater the value of the inductor, the bigger the voltage spike.
@@Mark1024MAK ha ha ha the cable used in this "experiment" is creating a rather effective LC filter that's most probably filtering the 100KHz that the device is looking for!
Hi JOHN, I have a 3 phase 10kw solar inverter connected to grid. It has 4 MOVS: S20K300 which keep blowing nearly every month or so. 3 MOVS are between 3 phases and earth and the 4th is between neutral and earth. Which external protection device should I use to prevent these from getting damaged? Currently there is a 6 Amp mcb connected to 4 pole contactor, which is then conncted to inverter. Panels installed are 3KW hence 6 Amp mcb is fine. I am also curious if our grid has some issues or if my earthing has some problems. Apparently earthing looks fine as with multimeter it is measuring 2 volts more than neutral when checked against phase. Appreciate your advice. Thanks
Nice tests, but that lamp arcing has such low energy, would you think that would burn something down? Someone below is asking about a hairdryer, that would be in the burnable category i think, curious if that one would trip it. The overvoltage function seems something someone from marketing dragged in to show the thing off :P
Interesting, I was chatting with a colleague about AFDD's and we were speculating some what as to how they work. Having looked into the info from manufacturers its my understanding that an prevalent fault with arc has quite a number of unique current draw signatures and I gather research is ongoing. the device looks for a rise and decay pattern linked mainly with loose connections. Looking forward to seeing more about these devices as the use of them has not been as widespread as it looks like it will be in the future. regards Sean
It would be interesting to see the results of a test like this across a variety of different brands of AFDD and perhaps also put the AFDD into an arrangement with other protective devices (RCD, MCB) to see if inbalance or increased current is detected quicker than the AFDD can detect the arc fault. My suspicion is the AFDD is a pointless addition to a CU, taking up space unnecessarily.
It looks like Eaton has addressed the nuisance tripping issue very well. Unfortunately, in real life it is difficult, if not impossible, to tell the difference between a loose wire on a 100 watt load and a drill or any tool with sparking. The problem is more difficult in the UK than in the US because the current level in the UK is only half as large. Tripping on low current loads will probably not be satisfactory in the market place, it's just too tricky. Loose wires on switches or connectors are potentially very dangerous but nuisance tripping makes people skeptical about the value of AFDD's. Manufacturers have been working on the AFDD for many years and they still have room for improvement.
That is not a fail - many light switches can arc if you press slow on them …. And i dont want my power/lights in my house to go down in that case + the risc of fire hazard is negligible :)
Thank you for another interesting video on this AFDD. It’s a fair assessment to say this device doesn’t work. Any chance of you comparing it with another one?
Only problem there is that the guidance (looked at in Part 3) states that AFDDs do not trip on overheating conductors such as the classic shower isolator failure, they only apparently work when actual arcing is present.
@@jwflame In your experience would you not have said you'd normally still get some amount of arcing in that failure? I've only had the pleasure of looking at one that's failed that way but it did at least look as if some arcing may have occurred.
No - it's AC, same current flows in each conductor, and the direction of that current changes 100 times a second in both of them. Device has no earth connection so would be impossible for it to determine which was L&N.
If you look at the Eaton promo video it shows it tripping in about 2 seconds, with a caption stating "The AFDD+ will immediately trip" - ua-cam.com/video/UtOmau4ym1Q/v-deo.html
Brilliant video John! That’s really opened my eyes. Maybe we will have some future amendments on the horizon? Maybe variable sensitivity? What if several devices are daisy chained on a large installation? ( discrimination type faults) Maybe different modules for different types of circuits? I doubt very much that the device itself is faulty. I think this is a cross the board example of them all in their entirety. Like you say, lighting loads are very low nowadays. A few mA? I wonder if a back feed from a lightning strike would trip it? Lol! I wonder what an arc welder would do? It’s looking like more research is needed regarding suitability. I wonder if these new regs should have been postponed for a while. Your brilliant work has proved so much! Thanks for the well presented and professional approach to this new subject.
The thing with AFCI/AFFDs is that they are a fantastic safety device IN PRINCIPLE. Unfortunately, they were rushed into mandatory service several decades before the technology was anywhere near mature enough to actually deliver on the promises. AFCI breakers are about as mature a technology today as the early GFCIs were, perhaps, 40-50 years ago. Sadly, they are mostly only good for excessive nuisance trips (the most popular brand in the US, Eaton, is notorious for tripping in the presence of even fairly weak RF signals, so one amateur radio operator in a community transmitting at just 100W can potentially knock out power to entire subdivisions equipped with these faulty AFCIs, and they are out there by the millions now). What the NFPA was thinking when they convinced the NEC to rush them into mandatory service decades before they were a mature technology is anyone’s guess. The reality is that many homeowners, after paying thousands of dollars to outfit their load centers with AFCI breakers, are at some point down the road very likely going to have to pay many thousands more to replace them all two or three more times before they are finally a mature and reliable technology...or they will have to put up with all the nuisance trips that they can produce. After all, having your refrigerator circuit repeatedly deactivated because someone drove by with a CB radio (or any other radio of at least a few watts power) is liable to get old, and costly, after awhile. That and one has to wonder if the circuitry in these was genuinely engineered well enough to provide decades of service without the circuitry itself failing at some point as homeowners do not ordinarily replace circuit breakers for many decades at a time - visit a 40 or 50 year old home and you are likely to find 40 or 50 year old circuit breakers, I just don’t see these lasting anywhere near that long on average.
