Failed 3-phase RCD/GFCI from a Rolec electric car charging pillar. (with schematic)
Вставка
- Опубліковано 27 тра 2019
- One of the most interesting things about this RCD/GFCI is that it is a type-A designed to detect pulsing DC from a mains voltage rectifier. Something that can defeat standard AC RCD's/GFI's in some fault scenarios.
It appears that Rolec have had quite a lot of these breakers fail in the same way, including the single phase ones, and usually on the same terminal.
I mentioned the large test resistor and then got distracted. It's basically a resistor that deliberately causes a test current to bypass the sense coil so it detects an imbalance. When you press the test button on an RCD/GFI like this it emulates a fault to show the breaker is detecting it electronically and then tripping mechanically.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:-
www.bigclive.com/coffee.htm
This also keeps the channel independent of UA-cam's advertising algorithms allowing it to be a bit more dangerous and naughty. - Наука та технологія
The magnetic effects of fault conditions are fascinating. Here they're used to induce a current and trip a breaker. I used to build distribution boards with big copper busbars in them. We had to put supports every 300mm because, under certain fault conditions, the opposing magnetic fields could "cause the bars to slap together like rubber, which wouldn't be good for anyone nearby" as my boss explained it to me.
One megawatt chargers for trucks: "huge potential for exciting electrical incidents."
I see what you did there.
Lots of new employment opportunities for "OH&S" Officers there - trying to reduce the "excitement potential" to an acceptable level :-D
@Dave Micolichek Lets hope the truck drivers will be more careful than the run of the mill driver and they put big pipe guards around the charging post . OH could it be it failed because its the contact furtist from the screw that holds it together ?? not getting tight on the contact ?
@Dave Micolichek A truck stop with 25 charging positions for trucks would need more power than a small city. Or a few acres of solar panels. Gut level feeling is that Tesla will be giving a lot of money back on the pre-orders (for publicity) for the big over the road truck....
What the heck would you connect to that with? 4" diameter copper bars?
A megawatt charger is starting to get competitive with a gas (petrol) pump, although my calibrated memory and a bit of Googling indicates that a gas pump is delivering about 24MW.
@@lasdkfh0la Very unlikely they would go over a kV. I'm guessing maybe around 700 V, whatever you get from rectifying 3p 480V. Say it's a kV. 200 kcmil = 1010mm² = 35mm diameter gets you to 665A. I'm guessing at least 2" diameter.
Many EVs only support single-phase charging (typically 32A) , so the burnt pole may be the neutral - Rolec had lots of issues with burnt neutral terminals on home charging units. My guess is that the spring being at the end furthest from the burnt contact means there was reduced contact pressure
Looks like in this one the burnt one is marked 1, with the neutral being at the other end, next to the switch lever
Yeah, I'd guess a mechanical imbalance too, if it's true that it's always the end contact the burns out.
The load was mostly single phase 32A, the L1 got most of there work to do. With these RCDs the neutral always seems ok though (despite the single phase load) but the single phase versions burn up even more quickly. My guess it is a combination of contact pressure and high resistance metal used on the contacts, they are just not suitable for 32A for 7 hours each night, night after night. This one lasted 3 years, but the new one that Rolec sent is a different design and I've not heard of the newer ones suffering this issue.
@@G8YTZ
Hmmmm. Depends what you mean by *mostly* single phase.
If it were a balanced 3-phase load then the neutral current ought to be zero. If it were a pure single-phase load then the neutral current ought to be the same as the L1 current. Anywhere in between balanced and L1 alone, the neutral current is going to be less than L1 current.
Doing phasors in my head (and probably getting it wrong) a 32A L1 with 16A L2 and 16A L3 is going to result in 16A in the neutral. 32A L1 with 31A L2 and L23is going to result in 1A in the neutral. 32A L1 with 1A L2 and L3 is going to result in 31A in the neutral.
The L1 current could be significantly higher than the neutral if it's mostly single phase. Depends what you mean by "mostly."
@@bdf2718 The supply and charger are 3-Phase, but some cars have three phase chargers and some cars (the majority) have single phase chargers. This is why 4-Pole Type A RCD's are used as the neutral is required for single phase cars. (Type A is to detect pulsing DC faults on the DC side of the carer as Clive states).
So if you plug in a single phase Car (for example a 60Ah BMW i3) then 32A will e drawn between L1 and Neutral. (94Ah and 120Ah i3's have 3-Phase chargers, if their connected to a 3-Phase supply they will draw 16A/phase (12kW), if connected to a single phase supply they will draw 32A (7.4kW).
Most EV's have single phase chargers, hence the term "Mostly" so the L1 contacts will be more stressed rather than L2 and L3. The neutral contact in these RCD's and in the single phase version of this charger type for some reason never burn, but the L1 (or L contact in single phase version) contact always burn, even though they are running well within their rated current, but the very latest examples seem to have improved contacts.
