@@NETBotic He made a video a while ago explaining what happened. In short, he got married, worked his back off to get her to UK, failed. Trying again...
@@NETBotic he's alive and well, he just doesn't have time to make videos. Last time I saw him was when he was giving a mercury arc rectifier to The Channel Of Random Crap, th-cam.com/video/VRVKiIkuKko/w-d-xo.html
Great video. The 'arc chutes' you mention are unlikely to be intended to help break the arc because once the contacts have rotated to that point the arc will have broken. I believe they are there to release the burned gas, and cool it down as mentioned in the video. Real arc chutes are metal, and they work differently. The arc connects to them breaking the arc into many small sections of arc in series. Each initiation and termination of the arc has a fixed voltage drop, which all add up. This exceeds the arc voltage and kills the arc almost instantly.
Never ever assume that the way the wires go in at the top matches the way they come out at the bottom. I have an isolator that swaps the conductors and only because I measure everything after connecting stuff together did I catch that. A gander at the datasheet would have helped but how complicated can a 4 terminal DC switch be ? I wonder how many people blew up their inverter after making that mistake. 😬 [edit] Lol, just got to the part where you also noticed the conductors swap over.
If I read the pin numbers on the label it would have been obvious. It would have been nice if they didn't show the switches in the schematic on the label going straight across!
Many DC breakers are also polarised as the internal arc suppression is on one side only. Trap for young players. Many were installed backward in our distribution area and a large program was initiated to check for compliance after some blow ups when operating.
@@TobyRobb DC breakers are typicaly polarised because they use a magnetic field (can come from a coil or magnets) to push/blow the arc into the chute/chamber and reversing the current would push the arc the wrong way.
Hey Dave, try some "high" DC current with a big inductor in series, you might get some interesting arcing. Maybe long exposure (photo or even higher shooter angle, lower frame rate). I'm guessing about those 4 to 10A with a gapped inductor in series with the switch, and then open it, the faster the better, maybe. You don't need high voltage, or energy fwiw, the magnetic field can store it for you. I just want to see it arcing!! If it works and you want to get the geometry shining, maybe an acrylic cover can gide the arcs to the channels.
Not even a microwave oven transformer? They have all the stack from one side, might act as gapped, or are easier make gapped.. At work I have a spare 2kA inductor, capable of 100kV kick back, but I guess it isn't standard office furniture. We do live off making 40kA arcs here 😅.
The Hager SB432PV looks pretty much identical to this disconnector. The Hager is probably just a Benedikt with a different label. I have 3 of the SB432PVs in use and haven't had any problems with them yet. My neighbour bought his DC load-break switches from Amazon (China brand) and already had one that melted.
This is such an awesome demonstration, never thought of the design of these DC Isolators, super interesting. Love your videos as always EEV, keep it up!
those switch contacts could be beryllium copper. Ex rated tools are usually made from it, its more wear resistant, wont oxidize as quick and has better properties than plain old copper.
Moving parts probably are, from the colour, but the fixed contacts are plain tinned copper, as the wiping ensures any oxide is removed. they do grow whiskers though with time, unless the tinm has at least 5% lead in it as alloying metal, which inhibits this.
@@SusanPearce_H You only need a small amount, well under 1% to get the improved properties. Copper is toxic anyway as a pure metal, so not much added risk. 0.1% in a batch of scrap copper will make that entire melt non malleable, and require the furnace to be relined to remove it.
11:24 seeing similar "staining" on 400v 63 amp 3 phase rcd's if the terminals have been a bit toasty. Especially on the neutral if it has been loaded a bit much and the terminals aren't tight enough.
I assume that a big part of the operation is in the shaft assembly. It looks like it's designed to snap the contacts around very quickly. Also would have been interesting to see the contacts around the burnt/corroded switch.
That plastic lines are almost certainly just an optimised mechanical support and ventilation. I'm sure you can muscle up some electro-mechanical switching design engineer as a guest to comment on this and promote the company's products.
I think you missed the fact that the black disk is actually hollow. The plasma doesn't need to sneak around any gaps. Just as the wiper is near the contacts, plasma still can find its way to the chutes through empty space in that disk.