To touch the copper conductor on line @8:39 whilst plugged in and the device live shows great confidence in having actually remembered to turn off the switch on the outlet... I have grave doubts about the utility of devices like this, as it seems to me they are either going to be so sensitive that there will be lots of nuisance trips or so insensitive to avoid that, it will not provide any functional benefit anyway. However, I can still imagine the industry lobbying for these just to make more money. If they were to be compulsory, there's also the little issue of space as if a three module width unit is required for every circuit, I doubt many installations would have enough space, and in many properties there won't the enough room in the current location to install a much larger one.
Have seen that experiment (it´s german) ua-cam.com/video/CWqVVDEvd0Q/v-deo.html He says, that the temperature carbonizes the isolator and this leads to a stable arc which then triggers the AFDD (in the second trial instantly). He states that that is the error condition the AFDD should actually trigger, nothing else.
I understand people asking for a higher load to be tested which I would like to see, but they are not as far as I’m aware marketed as high current arc fault devices. I personally detect chinesium in affect.
Very interesting, and comprehensively tested. Clever looking for DC current when arcing. I am sure some local council authority (UK), will be convinced, or cajoled by nefarious means to have these as standard fitment in new builds etc. I can see the rationale behind the arc detection, I assume this feature is used in line power applications. Thanks for sharing.
I am shocked that the device didn’t trip after all that effort . It’s going to be difficult to look a customer in the eye and say they need to spend a £1000 on AFDDs on each circuit when the test results so far are utterly unconvincing. Is it the brand? What about other manufacturers’ devices? Is it just that particular device? I can feel a destruction test coming on John .......? Adrian
Perhaps we're all confused - it's actually meant as a device to protect against faults that could develop in an arc welder on the circuit. John must have missed that in the small print or some such. I mean it's not as though he's known for being fastidious now is it! /sarc off
Thanks for the video. Either that example you have there is possibly a dud, or the concept is. One of the methods the devices use to detect arcs, is to sense the presence of power harmonics. So in the case of an arc on 50Hz, they are supposed to sense a 100Hz signal on the cable, with the ‘signature’ of a typical arc. Presumably then, the on-board software takes note of the duration and repetition of the 100Hz and decides whether, or not, it is an arc. I think this demonstration shows, assuming that the example under test is OK, that an AM radio is, at present, a more reliable electronic device to detect the signature of arcing.
Well, as of this week (21/9/2020) the draft Amendment 2 to the UK regs, BS7671 is proposing to make them compulsory for all circuits of 32A and below that are not lighting. standardsdevelopment.bsigroup.com/Projects/9020-04654/#/section is the draft for public comment, section 421.1.7 is the key one. (about 2/3 through the first of the 4 parts) You do need to create an account with the BSI to leave a comment, but anyone can do that. I for one will indeed be commenting, that I think this is not a good thing to add to the regs. Others may care to do the same.
Having watched your videos and David Savery's videos about AFDDs I will not be rushing out to spend loads of money on them. From what I have seen, they are not fit for purpose.
This seems to be whats required to trip a AFDD. They seem to be pretty insensitive: ua-cam.com/video/JRXFZHKGaXg/v-deo.html . Seems they are deliberate that insensitive to not false trip when running brushed engines and such. It would be smarter if AFDD manufacturers could include a "learn" mode where you push a learn button and then you run said machine/thing and test lightswitches and so on that is connected to said Circuit, and then it will learn a threshold so it don't false trip but still trip at unnormal arcing. Same with GFCIs, Personally I think 30mA is pretty high and it would be easy for GFCI manufacturers to include some "learn" mode here too, where it will Always trip at 30mA, but the "learn" mode can be used to "learn" a lower trip level into the system, so on a good circuit with low Earth leakage, it could even be learnt to trip at like 10mA or even 5mA. And in such a combined device, the "learn" button could then be responsible for both Learning AFDD And GFCI parameters into it. Or even a fully automatic learn mode where it will during the course of a week of normal usage, automatically learn the "normal parameters" of the Circuit (arc signature and Earth leakage current) and if anything gets unnormal, it should of course trip.
Not so sure about that "learn mode", install one on a circuit with an existing fault, it learns to ignore it from day one. Just sounds like a way to treble the price.
The arc fault breakers in the US are generally rated for around 2W consumption. This is kinda absurd since it’s not uncommon to have 15-30 of these in a larger new house.
I am informed these devices only detect an arc if the load is > 2.5 amps - so a broken heating element in an electric blanket arcing and setting fire to your bed is OK !!!!!!!!!!!!!!!!
Not really much of a surprise that this AFDD failed to trip on a 40 watt load, arcing or not... Given that it failed to trip effectively even with > 2 Kw load. Yet more money grabbing nonesense.
Don't You think they should re-introduce incandescent lamps for Arc Detection! But seriously though these are deemed as safety devices, and too many people already think a RCD will protect them if they grab the live and neutral.
What AFDDs are supposed to protect against are arc faults that could start a fire. At the same time they shouldn't trip on conditions unlikely to start a fire.
Robert Grant - And how do you tell the difference between an arc that can or cannot start a fire? If we can’t say, how do expect a “dumb” microcontroller chip to tell the difference...