A Tesla car also have 3-Phase charger(s) Depending on the model and age either 16A, 24A or 32A, but they all draw a maximum of 32A when connected to a single phase charger.
Note the function of the charger's Control Pilot pin (set on installation) defines the maximum charging rate according to the circuit and/or charging cable capacity. The car takes its loading instruction from the CP pin.
About the lack of active circuity: A good RCD is supposed to trip on the current difference regardless of the operating voltage. So the tripping trigger has to be powered by the pickup coil only. This makes it very delicate and expensive. In the cheap RCD the current from the pickup coil is fed to the thyristor which uses the main voltage to activate the electromagnet of the tripping mechanism. This makes the mechanism much cheaper but won't work without proper voltage. This type of RCD is not allowed in my country.
Anyone else notice the fantastic sound quality. It's like you're sitting beside me big Clive
Dear Bigclive
I also opend up an older RCD (already a DC Pulse capable type, same logo as yours) which had a similar PCB inside with Diodes and Tantalum capacitor, but there the sensing winding on the maintransformer had a connection in the middle of the sense winding, which was connected to the minus of the releaser winding. The 2 diodes formed a half bridge rectifier, for positive and inverted negative pulses to charge up the capacitor with a positive voltage each and it was connected to the positive connection of the releasing coil. The capacitors negative side had a connection to the minus of the releasing coil too. The releasing coil weakens the magnetic field of the little permanent magnetic pill. The magnet, which sits on top of the coil when the contacts are closed, and holds the mechanism, separates from the coil and the RCD will trip, if any pulses charge up the capacitor. Regarding the burnt contact you have to keep in mind that this RCBO sits in a housing where moisture in the air is one topic, due to the fact that the charging post usually sits in this box in ice cold and sunny hot weather, unclimated. On top of this you have a high current load, which will heat up the wires with every charged car. As mentioned in annother comment some cars only use a single phase, so the wires for phase1 and neutral will get the highest tear and wear. Together with a maybe weak torque while mounting the cable into this cage clamp the cable corrosion can start in the clamp, with more oxidation on every charging cycle, and therefore it gets everytime a little bit hotter the more the oxidation advances. Furthermore there is annother topic: The contact in the RCBO has a silver coating which can get the black corrosion on it. This black silveroxide has a high resistance. In a stairway timer light switch the corrosion is pushed to the side or burnt away with every triggering and release. But how often is a RCBO contact opened to have a cleaning effect on the oxidation layer?
Yes, they are tantalum smd capacitors. The bar marks the positive end, because, why not do the total opposite to electrolytic caps, where it marks the negative ?!
Trap for young players, eh? Me included... :D
Jonathan Kay - through hole wire ended tantalum capacitors have always (well, mostly) had their positive terminal/lead marked with a + (plus) symbol. Through hole wire ended electrolytic capacitors normally have their negative terminal/lead marked with an arrow that incorporates a - (minus) sign. But older can type (non-PCB mounting) electrolytic capacitors have used a variety of marking systems (often with different coloured paint spots). I don’t know why, but it has been this way for over 35 years.
@@Mark1024MAK the likes of you and I, who've been in the business for 40 years, know this stuff. But "a trap for young players" as Hanetar and the mighty Dave L. Jones would say. It was more an attempt at irony, on my part.
Jonathan Kay - Ahh! The thing that does trip me up is the extra small and hard to read labelling on modern components, especially SMD... And some chips now don’t appear to have pin one markings 🙁
Ive never understood why electos mark the -ve. Schematics n pcbs always mark the +ve. Odd...
Reverse polarization on a tant is rather energetic. A mate was having trouble getting reliable ignition with his candy rockets with his DIY ignitors. For whatever reason, at work we have about 30,000 small 6v tants that are not used in any design. Gave him a few to try. Given 12v reversed, they rather pop. 100% reliable, and instantaneous, unlike his resistant ones. Rather expensive if your buying retail, but definitely go off.
If the gardener can kill the protective housing, I expect the whole device is built to a very low price, to compensate for the price of installation.
i'm now wanting to know what would happen if dogs start pissing on them, would it be a new hot dog cooker for clive?
It's not just weed whackers either.
My local council managed to mow to top off a power company pillar with a tractor mount mower.
I happened to be watching it happen, the weeds were taller than the 1 foot tall plastic box, so the mower reversed in and mowed the top off everything.
Definitely made contact with street mains, I found a 100A fuse and fuse holder with blade impact marks.
@Dave Micolichek No flash as is only connected to one phase, lots of loud crunching noises though from the now lidless plastic box.
@Dave Micolichek I too was once near a transmission line that was hit and broke. I didn't witness it, but sure as hell heard it. Years ago I worked in a call center that shared a building with a wind turbine blade manufacturer. The place was located in an industrial park on the edge of town and a transmission line ran above the parking lot. This place makes 100+ft long turbine blades, and while loading a finished blade onto a semitrailer the telehandler they were moving the blade with hit the transmission line.