Nice to see something that is built of modular components that are not glued together and come apart easily. Didn't know something like that still existed! :)
Those chutes are there as a vent to keep the switch from exploding from gas pressure. . There is an experiment which would be quite interesting to see. I have tested DC breakers with not much more than 100V using a power supply and load resistor. Pretty interesting that on a double breaker the arc doesn't stop till the load reaches about 35V and there is very little difference in that ending voltage between single and double breaker as well as the arc time which is about 400 microseconds. My most accessed reference book is GASEOUS CONDUCTORS by James Dillon Cobine. Usually after the fact to explain what I've seen.
Did a project on cryogenic coolers for the dc link between mainland Japan and the Island of Hokkaido. They have to be able to absorb the inductive current of thousands of amps in miles of dc cable. They used a lot of high energy absorbers that they wanted to get close to superconductivity to absorb the current so it could be slowly dissipated. It was a fun project. The issue is the inductance of the cable from the pv to your inverter carrying 10 amps or more. You can't just turn it off as cable inductance will generate an arc.
Designed such that you use a single plastic part for the switches, with only the actual conductors being different. Single unit for forming the base and the top of the next switch. Then a top and bottom, with the terminal ends providing both arc covers and vents, plus the terminals. The heavy operation only from the spring, the actual contacts being relatively easy moving, though the spring does provide a rapid engagement when closing, to aid in reducing arcing, and same for opening. Looking at some online sources, 2 full load operations are guaranteed, and at least 4000 mechanical operations. Full load is breaking the rated short circuit current, for most breakers around 2kA, the regular overcurrent trip is included in the mechanical life. For the disconnector they are not rated to break current, or for operation under load, but only for service use. Damage in the switch is from moisture, probably some that got in the case when assembled, or that wicked through the seals from rain hitting the hot box, and being sucked through the glands and case seals, because nobody will actually use proper IP68 rated glands, with a silicone rubber seal on the gland side, because of the cost. Instead use the plain IP3X cheap glands instead, that do not provide any water protection, unless you use PVC weld to seal the gland into the plastic case. They might even have tightened the nut full,y but often you will find them either loose, or they are so badly made they split with time.
5:15 the Numbers at the Terminals are there for a reason 1/3/5/7 are line in and 2/4/6/8 are line out or throw. The current enters through terminal 1 and exits through terminal 2 and so on Greeting from a sparky out of Germany
Adding switches in series sure feels like an inappropriate solution for scaling towards higher switching voltages. Since mechanical separation tends to be fairly slow and have fairly large tolerances. Ie one contact will take the bulk of the burden. Paralleling contacts has the same issue. High voltage, and especially combined with a decent bit of current makes switching wonderfully non trivial.
The designation DC21B on the side is the standard rating it’s tested to. It will specify the current to be switched as well as the number of cycles. It’s common for switches tested toIEC standards to feature such a designation.
The Fronius snap inverters use a benedict isolator switch/mating block, the base plate contains the isolator with a set of pins, ac and dc the inverter just slots in. Good kit made in Austria. Our cable fault location kit is Baur also from Austria also good stuff.
Is it possible that the rusty/discolored tabs were due to the plastic "sweating" due to high temps caused by the current flow. See if the wetness smells like diesel fuel.
My guess would be that this was a roof mounted isolator (a very very stupid standard that ONLY Australia enforced), and it had water ingress along one of the cables, high contact resistance, and enough current to cause it to heat up. These things cause many roof fires.
@@EEVblog Had a little play with the Enphase gear; More of an industrial sparky here, those solar installers are like gym junkies! There's a lot of manpower that goes into them, I'm mostly hired for my knowledge to find faults fast.
Hi Dave, you need much less voltage than you think to get a decent DC arc, switching slowly is the key here. 100V DC, if switched slowly, with a load that draws a few amps can already create arcs of a few centimeters. But I doubt you would see anything usefull when those "modules" are open on one side.
Try connecting a DC welder for getting a sustained current arcing because that HV tester with its internal tiny current limit is acting as a gap capacitor charger, you can even hear the crackling of the high speed sparks
For shutting down a PV inverter, you are supposed to isolate the AC side first, so that you can then break the DC without it being under load. Then the reverse for startup. This is usually signed as such as well.
@@shaunclarke94 Yes, of course, but that's not the point of the video. In fact the entire point of the video is about what happens when you do break a HV DC load.