We can say. We know the ignition temperature of PVC, we can calculate how much energy is required to heat the conductors to that temperature and we can measure or calculate the high frequency spectrum created by such an arc and design a detection / protection device accordingly.Such a protection device is not going to be 100% reliable in preventing electrical fires but neither are RCD / GFIs at preventing every possible accident from an electrical mishap . Here is Au the RCDs used for most domestic installs are rated to trip at 30mA. For the majority of humans that will protect them from electrocution but will not prevent every possible death from an electrical mishap. It also means there's some issues with parasitic tripping of RCDs from condensation etc. The general consensus is this is a reasonable compromise.
I wouldn't have said "bullshit". Certainly they will not protect against every possible way electrical energy can start a fire just as RCDs will not protect against every possible way to be get an electrical shock. The same goes for seatbelts in cars and helmets on the heads of cyclists.
The brick wall he had in the background of himself in the first clip looked quite realistic despite the fact I assume it was CGI, as in the last clip we get the real background of his study/home office
I would repeat the test where the voltage increase is sharp and quick. I.e more analogous to an actual "spike" than a slow, creeping voltage increase, which is how you tested it. I make this comment because, of something interesting I observed that happened recently in my kitchen when I was carefully following some recipe, (Please stick with me on this. It really is relevant). Well, I was precisely measuring out some salt, and It got to the point where I was almost up to weight and I was adding extra salt practically grain by grain. After a few minutes of the digital scales holding steadfast, I was becoming quite annoyed. "Surely, I have must added more than a gram by now" I said to myself while the scales were still stuck on 3.9g. (I was aiming for 4). I ran off to get some other digital scales (equivalent accuracy of 0.1 g) and dumped the contents of the first pair of scales onto the second. Well, well. What a surprise: 5.7 grams! Try it for yourself. Measure out say 10g salt or sugar onto your scales and then gradually add to it grain by grain. You'll be there for quite some time before those scales move. Discalaimer: It also could be just that my scales are cheap. But it would be interesting to know if the same thing happens on a "quality" pair of scales. (PS: It's a long time since I've seen a Hospital Saddle.)
its not inly on cheap scales at wor somethimes i have to count leds with a rely expenciv scale with a resolution of 0.01g and a countig funktion some leds wont get counted but if you lift the tray and place it on the scale again it has counted the led
It's probably caused by stiction, ie the fact that a bearing or sliding surface requires a higher amount of force to start it moving than the friction it sees once it is moving.
Except that most electronic digital scales use a strain gauge - no bearings to get stuck. A typical strain gauge is a resistive sticker put on a bar of metal. As the metal moves slightly under load, the resistance of the strain gauge changes. This is measured by a microcontroller.
Well, if I ever want my house to burn down from bad electricals... this is the device to use!
And the pretty green light stays on while that's happening!
Well they look really effective. Maybe it will trip a week after a shovel goes through your extension lead with a 3kw heater on the end of it in the rain. :-/ Junk
You have the Li/A version which has a delayed trigger mechanism so you need 30mA of high frequency current component while using 500mA for a longer time, up to a second. The higher the load current, the faster it trips. If you put ~5A in parallel it should trip faster and more reliable. These things are tricky to do (and to trigger) since there is a certain overlap in characteristics of ordinary switches and those faults. The spec sheets and curves are somewhat hard to interpret too for certain manufacturers.
I'm not fully up to speed with your points Dennis, although I think I get the gist - a load is required of a certain characteristic which corresponds to the characteristics of this AFDD.
However, shouldn't we be testing these things under actual/'normal' operating scenarios? If we had a domestic lighting circuit with no load, but damage occurs resulting in serial/parallel arcing then shouldn't we expect this to operate?
Or is your point that John is testing a type of AFDD not typically installed in a domestic lighting circuit?
Cheers.
I sort of get your point on the arc detection though I do agree with Coubertin67s comment. However that doesn't excuse the over voltage fail.
If we look at the evidence from the US these things are a nightmare; either failing to trip when they should or continually nuisance tripping. Also the price of them is ridicules. I'm about to put a new RCBO board in my own house at a materials cost of just shy of £200. If I were to use AFDDs I'd be putting another 0 on the end. The argument that the price will come down with increase in unit sales is bogus, again I point to the US where they are still very expensive indeed and have been code for a while now. (And I'm expected to pay for the juice they use!)
@@Raysnature I have never bothered too much about the characteristics of its over voltage protection abilities, but I know that certain manufacturers don't use AC RMS but rather the maximum voltage peaks as a trip criterion, so in case of using his transformer setup I would wonder what the waveform looks like and if it maybe doesn't go all the way up to 380V
Dennis Lubert - transformers don’t change the waveform shape that much. The output will still be an approximate sine wave with the peek voltage being 1.41 times the RMS voltage.
When John says the overvoltage criterion for operation is 380V AC is he getting this from the spec sheet? And does it refer to RMS or peak V?
Hey JW, since Eaton's is a major fail, how about trying some others like Hager etc? See how they perform
Excellent analysis of these AFDDs. Important and very useful info!
Hard work getting that thing to trip. Don't think I'll be using anytime soon.
Thanks for the video John.
I would like to see a test done on a hair dryer. These are the devices that the cords end up fraying on the inside of the insulation and spontaneously flashing into a hot white plasma fire on me. If an Arc Fault Circuit Interrupter can't protect you against that, then these devices are just a load of snake oil.
As an electrician I can say that these are a massive scam. Totally unnecessary.