The driver wasn't injured, but the telehandler was utterly destroyed, the current even blew the tires on it. Also dropped and broke a $250,000 turbine blade. We must have been upstream from that transmission line as we never lost power, but when that line was hit every CRT monitor in the place lit up like a christmas tree.
Yes that was quite the display of power, especially when you consider that only one out of three phases was hit. This was around 15 years ago so we were still using CRT monitors for the computers; when the line broke every monitor started flickering as if you had activated the degaussing coil to remove magnetization from the screen. Surprisingly it did not seem to have any effect on the computers themselves, they must have been sufficiently shielded.
If the actuator pushes on the far right end of that shaft, and the contact on the far left is messed up, my gut instinct is the shaft is twisting a but too much and not applying sufficient force to the contacts. Smells a lot like a basic design flaw to me.
@Dan Bowkley I'd go even further and say that the shaft shouldn't be made out of plastic.
I would have thought the individual contact springs might have helped. Unfortunately it's hard to test without getting one that's not pre-melted. But the burn pattern on the contact hinted at an angular connection.
Bingo. Nylon in high humidity = significant creep deformation rate.
Also wondering about the single screw in the back. Would over-tightening that cause the innards to be (slightly) out of alignment?
good point, that could cause slight make/break delay between them also.
Clive, great video! It's a three phase unit, but many cars only use singe phase so L1 gets most of the work to do.
Good to see that your hand is healing up nicely!
As per usual.
Very interesting Clive.
Thanks for the video, love the way that the trip mechanism works.
Bet that solution did not come to mind overnight.
As for the failure of the end contact, I would guess that contact pressure on the end contact is not high enough and it runs hot.
But the video stimulated my thinking about a problem I have to solve at work and now I have an answer for the folks tomorrow.
We work with variable frequency motor drives and ground fault detection on the load side is tricky as the fault current on the ac side of the vfd is dc due to the nature of the conversion process, when a ground fault occurs on the output of the vfd.
So a core balance does not work in that situation.
But, a transformer type core with two winding's will work, because the output voltage from the measurement coil changes when dc current magnetizes the core.
Is what we get up and go to work for in the morning, dealing these interesting and stimulating works tasks to deal with.
And then sharing the experience with others, like you and other creators on YT do.
Kind regards, D
The little trip plate is surprisingly similar in design and function to the trigger and hammer sear in a firearm's trigger group: a very small, weak motion in one direction unlatches a catch that allows a much larger and more forceful motion in another direction. Cheers!
This reminds me of when I used to work in tv repair - Back in the 80s we used to see a lot of those old Ferguson TVs that used the Thorn TX9 chassis. These had an unusual thyristor-based PSU, and the most common issue with those was a burned up connection inside the mains plug, and it was ALWAYS on the neutral pin. Usually just cleaning up the connection (or in bad cases just replacing the whole plug) would result in a long-term fix but it was a bit weird at the time why it would always be the neutral end that burned up rather than the live.
So much mechanical components and magnetic hacks which are life saving - scary.
Interesting. AC type RCDs are forbidden in Germany because they don't detect any DC faults.
That's why it failed here.
You actually should use a type B RCD for charging stations.
Unfortunately B type 40A/30mA costs around 400€ compared to A type 40A/30mA which costs around 30€
@Dave Micolichek maybe it wasn't installed properly, that's the only answer left. Nothing was wrong with the inside, you seen it in the video.
This breaker wasn't in the feed to the charging pillar, it was inside. Granted, type B are much more expensive and for obvious reasons but the threshold is still mA. If you suddenly apply a fault current of *many amps DC* (as would need to be the case for this damage) to a type A, it will trip. Maybe even a type AC would.
With the activator mechanism on the right, i wonder if there was enough twist in the bar so the left contact isn't as tight as the others.
Others have commented on that too. I would have thought that the individual contact springs might have prevented that. The contact burn hinted at maybe a slight angle of the contact.
Torsion was near the top of my list as well,
The neutral gets interrupted first in this case, usually the neutral is done last for safety reasons, so the current is balanced over the three contacts, the last contact collects all the current in this case
Torsion indeed, AvE said last week "Everything is a spring". If you had a way to accurately measure the contact pressure of each of those 4 contacts when engaged, I would bet that the pressure decreases on each contact as you get further to the left.
The little bit of insulation fallen off blocked the left hand contact causing the twist to get worse.
Most likely reason is that the first contact is furtherest away from the actuator, and the flex in the plastic is enough to mean the contact pressure is low. Combine it with the stiff cable and the contact pressure is low, so the resistance as the contact ages rises, due to not enough pressure to wipe off the oxide layer, so the small area overheats, and then results in more oxide forming, and this then heats up more as the actual contact in electric contact reduces.
