ฝัง
- เผยแพร่เมื่อ 13 พ.ค. 2024
- There's been a lot of fuss recently about bidirectional RCDs and how renewable generation must use them.
Most of this fuss seems to have been coming from Proteus/CEF.
Is this just a marketing thing? Am I missing something key here?
I have two type AC RCDs here, one with clearly labelled "in" and "out" terminals, one without. - วิทยาศาสตร์และเทคโนโลยี
Yes, not surprised by your findings, the RCD protection will continue to function if live and neutral are swapped or designated input / output terminals are swapped. The issue is around voltage remaining on the output terminals of some RCDs when there is a fault still present that causes permanent energising of the trip coil / electronics and may lead to their failure. The likelihood of this is quite slim as the G99 protection shuts down the inverter on loss of mains.
There is a document from BEAMA that pushed for the amendment from the IET, can't leave a link to it here. That also claims that MCBs may not have the same arc extinguishing properties and/or the short circuit rating is reduced if they are connected incorrectly.
Will be interesting to watch how this one develops and see if the amendment leads to RCBO swapouts on the multitude of PV installations already in. I think its just going to end up costing consumers more, same as AFDDs and surge protection.
Yes, I read the BEAMA document.
There's a bit about inverters sharing RCCBs on split load boards - that makes sense. I don't think it's a good idea, and if it's a split load board it's probably a type AC RCD anyway.
Then it goes on to having voltage present on load terminals of a switched off RCD. As you say, the G98/99 relays in the inverter would switch off at that point but even if they didn't, there's no current and the RCD is energised by a differential current. I think it would be more of an issue of the RCD remaining powered by the mains if it was installed with the load side as the consumer, rather than the inverter, that's pretty difficult to achieve given the bussbar location. I guess this is my brain being too small in trying to see how that could cause a problem.
I know RCDs can become faulty, but that's why we test them (or in my case, I don't much because we don't tend to use them!).
Next is a section about extinguishing arcs, which is normally more of a DC fault issue than an AC fault issue on account of the zero crossing point. Again, I'm struggling to imagine the issue there because the current is flowing backwards and forwards multiple times a second and the arc doesn't care which way the energy flow is.
I just think this is all a fuss about nothing TBH.
As for AFDDs, I think I like the idea but I'd be putting them on electric shower circuits first! Quite a lot of inverters have built in arc fault protection on the DC side which again is a nice idea. However I have sat watching an MC4 connector quite happily arc away melting itself to bits on a system with a SolarEdge inverter with AFCI enabled, so I'm not really convinced about their effectiveness. I'm also not sure of the difference between the cheap US spec ones and the apparently fancier Euro spec ones.
@@tceng9449 the european mainland type of RCBOs and regular RCDs don't have a designated input or output side and it's usualy assigned in the way it's more convinient to connect to 😅
@@circuitbreaker1434 Yes, but I know for a fact that several popular brands of devices sold in Europe are 'directional' in that the device can be damaged by long holding of the trip button. As I said in my long winded post to the video, no RCD appears to latch earth faults so if you switch the neutral too it's a non-issue bar the test circuit problem. Solution: find something else to do all day other than holding the button!
@@retrozmachine1189 there are some older designs wich don't disable the test button when tripped with this it would be possible to burn out the test resistor if the button is held down too long
@@circuitbreaker1434 Yes, as I said in the comment you are replying to...
Sorry for the long long post but I don't think it's an issue that can be covered with a couple of sentences.
Aussie here, so I'm looking in from the outside. The issue is confusing to me, and even though I am fairly new to RCDs since they weren't really a thing in the 80s, there seems to be a lot of misinformation going around. I've even seen one larger youtube channel state that if you wire supply and load on the wrong side the device simply won't function at all. The issue here for 'directional' RCDs has always revolved around the test circuit and is well known. If the test button is held in for an extended period with supply on the 'wrong' side the electronics in an electronic RCD will leave the trip coil in the high current state, ie for as long as the button is pushed, potentially burning out the test load resistor or the trip coil or for that matter the semiconductor that causes the high current draw.
From what I can see no electronic RCD latches the earth leakage state, ie as soon as the imbalance goes away the circuit releases the trip coil current back to idle. I've tried my own test with a few brand name RCBOs I have here that aren't 'bi-directional' but are neutral switching. Due to the circuit construction there is always some current in the trip coil as it forms part of the power supply for the circuit. Not one of them has fired the trip indefinitely. Yes, if supply is on the wrong side you can get a few hits of the trip until the fault state is cleared but this is not going to kill the device even in the long term.