Yes we have managed to live without them for quite a while. Snake oil really. Create a scare and provide a magical device to stop it all, scare the public and we have this magical £200 thing that will save you from burning alive. Yeah right. AFDDs seem to trip with harmless brush sparks but will ignore a flashing light fixture with arcing between the millimetre thick aluminium strip that consists of "Crabtree" pendant light contacts. Those Craptree luminaires lasted 10 years, all replaced because all arced thanks to the crap 1mm thick aluminium terminals. New ones have heavy duty copper terminals with a strong spring instead of a coke can's tab doubled back on itself for a bulb contactor. Arced with the room lit up all flashing like a rave until the room stank of ozone and the lampholder was brown and smoking. What do you think the AFDD did? Ignored it. Tried it with all 3 failed Craptree luminaires and also a demo arc I set up. Brush arc in an antique Singer (perfectly normal, these old Singers really spark) and it tripped instantly. The arcing luminaires are a fire hazard. The Singer isn't. Oil of adder, eye of python.
I agree - make your comments to the consultation document before 11 December and we can try and stop this scam
I would like to hear what the manufacturer of these AFDD's has to say. Could the high voltage trip need to be a spike within milliseconds? In order for it to trip, and the arcing need to be a heavier load? Is there a spec sheet with these AFDD's? Seem pointless for domestic installations. Looking forward to your next report captain JW great video.
A constant overvoltage condition is what *should* trip it. A spike should be irrelevant - that's why MOVs exist.
Forget the spec sheet, there's no point if all it does is detect some arbitrarily defined arc test that never occurs in the real world. And we all know angry pixies can't read spec sheets anyway. If it can't detect a realistic fault as tested here, then what's the point?
I would say the arcing needs to be a heavier load.
Call me cynical but I wonder if the manufacturers of such BS devices have an influence on standards and amendments...
Good point and I'm guessing they have suckered the authorities as usual..
They probably give money to and lobby the standards people.
Yeah, I was wondering how much they got lobbyists involved in this...
They're already "recommended" which is a strong hint that they'll be mandatory at some point in the not too distant future.
1,000% they do. Ooh evil arcing is happening, superhero AFDD will come and save you! Except it ignores a lot of arcs and will trip out with brush sparks on electric motors (sewing machines, drills, etc) that won't do any harm.
A National service chap. 👍👍👍
International!
Eaton are currently developing chocolate tea pots.
I was thinking you should try it with a larger load, since a 40 watt lamp is only 1/6 of an amp. Maybe try it with an electric kettle.
That or a space heater. The only time I've ever had a device that produced noticeable sparking and caused me concern, it was a faulty 1800 watt space heater.
why cant an arcing 40w load start a fire? i think microsoft have some early xbox psu connectors to sell
Cadwaladr In Canada, these are are required now in bedrooms, where lighting loads would be much more likely.
So, in short ( no pun intended ) these devices are useless and offer even less protection than a standard RCBO ( the tripping times are way higher than most). Well done Eaton.
I wonder if the device is rejecting the sparks and arcs you're causing because they're infrequent and are similar to what might be expected when switching devices on and off, particularly with a worn switch?
Could a test be devised where arcing is more or less constant? In real life, I'm reminded of the fizzing socket that's got a loose wire behind it.
Thanks for interesting and informative videos!
Narrowing operation down to very specific situation. Don't you think it should operate in any arcing scenario?
It looks like you could run the cables from an arc welder through this device and it still wouldn’t trip !
The overvoltage protection - would it make a difference if you had any current flowing?
I doubt it. A load would drop the voltage a little.
That shouldn't cause any difference i.n reaction
I would imagine it would just be the voltage the processor see’s
I wonder if the AFDD is calibrated to detect a much smaller amount of arcing over a longer period, such as might accompany a high-resistance fault that generates a lot of heat. The kind of arcing shown in the test seemed unstable; if left untouched, I'd expect it to either go open circuit (problem solved!) or revert to a high-resistance fault with less arcing. It would be interesting to just leave it as a loose connection, let it arc or not arc by itself and see how hot it gets before the AFDD trips. (If indeed it does trip.)
I'm thinking they are designed to not support nuisance tripping as people would just get annoyed and not use them. I would also expect the drill to produce more arcing than the bad connection for the light and again more when the drill is under load related to the amount of current going through it. I'm thinking they will only trip under high current arc's with some thinking being that if you can disperse the heat faster than the arc can generate it then there wont be any fire and these are are sold as 'fire protection?' Although you can argue that a socket full of dust will go up on a smaller arc than one that isn't full of dust. So it might be a careful balance of how annoying is annoying. Thanks for the videos, I have been keeping an eye on these for a little while now.
I thought there were standards for what the transients had to meet in order to trip. You wouldn't want them tripping on every single like you would have in a light switch or turning on the vacuum sweeper. You could put a scope across your two contacts when you make the arcs and see how they compare with the standards. Odd about the over voltage not tripping it though. Make me wonder how high you would have had to push it before it tripped. Interesting video. I thought these were smart enough to look at changes in the waveforms over time, almost like making a rolling signature analyzer. It would be nice to see you dive into the details on how they work and if your series arc should have caused the unit to trip.
5:10 Note how you can quickly estimate the cost from the watts. 0.84 W => £ 0.89. 1 W is roughly £ 1.
My electric is over 16p a unit
My electric is over 21p a unit
@@simonlewis7327 im not going to mention mine . Electricity is free its in the air . They scum us big time.