With the electronics, I have some old breakers from the 1970's, which use a BRY39 thyristor to trip the coil, using the current generated by the sense coil to charge up a capacitor, and then use the thyristor with a 6V8 zener diode as a fast pulse generator to fire the trigger coil, though there is also a parallel path with some 12V zener diodes for fast triggering on high current, so that the time to charge the capacitor up is not going to cause a delay on short circuits. Leakage current it will trip within 2 full cycles, but large leakage ( like a short to earth) it trips on the same cycle. Nice application of a 4 terminal thyristor, pity they are not really made any more, like unijunction transistors, but at least you can emulate then somewhat using only a NPN and PNP transistor tied together.
(Sorry if I’m spamming this, I posted a few places)
I started watching your videos a month ago, I stumbled upon your “A Simple Guide To Electronic Components” video when I learned about resistors after blowing all the LEDs in a 6v firework igniter box I was building.
Point of the story: I’m 31 and have worked in kitchens (and one porn dvd rental shop) my whole life, I’ve never had any career I wanted or any hobby that could be monetized, until now. I have always liked electronics but now I am in love! I watch your videos constantly, most of it is a foreign language but I’m slowly picking up on it. If all goes well I will be going to school in the next year for an electronics technician. I have already converted my workbench over for electronics and have ordered a half dozen kits off of Ebay.
Thank You Sincerely!
thrice1888
Great story there, great Oaks from small acorns grow, I hope you become an engineer extroardinaire!....and I've got loads of things need fixin', stay in touch!...just kidding but good luck and bon chance!
John Salmons Thank You Sir!
@@thrice1888 you reminded me of me a bit, I was a painter and qualified as a sparks in my late 30s now I own my own company and test and certify domestic to industrial installations, keep at it bro!
Wonderfully informative - thank you! And what a fantastically authoritative, yet lilting voice. Warmest regards and thanks for the insight. Eddie
I wonder if it's usually the left-hand terminal that burns out because right-hand electricians are installing them.
More likely it's because most EVs charge of signal phase.
@Andy Peek Andy Peek my theory if you tighten the termination up with a screwdriver in the right hand, holding the wire in the left, you may run out of room on the leftmost position against the housing, leading to a poor termination, leading to a HR connection, overheating and cooling. I think Conor is right though. First phase is the most loaded phase.
I think a poor design of RCD. Even if only the first phase is used, the RCD should be more than capable of handling the current. It should not overheat (assuming correctly installed) at its maximum rated current. And it is rare that such devices are actually ran anywhere near their rated current.
An occasional loose wire/terminal may well cause failure, but if as indicated, many of these are failing, that, to me indicates it’s the design or manufacture of the RCD that is the problem, rather than that of the installer.
They probably needed racing electricity. 87 electricity has too much resistance for the electric race cars.
Lol!
😆
But with less anti back EMF you get more current at high RPM.
It would be interesting to compare the Rolec design with other 3-phase RCD's, like Schneider or Eaton
Let me just pause while I take it to bits... Cut to explosion of metal parts... Mmm should have done that slower.. 🤣😁😂
the cut to the whole thing being apart with parts strewn all around the bench like everything just exploded made me audibly laugh out loud
I believe the green sleeve, being constrained by the housing, reduced the contact force, making it get hotter until it finally arcs. I've seen similar problems before, always by something external reducing the intended contact force.
Nice mechanical setup
Love the brick!
Scott Davis
Trump is a Common Criminal.
@@makeracistsafraidagain I'd wager much worse than that unfortunately.
ua-cam.com/video/R1ZU6UMDfgY/v-deo.html
Had a home Rolec charger. They are notorious for this fault. Replaced the RCBO myself. Apparently the newer ones are better.
So I've had a Nissan leaf for just over 3 years now. The high power CHADEMO chargers (400V, 120A, ish) have their own cable fixed to the unit and it does an insulation test before starting.
The smaller, 3, 7 &22kw chargers you usually use your own cable and no test is done.
I believe the contacter in the charger is fired when it gets a 9v signal from the car for EVSE chargers.
Ironically the charger is actually in the car so the things on the street, despite being called chargers, are just fancy sockets
DC chargers (CHAdeMO, CCS) are in fact chargers, not just sockets.
Points for actually using 'ironically' properly.
Yes quite right, the CHADEMO is more or less a direct connection to the battery. I rarely use them in fairness
Great video Clive
Obviously too much current on the contact. Raisins are 20% more conductive.
Johnny Be Lost... 🤣🤣🤣🤣 You made me pee my pants with your comment. I'm smashing down the BIG RED Emergency Button, that Clive showed in a video as an engineer's puzzle. I keep it next to my laptop while watching these videos to release frustrations on the funnies, dealt out by the likes of Clive, AvE, This Old Tony etc... and now by the likes of you.
Had to smash it every time I heard Clive say HAWAII instead of Waa-way-ee, which is the correct pronunciation of Huawei and also his in-focus or not in-focus and jhoom-in and jhoom-out mentions, while he's actually out of frame sometimes😂‼️
On the SpeakEV forums there's a large thread on failed Rolec RCBOs. Almost always the neutral and the seems to be a wonky angle where the contacts don't make good contact. Rolec blamed installers not tightening the screw terminals correctly but they have since issued a new green coloured switch version. Clearly made from better chinesium.