All that needs to be done to verify if there is an RCD that latches the fault state is to put supply on the 'wrong' side and cause an earth fault on the other side. If it trips and doesn't make a never-ceasing brrrr as the trip is repeatedly fired for ever, all good. Does anyone know of an electronic RCD that does latch, ie it only clears when the circuit has powered removed making supply / load connections critical. All that needs to be done to verify if there is an RCD that latches the fault state is to put supply on the 'wrong' side and cause an earth fault on the other side. If it trips and doesn't make a never-ceasing brrrr as the trip is repeatedly fired, all good.
The only other circumstance I can see an electronic RCD failing in such a way that the operation under normal earth fault ceases is for a device that does not switch the neutral and there is a N/E fault. When the device trips the N/E fault could cause sufficient continued imbalance in the CT to leave the trip coil etc in the high current state leading to a device failure if the supply is on the 'wrong' side, ie the circuitry continues to be powered.
I can see that magnetic type RCDs that do not switch the neutral would have a similar issue since the trip coil is operating straight off the CT current but don't know if it will cook the coil or not given the current flows involved. The test resistor burning out / coil failure on long hold in of the test button would be an issue for non-neutral switching magnetics of course.
The whole issue of arc suppression failure is complete rubbish. It's AC people. Current flow reverses as part of it. The arc breaking nature of the device is already capable of suppressing arcs with current flowing in either direction, be it MCB or RCD or RCBO.
I've made similar comments with other youtube channels. In my state in Australia, Queensland, the last step of testing by an electrician upon installation is to press the trip button, ensure device trips, observe behaviour for anything inappropriate, reset and repeat. Any abnormal behaviour is investigated.
yes, the zero crossing point makes switching AC so much easier on switchgear than DC!
I do wonder if we put too much emphasis on testing methods for RCDs, the "does the test button work" approach is fine for lots of things but apparently not RCDs. Thanks for the detailed comment :)
having been retired for 6 years , 50ms seems to be a long time tripping , on the reverse connection 19 ms is what I remember as being acceptable at 5x the tripping current as is 27ms at x 1, From your conclusion I think this latest regs amendment is another con to make you use the new type of RCD for use with sola panels ?
His instrument is set to 1x IΔn so tripping times could be up to 300ms for AC type RCDs and meet the standard. For 5xΔn it would need to below 40ms
@@mra813 yup, thanks :)
So you didn't test for the actual problem as described?
We all know the sense coil does not care about direction so your random test shows normal physics.
The actual problem is the RCD or RCBO circuit & solenoid remaining energised even when tripped on a badly designed RCD, when it is under a reversed load. Over time the now-tripped RCD circuit detection circuit burns out, turning the RCD back into an MCD. Good RCDs trip and disconnect the load and the detection/test/solenoid circuit (3 switched lines), no matter which side is connected. Bad ones (only 2 switched lines) disconnect the load only.
You can, of course, test an RCD to check for a latent or previous burnt-in failure simply by pressing the button that nobody uses at 6 months in the same way nobody tests at 3 months, or indeed following an actual fault trip.
Place supply on the wrong side of a neutral switching device, induce earth fault on the other side, observe for repeated striking of the trip (electronic type), if it doesn't sit there going brrrr it doesn't latch the fault state and does not have a problem. None latch as far as I can see. Try it with a non-neutral switching type and pass trippable levels of current through the neutral and see if it misbehaves. I think the latter is a valid concern, the former is not.
Trip button issue is valid concern, so don't hold it in excessively - duh.
@@retrozmachine1189great comment retro.. unfortunately mate 99.9% of people won't have a scooby doo what your talking about
@@retrozmachine1189 What we need to know is which devices actually have this problem, so we can avoid them or identify them in order to get them changed. Let's get a list of poorly designed devices. How often is this really going to be a real world issue and where are the stats that the changes are based on ? It's just an insurance policy for the bad manufacturers.
@@Mainly_ElectricalThat comment is in the territory of engineering so therefore it’s unnecessary for the “99.9% of people” to have anything to do with it (assuming they are in the electrical trade).
@@toms7431That is a prime job for the IET / IEEE to do. Unless you can look at the kite marks and hope to only use them when verified? Maybe only purchase from trusted manufacturers who sell in the UK and have the burden of being held accountable.
Just another way for the electrical industry to rinse the customer and publishing bullshit
100% sales people