Our's was increased from 11ppu to ----->> 38ppu !!! When the company owning the properties, changed ESP from British Gas to N-Power. The contract was changed, without consultation with the owner-occupiers or tenants, from a 'domestic' to 'business' tariff, which allowed N-Power to levy a higher charge, but giving the owners a tax loop-hole, that charge was then subsequently passed on down by the owners to us, the consumer.
.
The 'owner-occupiers' (this being a 'Park Home' and caravan estate) kicked-up, because they were paying by post-use billing and noticed the £200 quarterly charges... They refused to pay. The owners caved in and agreed to reduce the charge to 18ppu. Those of us on 'pre-payment' energy card meters - mostly the caravan tenants - have not been given any reduction. We still have to pay (what I feel sure must be illegal) inflated charges. But then I'm one of only a handful who knows that these charges were increased, most tenants on sight are oblivious, because the owners never communicate to the tenants.
.
If only we could find some legal team to look into this. My winter electricity bills are around £80 per month and that is with being tight on what is switched on, not forgetting ever more milder temperatures. In high winds (such as now), the caravan is draughty, the walls are about 35mm thick, no double-glazing (which would be pointless anyway). No gas heating or hot water... So my sole heating devices are two portable electric fires, one static oil, the other a small fan heater and an electric blanket. For cooking, it's the kettle, microwave, mini-oven/grill and crock-pot.
@@Deebz270 I am real careful with my gas. And this year I did not use it very much after February. I topped meter up a few weeks ago and wo. £89 of debt for not using any gas. And when I top up it takes £7 out of £10 for debt or £14 out of £20 for debt. This is insanely disgusting...The daily standing charge is annoying but to claw it back at 70% of a top up is really bad. Im with EDF.
"it doesnt trip this device but neither does anything else it seems"
that made me laugh John
The mind set of manufacturers is an AFDD in each circuit, so fitting in with (enlarged) conventional consumer units. Given that arcs emit an RF signature, there must be a better way. A single centralised receiver/analyser scanning the circuits would seem less costly and space hungry. Various developments on this theme are possible.
Maybe try a capacitor as the load, or a switching power supply (with some load). Those often end up with loose/broken connections. Also would be interesting to see if that will kill the input rectifier diodes in it.
I don´t have high hopes for it to trip though.
Great video going with the 18th edition I think they're going to start making AFDS compulsory lot of us don't really know too much about them at the moment so thank you for the video
Hello JW. I like that Drill... Are you still using it ? If not is there a Possabillity of Selling it to me ?
Hi John, you know I expected to see the results you got, these things must be set to a compromise otherwise they would probably trip everytime a worn switch is actuated!
It's a good idea and it does provide an extra layer of protection well at least in theory!
I think another way is to have a device that can detect extreme electrical noise generated within the installation, maybe with an adjustment facility for setting up against oversensitive tripping!
A kind of realtime monitoring of the health of an installation!
Kind regards Warren
I would suggest that the reason that the AFDD is not tripping is that you are not subjecting it to an arc. You are only generating a series of short sparks which the device is specifically designed to ignore. How about trying again using something inductive such as your Variac as the load. The drill doesn't trip it because it has an internal RFI suppression filter that prevents the RF from being conducted back into the mains as per the EMC regulations.
The Eaton AFDD is obviously not working properly and therefore should be withdrawn by Eaton. At the very least it should trip on overvoltage! Maybe you should report it to them and the local trading standards body to trigger a product recall.
I have the same exact model and I already had to swap out one faulty one, was tripping constantly and without any load. Having 2 adiacent ones, i can confirm also the "empathic" tripping, now the second one is also tripping falsly sometimes. Quite a dissapointment, I might switch back to the RCD's.
Question: Is it worth testing a heavy duty brushed power tool (drill/angle grinder) under heavy load. Assuming the tool is otherwise safe and maybe the noise suppression capacitor has seen better days to gauge how much 'nuisance' tripping would occur?
However, based on your findings so far nuisance tripping would be the least of our worries! 😊
How about an ARC welder ?????????????
Well performed tests.
A bit of arcing has under normal circumstances hurt no-one, except when combustible substances were near.
An other thing where these devices can't help is bad contacts.
I had once a situation where someone had not tighten the screws in a wall socket.
There was just no fire thanks to my quick reaction when I smelled the burning PVC.
I had to replace the wall socket and 3 m of wiring. And repaint the (wooden) wall.
Dear Mr John, Thank you for making the videos. I have chanced other AFDD videos. In which, the person had used higher ampere load like you did in first video. His AFDD tripped after he disturbed for a while. Another video has had cable fire before AFDD tripped. Could you share your opinion on those?
In both cases you created a nice inductor with the cable this would block a lot of the electrical impulses. In an installation you don't get that.
Tim Ballam - what rubbish are you talking? First, ALL wires ARE inductors (although in the wire that John used, the inductive value is so small, it can be ignored) . Second, inductance is likely to make an arc worse with a resulting bigger voltage spike. The greater the value of the inductor, the bigger the voltage spike.
@@Mark1024MAK ha ha ha the cable used in this "experiment" is creating a rather effective LC filter that's most probably filtering the 100KHz that the device is looking for!
@@Mark1024MAK ua-cam.com/video/JRXFZHKGaXg/v-deo.html
Is there any actual point to having these in a typical English house?
Just another way to give your hard earned income to the big wigs. These devices are slowly starting to creep into our rule book (AS/NZS3000)
How many are typically in the average house? Isn’t it usually just 2-4?