A possible reason why it's always the left hand contact: The switch is on the right hand side, therefore when re-setting the breaker, the mechanism is tilted slightly due to the switch being off-center, causing the left hand contact to arc slightly.
Hi Clive. Put a straight edge along the black plastic contact carrier and check for signs of twist. The burnt one is the farthest from the tension spring so has slightly less contact pressure. When the power ramps up the contact gets warmer and begins a self perpetual cycle of further reducing the contact pressure until it finally fails. My two cents.
Actuating shaft for the contacts twisting as it gets further from the toggle?
The neutral is probably loaded to the max and over, if you have stady current in every of the 3 phases the neutral current is the square root of 3 times the phase current. Then the neutral current can be higher than phase current.
Maybe worth feeding the circuit from a signal generator and seeing what does/doesn't make its way through to the output?
How about connection screw slackened off with heat cycling or wasn't properly tightened in the first place?
I think the end contact burning is due to Newton, it will take just a fraction of a second longer to breake contact due to inertial forces causing twisting of the contact bar. All the actating forces are at the other end of the bar so it will twist a fraction as the contacts are opened so the far end will be the last to open so the system will try to draw all the power of that one line.
Another name for the arc quenching part is an "arc chute". Just an fYI.
Thanks!
As usual, when I'm in the middle of a video I forget names like that.
@@bigclivedotcom No problem.... just wanted to help! Keep up the great work!
Is all the return current from the three phases going back through the burnt contact and overloading it?
The burnt contact is labelled "1" on the box;"N" is the right-most contact. It might not have been wired properly though......
And perhaps the supply into the box is only single phase, or only one phase is used in the vehicle.
Clive, the failure is most likely due to poorly attached wire in the terminal (externally), or a piece of insulation that wasn't properly removed... I see a lot of that out in the field, and it manages to cook the contacts the same way. Done a lighting relay last week that welded itself shut, because they didn't bolt down one of the terminals, and the wire just welded itself in place and eventually cooked the contacts.
That terminal looked OK. The main point of heat looked like the contact.
at least here in germany a RCD type B oder type A-EV, which are sensitive for (continous) dc leakage current are mandatory for EV charging stations. These are really expensive. They cost 10 times much as a type A costs.
Should have a notched hole to let the wires flex a bit more as well as centring the main switch to evenly try and displace pressure evenly along all contacts.
Your "Tramp Brick" is hilarious and clever.
Here in the US there have been problems with overheated neutral in 3-phase systems due to poor power factor caused by electronic power supplies.High frequency harmonic currents do not balance out causing the neutral to carry higher then rated current.
to me I think it is the spring is one end and the fault is the other , is it possible that it gets twisted over time and therefore that contact would be loosest and likely to arc?
Perhaps the edge-most wire gets pulled on and has insufficient strain relief, increasing the terminal resistance and producing heat?
I'm guessing here, but if one contact breaks first or closes last (compared to the others) it will arc badly. If there's something loose (or just plain cheap) in the pivot shaft, it can do both - break first and close last - double the failure fun. This is a huge issue with high voltage, large capacity breakers - so much so that industrial versions have special snap mechanisms that slam them open and closed fast and along with arc suppression coils that quench any arc quickly. Even so, every time I've serviced one there's always one contact that's a bit more pitted and burned than the others.
torsion on the contractor bar, it latches from the right side, meaning the left side has the least force keeping it in place, and most torsion on it.
RCDs are known to open or attempt to open under extreme overcurrent or short circuit conditions. So if the complete charging system (pillar plus cable plus car) causes a large current spike when plugging the car in, out or operating the motor while plugged in, this can cause more tripping than normal. Also, some RCDs are only rated for two such events in the lifetime of the RCD (there's a sticker telling homeowners to call an electrician if it trips from short circuit and to never reactivate it a second time).
So if the regular operation cycles that scenario as often as once per month, contacts designed for normal RCD use may simply wear out.
But why wouldn't the other sets of contacts also show deterioration in that case? The rest all looked fine to me.
@Cupid Stunt
This is not supposed to act as a fuse (overcurrent protection). The transformer primaries are fed by the live and the return path so it's the same signal shifted 180° so the sum induced in the secondary is zero. If someone or something with earth ground potential touches the live wire, the return path coil in the transformer is circumvented and thus, a current is induced in the secondary which trips the breaker.
However, sometimes in case of a short curcuit the manufacturing tolerances of the primary coils are enough to induce sufficient current in the secondary, which trips the breaker involuntarily.
or in the case of 3 phase, 120 degree phase shift :)
John Francis Doe - this device is an RCD not a over current circuit breaker. It normally will only trip if there is an insulation fault, or other current leakage to earth/ground. So unless there is such a failure at the same time as a full load current, or a short circuit (most likely due to deliberate damage to the output cable or connector), it will not be opening at full load or under short circuit conditions.
I think there is either a design flaw or a manufacturing problem. As it is unlikely that the current in normal use will be anywhere near the rated current of this RCD.
Could the neutral-wire, due to bad installation by the electrician, have been fed through "1" terminal on the RCD ? - combined with third-harmonic * currents from the charging-device raising the total current above 63A nominel current through "1" terminal?
* third harmonic currents gather in the neutral + third hamonics happens because of rectifying from AC to DC (just like in a frequency converter)
I know its a longshot :P
I wonder if because the damaged terminal is furthest from the trip mechanism, it opens slightly slower and causes more arcing and burning?
Is it not the neutral that is burnt? It handles return from the 3 other phases when using it for 230v and therefore more current?
Mayhaps that contact isn't held down as strongly because it's at the end of that plastic bar?
The breakers likely pull power, one after another from left to right mainly, which is why you'd see the left burnt/blown first on most units
Don't know if that's the case, but it's a thing to think about.
Are type-AC RCDs allowed to be installed in the UK? I'm asking, because here in Germany those devices are prohibited; Type-A RCDs are the bare minimum here... Greetings^^
@8:45 those are tantalum capacitors.
In asean countries, we call this RCCB or ELCB. Even tho they the same thing but you can see generic version or the cheapest version os designed like this. Tye cheapest ones after 2 years or so and you press the test button it will not trip.. Thats why i use high quality stuff for my projects
A guess would be when you reset this style of RCD the contact that is burnt is the last one to close (by a few ms?) because the lever is operated from the other side of the unit, the bar inside flexes, but in this application I can’t see that item being operated very often nor under load, so maybe other factors but that’s my guess. I have seen 4 pole breakers that have been pushed up from one side destroy equipment as the neutral connected momentarily after the phases. We confirmed this by testing it afterwards - was on one of our genset hired to a customer.
Due to harmonic distortion more current flows through the neutral than the phase conductors, (not be confused with power factor). This is why the DNO are oversizing the neutral on new installations, eg 50mm phase conductors 70mm neutral. All modern switching electronics from light dimmers to switched mode power supplies cause harmonic distortion.
Bigclive, a fully articulated passenger bus caught fire here while connected to it's overhead charging port, and was unable to disconnect from it while on fire (might have fused itself to the connector). Fire department was unsure whether power was still on and decided to wait for the power to be killed, all the while the back of the bus was burning and they watched (American kind of situation, actually).
I only got a couple of pictures of the bus on fire from a colleague, so I don't know how violent or interesting it really was, though it was a troublesome thing to see because the bus was merely a year old...
So I have a theory on why it's always the far left terminal. It is the furthest from the switch, so the flex in the mechanism would cause it to be the last to move away, and it may not move as far. Allow more arcing then the others.
If this is an RCD without the ability to trip on overload, I wonder what current the actual overload circuit breaker was rated at? Just wondering whether it got into a situation where the current was too great for the RCD but not quite enough to trip the actual breaker, so it sat there slightly above its rated load getting quite warm over some time.
I think this is spot on, but could also be that most electric cars are pulling from the first phase, so more continual wear on that one side. Could be a combination of things. We'll probably never know.
Keep in mind that the amp rating is affected largely by the need to quench the resulting arc when opening under load and the need to avoid excessive contact pitting that would degrade contact resistance over time. Given that in normal operation these contacts should spend all their time closed, which is also where they have the minimum possible contact resistance, minor overloading, even long term, really wouldn’t be expected to allow things to get this hot without some other defect in the design of the contacts or their closure mechanism - which a single consistent contact position failing would tend to suggest more so than long term overload. You can, for example, routinely push far more current than the rating plate suggests through a switch or relay/contactor if, and only if, you can guarantee that the contacts will never be opened or closed under that elevated load (note that I am not suggesting such is good engineering practice, but it is a possible thing to do). A well designed contact, particularly one with little age and wear, should be able to safely handle many times it’s rated current without overheating so long as it is never opened or closed under that elevated load since there is no arc or arc flash concern to degrade it. And if the charger as a whole is well designed, then the RCD/GFCI breaker should never be tripped unless something has gone horribly wrong, so there should be little routine wear on those contacts that might increase their contact resistance. So, to myself, it really screams “design defect” in the RCD/GFCI breaker itself.
its rating is 63A but the most load is 22kw or 7kw per phase so not overloaded . They just like to blame the spark for not terminating correctly , but i have my doubts about there quality , it would be interesting to compare a few under load with a thermal camera.
@Steve AtTheVanFirst off that was a three phase point so would've been a 22kW post.
Nissans may not draw more than 6.6kW, but Hyundais, Kias, Teslas, Renault Zoes and many other EVs can draw 7kW or more on AC. Some Teslas and all Zoes can draw the full 22kW from a type 2 connection.
@@ethanpoole3443
With relays/contactors/breakers the carry current is usually a lot higher than the break current. Which often doesn't do you much good, because it's usually the break current you're worried about.