@@ericspda. I've got 3 light circuits and 4 power circuits at my place. Plus cooktop, oven, air con etc on dedicated circuits too.
A.S isn’t that exceptionally uncommon? Would 2 be the expectation for a typical older house?
Obviously not.
Ya gotta love this guy thanks for your efforts Johnny.
Another great video from you John. I enjoy hearing your opinion and the odd sarcastic comments. Keep up the good work!!
Hi JOHN, I have a 3 phase 10kw solar inverter connected to grid. It has 4 MOVS: S20K300 which keep blowing nearly every month or so. 3 MOVS are between 3 phases and earth and the 4th is between neutral and earth. Which external protection device should I use to prevent these from getting damaged? Currently there is a 6 Amp mcb connected to 4 pole contactor, which is then conncted to inverter. Panels installed are 3KW hence 6 Amp mcb is fine. I am also curious if our grid has some issues or if my earthing has some problems. Apparently earthing looks fine as with multimeter it is measuring 2 volts more than neutral when checked against phase. Appreciate your advice. Thanks
What about doing the arc test with a highly inductive load?
Nice tests, but that lamp arcing has such low energy, would you think that would burn something down?
Someone below is asking about a hairdryer, that would be in the burnable category i think, curious if that one would trip it.
The overvoltage function seems something someone from marketing dragged in to show the thing off :P
Interesting, I was chatting with a colleague about AFDD's and we were speculating some what as to how they work. Having looked into the info from manufacturers its my understanding that an prevalent fault with arc has quite a number of unique current draw signatures and I gather research is ongoing. the device looks for a rise and decay pattern linked mainly with loose connections. Looking forward to seeing more about these devices as the use of them has not been as widespread as it looks like it will be in the future. regards Sean
It would be interesting to see the results of a test like this across a variety of different brands of AFDD and perhaps also put the AFDD into an arrangement with other protective devices (RCD, MCB) to see if inbalance or increased current is detected quicker than the AFDD can detect the arc fault. My suspicion is the AFDD is a pointless addition to a CU, taking up space unnecessarily.
As always, very detailed.
I hope someone from the manufacturers get to see this
283VRMS is 400VDC. Unless the capacitors in the power supply are >450V rated, the device is going to fry its electronics before tripping.
Very good videos so easy to understand, I’ve always had a lot of interest in electricity and I am going to train as a fully qualified electrician 👍🏼
Thanks, John for demonstrating yet another IET sanctioned gimmick.
It looks like Eaton has addressed the nuisance tripping issue very well.
Unfortunately, in real life it is difficult, if not impossible, to tell the difference between a loose wire on a 100 watt load and a drill or any tool with sparking. The problem is more difficult in the UK than in the US because the current level in the UK is only half as large. Tripping on low current loads will probably not be satisfactory in the market place, it's just too tricky.
Loose wires on switches or connectors are potentially very dangerous but nuisance tripping makes people skeptical about the value of AFDD's.
Manufacturers have been working on the AFDD for many years and they still have room for improvement.
You should try to disconnect Neutral wire in terms to cause the arc.
That is not a fail - many light switches can arc if you press slow on them …. And i dont want my power/lights in my house to go down in that case + the risc of fire hazard is negligible :)
Thank you for another interesting video on this AFDD. It’s a fair assessment to say this device doesn’t work. Any chance of you comparing it with another one?
The only circuits I'd be thinking about installing an AFDD on would be showers and similar.
Only problem there is that the guidance (looked at in Part 3) states that AFDDs do not trip on overheating conductors such as the classic shower isolator failure, they only apparently work when actual arcing is present.
@@jwflame In your experience would you not have said you'd normally still get some amount of arcing in that failure?
I've only had the pleasure of looking at one that's failed that way but it did at least look as if some arcing may have occurred.
Thanks for the video :)
What about arcing on the neutral side, would that make any difference?
No - it's AC, same current flows in each conductor, and the direction of that current changes 100 times a second in both of them. Device has no earth connection so would be impossible for it to determine which was L&N.
It should be call Arc DECEPTION device
Fire Fire! Interesting to see the manufacturing test sheet for this!
If you look at the Eaton promo video it shows it tripping in about 2 seconds, with a caption stating "The AFDD+ will immediately trip" - ua-cam.com/video/UtOmau4ym1Q/v-deo.html
Brilliant video John! That’s really opened my eyes. Maybe we will have some future amendments on the horizon? Maybe variable sensitivity? What if several devices are daisy chained on a large installation? ( discrimination type faults) Maybe different modules for different types of circuits? I doubt very much that the device itself is faulty. I think this is a cross the board example of them all in their entirety. Like you say, lighting loads are very low nowadays. A few mA? I wonder if a back feed from a lightning strike would trip it? Lol! I wonder what an arc welder would do? It’s looking like more research is needed regarding suitability. I wonder if these new regs should have been postponed for a while. Your brilliant work has proved so much!
Thanks for the well presented and professional approach to this new subject.