Some relays and contactors, though, may have a much higher one-time break rating. Not as high as the carry current, but quite a bit higher than the current they can repeatedly break. There are applications when you need this capability.
Looking at it and seeing that the springamathing is on the other sideo of where the burnt contact is, it could be the contact with the least pressure.
since its the rcd and no switch/fuse it shouldnt have wear from on/off arcing. (of course it gets tested each month...)
aaand perhaps over phase L1 there is running more power than over 2 and 3. or even only on L1 depending on car charged.
Bad connections get higher resistance. Higher resistance means more heat. These failures, in All power distribution panels can be attributed to a starting bad connection
Wonder how branded one look inside.
I have type A ones installed at home, they are Hager brand and cost a lot more than this Rolec
That should not really do very much switching so I'm left with guessing between some sort of vibration mode parting the contacts slightly or just a poor contact with not enough pressure on the it to cause overheating during high current draw.
I think low contact pressure is a good possibility. The contacts are held closed by a torsion spring at one end of the molded bar that carries them. The failure of the far terminal is likely due to creep deformation of the bar (or its bearing surfaces in the case) resulting in low contact pressure. The deflection (and pressure degradation) would be the worst at the end furthest from the spring. The creep rate of nylon is heavily dependent on atmospheric humidity. Even glass-filled nylon is prone to creep deformation at elevated humidity. The effect of self-heating would only increase the creep rate. I can't really see if there's any noticeable twist to the part. I'm sure there are other possibilities.
That said, creep deformation is something that some people never seem to think about in design. It's the primary reason why SATA power connectors are prone to causing arc damage to drives. It's also why data connectors tend to get loose and fall off. Adding retention clips won't fix the problem if the contact pressure is zero.
I was also thinking that the conductor, being bare braided copper and just shoved into it might have possibly shorted out along the length at different points and over time caused the conductors resistance to change just enough to cause a problem.
What would it short against? IIRC, the contacts (and wire terminations to the moving contacts) are separated by barriers. Any sort of short would be a catastrophic failure mode, not something that happens gradually.
Hysteresis seems a good possibility. Or secondary effects due to the mechanical damping on that portion of the device attributable to the clamping of that section of cable. Otherwise, it _is_ int'resting that this device seems _reliably_ to fail in the same way at the same location. By _very_ clever design? Dunno.
When I was a nipper the points in my car were platinum coated. Over the years this coating got thinner and thinner. So whereas you could have sanded your points nowadays there is almost zero plating on contacts.For example, Italian made fridge thermostats will fail after 13 months.
Convenient then to have a 12 month warantee.
@@jimnelsen2064 You are dead right! I don't have the equipment to test this but I'm guessing that these contacts are minimally silver plated.
Might be because the set of contacts are driven from the opposite end. Any slop or deflection on that shaft could result in less pressure on the far side contacts.
A stab in the dark, AC->DC we are well aware of but the AC->DC->DC thingy with highest current/best efficiency possible maybe a new ground?
All high efficiency DC->DC conversion actually uses DC->AC->DC (the AC component being pulsed DC). Modern AC mains powered DC power supply units are actually doing AC->DC->AC->DC.
Probably burns that contact because of play in the mechanism, with the activation system being at the far end any play could result in a difference in the contact action, it is probably the last to make/break.
I'm actually not suprised about the circuitry. I have recently disassembled an old RCD that was installed around 1985. It was kinda the same mechanism. The trip coil was connected direcly to the detection windings without any circuitry at all.
Meanwhile in Sweden most of our charging poles are galvanized steel lol. Try chipping that with a weed whacker xD
Im waiting for America to use corten steel for charging stations.
Can’t beat hot dip galv for durability
@@DurzoBlunts I think I saw someone manufacturing chargers that are robust enough to withstand smaller calibre rounds fired at it.
Is it always the contact that's furthest from the switch actuator? Maybe the springiness of the plastic causes it not to close and open as positively as the contacts nearer the actuator.
I don’t like the way it armature picks up the contacts from only one side and not the center for balance. The contact on the end may not be getting enough wipe due to lack of rigidity in that arm type arrangement. Insufficient wipe (contact pressure) can definitely cause this problem too. I like your observation on the shunt being impeded and not being able to move freely. That seems to be the most probable cause.
You say the left hand one always burns, on 13 amp plugs the live terminal screws always seem to loosen themselves,strange.
I wonder if the reason the left one fails most often is that the geometry of the pivot is a little sloppy on that end, slowing the break and allowing more arcing. Or, perhaps the slop makes that end a lighter connection with more resistance.
I almost wonder if the station is wired correctly inside. could one of the phases be under a higher load where the RCD is connected to the source directly instead of through a master breaker, and the charging circuit and control circuit combined are close to exceeding the rating of the RCD, if not exceeding it?