The thing with AFCI/AFFDs is that they are a fantastic safety device IN PRINCIPLE. Unfortunately, they were rushed into mandatory service several decades before the technology was anywhere near mature enough to actually deliver on the promises. AFCI breakers are about as mature a technology today as the early GFCIs were, perhaps, 40-50 years ago. Sadly, they are mostly only good for excessive nuisance trips (the most popular brand in the US, Eaton, is notorious for tripping in the presence of even fairly weak RF signals, so one amateur radio operator in a community transmitting at just 100W can potentially knock out power to entire subdivisions equipped with these faulty AFCIs, and they are out there by the millions now). What the NFPA was thinking when they convinced the NEC to rush them into mandatory service decades before they were a mature technology is anyone’s guess. The reality is that many homeowners, after paying thousands of dollars to outfit their load centers with AFCI breakers, are at some point down the road very likely going to have to pay many thousands more to replace them all two or three more times before they are finally a mature and reliable technology...or they will have to put up with all the nuisance trips that they can produce. After all, having your refrigerator circuit repeatedly deactivated because someone drove by with a CB radio (or any other radio of at least a few watts power) is liable to get old, and costly, after awhile. That and one has to wonder if the circuitry in these was genuinely engineered well enough to provide decades of service without the circuitry itself failing at some point as homeowners do not ordinarily replace circuit breakers for many decades at a time - visit a 40 or 50 year old home and you are likely to find 40 or 50 year old circuit breakers, I just don’t see these lasting anywhere near that long on average.
To touch the copper conductor on line @8:39 whilst plugged in and the device live shows great confidence in having actually remembered to turn off the switch on the outlet...
I have grave doubts about the utility of devices like this, as it seems to me they are either going to be so sensitive that there will be lots of nuisance trips or so insensitive to avoid that, it will not provide any functional benefit anyway. However, I can still imagine the industry lobbying for these just to make more money. If they were to be compulsory, there's also the little issue of space as if a three module width unit is required for every circuit, I doubt many installations would have enough space, and in many properties there won't the enough room in the current location to install a much larger one.
Have seen that experiment (it´s german) ua-cam.com/video/CWqVVDEvd0Q/v-deo.html
He says, that the temperature carbonizes the isolator and this leads to a stable arc which then triggers the AFDD (in the second trial instantly). He states that that is the error condition the AFDD should actually trigger, nothing else.
I understand people asking for a higher load to be tested which I would like to see, but they are not as far as I’m aware marketed as high current arc fault devices.
I personally detect chinesium in affect.
Thanks for the video John much appreciated! I guess they didn't carry out the appropriate tests on the device before bringing it to market!!
Very interesting, and comprehensively tested. Clever looking for DC current when arcing.
I am sure some local council authority (UK), will be convinced, or cajoled by nefarious means to have these as standard fitment in new builds etc. I can see the rationale behind the arc detection, I assume this feature is used in line power applications.
Thanks for sharing.
I am shocked that the device didn’t trip after all that effort . It’s going to be difficult to look a customer in the eye and say they need to spend a £1000 on AFDDs on each circuit when the test results so far are utterly unconvincing. Is it the brand? What about other manufacturers’ devices? Is it just that particular device? I can feel a destruction test coming on John .......? Adrian
Another great vid...thanks.....Have Eaton seen this ....or commented....
I think AFDD actually are cool but the DSP inside makes them way too expensive. Maybe in another ten years.
Perhaps we're all confused - it's actually meant as a device to protect against faults that could develop in an arc welder on the circuit. John must have missed that in the small print or some such. I mean it's not as though he's known for being fastidious now is it! /sarc off
Thanks for the video.
Either that example you have there is possibly a dud, or the concept is.
One of the methods the devices use to detect arcs, is to sense the presence of power harmonics. So in the case of an arc on 50Hz, they are supposed to sense a 100Hz signal on the cable, with the ‘signature’ of a typical arc. Presumably then, the on-board software takes note of the duration and repetition of the 100Hz and decides whether, or not, it is an arc.
I think this demonstration shows, assuming that the example under test is OK, that an AM radio is, at present, a more reliable electronic device to detect the signature of arcing.
Well, as of this week (21/9/2020) the draft Amendment 2 to the UK regs, BS7671 is proposing to make them compulsory for all circuits of 32A and below that are not lighting. standardsdevelopment.bsigroup.com/Projects/9020-04654/#/section is the draft for public comment, section 421.1.7 is the key one. (about 2/3 through the first of the 4 parts) You do need to create an account with the BSI to leave a comment, but anyone can do that. I for one will indeed be commenting, that I think this is not a good thing to add to the regs. Others may care to do the same.
Does it detect an arc that is directly between live and neutral, if that is all it does surely all that is required is a current trip.
Is it possible you have a faulty AFDD as it did not trip when over voltage?
I wonder if Tesla's spark gap machine would trip that thing, considering a obvious spark fails to trip the device. What aload of crap.
Having watched your videos and David Savery's videos about AFDDs I will not be rushing out to spend loads of money on them. From what I have seen, they are not fit for purpose.
This seems to be whats required to trip a AFDD. They seem to be pretty insensitive: ua-cam.com/video/JRXFZHKGaXg/v-deo.html . Seems they are deliberate that insensitive to not false trip when running brushed engines and such. It would be smarter if AFDD manufacturers could include a "learn" mode where you push a learn button and then you run said machine/thing and test lightswitches and so on that is connected to said Circuit, and then it will learn a threshold so it don't false trip but still trip at unnormal arcing.
Same with GFCIs, Personally I think 30mA is pretty high and it would be easy for GFCI manufacturers to include some "learn" mode here too, where it will Always trip at 30mA, but the "learn" mode can be used to "learn" a lower trip level into the system, so on a good circuit with low Earth leakage, it could even be learnt to trip at like 10mA or even 5mA. And in such a combined device, the "learn" button could then be responsible for both Learning AFDD And GFCI parameters into it.
Or even a fully automatic learn mode where it will during the course of a week of normal usage, automatically learn the "normal parameters" of the Circuit (arc signature and Earth leakage current) and if anything gets unnormal, it should of course trip.
Not so sure about that "learn mode", install one on a circuit with an existing fault, it learns to ignore it from day one. Just sounds like a way to treble the price.
The arc fault breakers in the US are generally rated for around 2W consumption. This is kinda absurd since it’s not uncommon to have 15-30 of these in a larger new house.
Now I know why Eaton get so many returns
When we're talking about arc flash, shouldn't we be looking at faults at the kA level and not 40W GLS lamps???
It’s probably designed to trip when used with an arc welder hahaha
Is there a unit of measurement for arc flashes... At what point does the manufacturer claim the device will operate at?
Safe and effective! Mandated!
is this breaker a fake or have eaton gone crap?
I wonder if anyone at Eaton has seen these tests. I'd be dying of embarrassment.
At 11.4 p / kWh one watt costs one pound in a year. That is a good rule to remember when thinking about the costs.
Did not expect those results.
Really! How much R&D do these companies do before letting lose on the world
They spent most of it on booze by the looks of it.
What's the tolerance on these things, +/-100%?
I am informed these devices only detect an arc if the load is > 2.5 amps - so a broken heating
element in an electric blanket arcing and setting fire to your bed is OK !!!!!!!!!!!!!!!!
Not really much of a surprise that this AFDD failed to trip on a 40 watt load, arcing or not... Given that it failed to trip effectively even with > 2 Kw load.
Yet more money grabbing nonesense.
Don't You think they should re-introduce incandescent lamps for Arc Detection!
But seriously though these are deemed as safety devices, and too many people already
think a RCD will protect them if they grab the live and neutral.
I eagerly await EATON's response...
Hokum. As was harmonisation. Fyi suppliers didn't change their incommer colour code.
What AFDDs are supposed to protect against are arc faults that could start a fire. At the same time they shouldn't trip on conditions unlikely to start a fire.
Robert Grant - And how do you tell the difference between an arc that can or cannot start a fire? If we can’t say, how do expect a “dumb” microcontroller chip to tell the difference...
We can say. We know the ignition temperature of PVC, we can calculate how much energy is required to heat the conductors to that temperature and we can measure or calculate the high frequency spectrum created by such an arc and design a detection / protection device accordingly.Such a protection device is not going to be 100% reliable in preventing electrical fires but neither are RCD / GFIs at preventing every possible accident from an electrical mishap . Here is Au the RCDs used for most domestic installs are rated to trip at 30mA. For the majority of humans that will protect them from electrocution but will not prevent every possible death from an electrical mishap. It also means there's some issues with parasitic tripping of RCDs from condensation etc. The general consensus is this is a reasonable compromise.
I wouldn't have said "bullshit". Certainly they will not protect against every possible way electrical energy can start a fire just as RCDs will not protect against every possible way to be get an electrical shock. The same goes for seatbelts in cars and helmets on the heads of cyclists.
You only pay 12p per kwh? What energy company are you with? My supplier charges 21p per kWh.
Currently paying 13p, Scottish Power Online Fix And Save April 2019.
@@jwflame Wow. I think I need to have a word with them. I'm with Scottish power too.
So if you have a Poltergeist flickering the lights, they won't trip off leaving you in the dark? Good to know. lol
The brick wall he had in the background of himself in the first clip looked quite realistic despite the fact I assume it was CGI, as in the last clip we get the real background of his study/home office
That's Johns Victorian period brainstorming room.
I would repeat the test where the voltage increase is sharp and quick. I.e more analogous to an actual "spike" than a slow, creeping voltage increase, which is how you tested it.
I make this comment because, of something interesting I observed that happened recently in my kitchen when I was carefully following some recipe, (Please stick with me on this. It really is relevant).
Well, I was precisely measuring out some salt, and It got to the point where I was almost up to weight and I was adding extra salt practically grain by grain. After a few minutes of the digital scales holding steadfast, I was becoming quite annoyed. "Surely, I have must added more than a gram by now" I said to myself while the scales were still stuck on 3.9g. (I was aiming for 4).
I ran off to get some other digital scales (equivalent accuracy of 0.1 g) and dumped the contents of the first pair of scales onto the second. Well, well. What a surprise: 5.7 grams!
Try it for yourself. Measure out say 10g salt or sugar onto your scales and then gradually add to it grain by grain. You'll be there for quite some time before those scales move.
Discalaimer: It also could be just that my scales are cheap. But it would be interesting to know if the same thing happens on a "quality" pair of scales.
(PS: It's a long time since I've seen a Hospital Saddle.)
its not inly on cheap scales at wor somethimes i have to count leds with a rely expenciv scale with a resolution of 0.01g and a countig funktion some leds wont get counted but if you lift the tray and place it on the scale again it has counted the led
It's probably caused by stiction, ie the fact that a bearing or sliding surface requires a higher amount of force to start it moving than the friction it sees once it is moving.
Except that most electronic digital scales use a strain gauge - no bearings to get stuck. A typical strain gauge is a resistive sticker put on a bar of metal. As the metal moves slightly under load, the resistance of the strain gauge changes. This is measured by a microcontroller.
Will not be fitting them unless I am forced to do by regulations.
Take it apart John I don't think there is anything AFDD in it