Well in Germany Type A RCDs are required for all circuits with socket outlets under 32A that are in the reach of non electrical personell. Type B is recommended in circuits with VFDs, inverters, charging poles for electric cars and such things. Type F is recommended for circuits with single phase VFDs like modern washing machines. Type A is quite cheap here, like 30€. Type B is like 300€ upwards. Type F is like 100€ upwards. Type AC is forbidden here in Germany, if we see such a thing we have to shut down the supply until its changed, because Type AC is a health and safety problem and people are in danger. With all the modern electronic in the houses or businesses today the Type AC can't protect sufficiently enough. I'm a little bit shocked, that RCDs Type AC are common in the UK.
Maybe it's whatever goes into one of the contacts? Like one lead or bar being aluminium for some reason (price) and running into contact issues?
Hi Clive, Any chance of a video about variable frequency motors ?
Wife liked film about mosquito !!!!
Regards John
The fault could be the fact that the left most terminal is the furthest from the "breaking torque" on the other side. Meaning when tripped or going into the off state the twisting force isn't enough to fully or properly disengage the left most pole. Perhaps why it's common to see the left side with problems. BTW, I think it's pronounced Wah-Way.
When thinking about the failure of the single contact: these contacts shouldn't be making and breaking very often at all. They should only open on a fault - right? And when they close they shouldn't be under load. There's another contactor that does the normal on/off function. I'm going down the path of a bad connection overheating. Was there a "fault" in the pillar or cable/vehicle that caused this device to trip, or was the only problem the failed breaker? Maybe (like you said) the cramped space for the wire was preventing the contact from seating firmly - causing it to over heat.
A bad contact causing a low-value resistor in the contact block causing heating is what I was thinking as well. Stands to reason that if it were a bad physical piece of contact metal it would be on at least two of the block sections (due to identical contact components) but it's just the one, implying bad seating or the like. As for why? Either bad assembly in a batch of those things or a bad design. Hard to say without difficult or expensive scanning while fiddling with it (like x-ray or some other radio scanning at the internal mechanism while assembled).
@@IncertusetNescio Could a thermal image be able to see something through the outer cover? We use to do thermal image inspections of control enclosures and switch gear on the US Navy subs I was on in the 90's. Thermal imaging has come a long way. Our rig was literally something you wore. It had a CRT monitor hanging from your chest!
@@SueBobChicVid Modern FLIR has pretty good resolution, and I'd tend to agree that this would seem an ideal application. We need BC to get the FLIR out and have a look at another failed module (if at all possible).
Did it break and not work afterwards or did it failed without tripping? Maybe it was just an too high current from a short on that channel while tripping?
nice explain... can't wait to make a party and invite all there, meet and drink and enjoy....(selfpay)
The focus was fine, btw.
Rolec RCBO,s have always been on the very cheap side compaired to other makes and the the caravan hookups the same.
I would guess the spring that holds tension on the 4 contacts is either not strong enough when the RCD gets hot or the arm that hold the 4 contacts slightly twists causing less pressure to be applied to the final contact
Arc breaker, your time has come
Can't take your evil way
Go away Arc breaker
Arc breaker
Arc breaker
Arc
Followed by communication breakdown. ;)
Vogon poetry?
Perhaps there is a slight delay when the circuit is broken and the one on the left is the last one to break. That may spike the current on that leg as the load is still trying to draw current.
very interesting, at 2:55 the tester lamp goes on while you sting it into the back hole of the rcd...
I've been involved with electricity and Electronic's since age 5. Here going 62 I'm somewhat taken back at the complexity of these things. I always assume they were all just a bi metal that heated and warped. LOL Exception of the GFI. That I figured did in fact hold some Electronic's wizardry .
It's FM (Freakin' Magic). :-)
@@mcgyver272000 Yep! And Angry Pixels.
Indeed, the answer is simple, they let out the magic smoke, that which makes all things electric work.
The control circuitry may be running off the left hand terminal, the control circuitry might draw a lot. Have not seen any of these charging stations, do they have a street light or anything attached?
Well Clive if you're puzzled then I've got no chance. But did you check for gremlins?
Your profile pic is disturbing 😰
@@mrfluffytailthethird
So is he dark side of the force.
Hey Clive , back in the late 80’s early 90’s prior to the introduction of RCD’s in Australia, we were using voltage operated ELCB’s which used two earth electrodes at a predetermined distance apart. Are you familiar with this type of protection. The concept of operation was that an earth fault would cause a potential difference between the two electrodes of approximately 22volts which would cause the ELCB to trip. I was only a first year apprentice at that time and didn’t really understand. If you are aware of these products please enlighten us. They weren’t around too long probably because of the complexity’s of there installation and the simplicity, reliability and effectiveness of the RCD. Cheers jj
Exact same thing happened to me on the single phase version. Burned out left hand neutral terminal on an EV charger
We had the same problem with a new apartment in China. Twice they replaced a Siemens RCD before ripping out all the wiring and replacing it with better quality not a small job. Might be a bad batch of cable?
what is inside and how it working sony crt tv