Some high voltage switches use alloys with cadmium for the contacts, you don't wanna lick that
2 หลายเดือนก่อน +1
Interesting, i expected more metal parts. I imagined there was a discharge chamber with many metal fins to spread the spark like in DC breakers ... the more you know. ^^
Interesting! I have never seen plastic arc chutes before. I would have expected the heat of an arc to carbonize the plastic and maybe form a conductive path across the switch
These are not rated for on load disconnection, only to isolate the panels for service, so switching should not involve current flow, only the arc from the capacitance and inductance in the wire. Plastic arc chutes are common though, most breakers use a combination of thermoplastic and metal arc chutes these days, at least using plastic parts for the channels and any arc diverters in the switch.
If someone wonders if they can use a regular fuse or one of those din mounted automatic AC circuit breakers as a DC fuse/ isolator because it's cheaper and "works" the answer is absolutely not. DC arc is nasty and much harder to extinguish compared to AC ok, exception being low amps and voltages, then use whatever
Most inverters should offer a software shutdown that should be used for maintenance before switching the actual DC switch. Helps with contact wear-out.
solar DC isolation switches are not recommended here as it increases the risk of installation failure. Most inverters have this built in so you dont need extra failure point.
A question to you please, how would you prevent Switching bouncing in a high voltage DC? For low voltage DC it is easy but for high DC voltage (120V and up ) it wouldn't be easy .. taking in consideration that you need to have a ON voltage instantly .. i.e. you should be able to take transient characteristic sof a DUT using that DC voltage.
Why does it need to do fancy denouncing when it makes contact? At nominal current and typical arc voltage of ~50V it's not that much power, and only for a brief moment. Besides that, it shouldn't be hard to design the contact pressure delivery setup such that the ringing from the impact is dissipated somewhere else other than the mating surface.
@@namibjDerEchte If you are testing a device or a board, and the switching ON is captured by an Oscilloscope, i.e. the transient. That is the situation I am talking about. You need to have a pure, quick high voltage ON to the circuit. I thought about Solid state relay, but they are not without delay.
It is real problem with arcing. I wonder why not to short PV input with a IGBT first, and then switch it off, and then release IGBT short? Instead of dealing with full PV voltage, shorted PV will have voltage equal to Vce which is about 3V
Because electrical standard committees are allergic to solid state _disconnects._ Note how you'd have to use two switches (but one need not be capable of breaking operating voltage/current, only carrying fault current while mated) with the IGBT in series with the weaker one, and sequence "break main; wait for arc to extinguish; turn off IGBT; when current has subsided, break auxiliary". In practice for DC situations (with >=400V) you would want to use SiC JFETs that are always in the current path. The reason is that you can trip them extremely quickly before a large fault current has built up (like, literally in single-digit microseconds), which also conveniently allows selective breaker tripping to isolate the fault without interrupting the rest of the local DC grid (e.g. panels with power optimizers feeding a DC bus to which inverter, battery, and straight DC loads (inverter HVAC units and the PSUs for modern electronics (computer, laptop, TV) could run off of a couple hundred volt DC with minor or even no modifications) are connected).
With all due respect Dave, it looks like some artist created some mold designs that look the part and eventually one such design passed some specified tests. For one energy for sustaining the plasma can only come from the external circuit, not the rotating contact which has all the elaborate designs for plasma control in the off position. Maybe I'm completely wrong but I have not seen enough to be convinced otherwise.
Huh.... I wouldn't be able to tell, only have a Civil Engineering and Ocean Engineering background. At least in that you can pretty quickly tell where architecture meets the actual Engineering design principles... I am unsure how one would be able to analyze a design for electrical engineering. Since a huge portion of designs go unseen in our day to day lives I don't know how one would tell good design from bs.
The mode of operation is to lengthen and cool the arc. Used also in substations for AC they used to use (SF6) gas to blow the arc away to lengthen it and also cool it at the same time. Not sure what they use now as it is not good for climate change.
@@FireballXL55 ive never heard of sulfur hexafloride being used to "blow away" arcs, I've only ever heard of it being used as an insulating medium inside sealed enclosures.
@@EEVblog The new update to AS5033 has the rules around whether a DC panel string requires over current protection, If the cable sizing is below the max fault current including any chance of battery backfeed etc. and some other conditions. Luckily most solar is low-ish current but the voltage gets up there and that's where the problem's have been. With the more modern grid connected inverters the voltages get well over 500 volts making them extremely dangerous. Many fires have started as a result.
@@TobyRobb I've never seen a typical home solar install with a DC string side breaker. Why would you use undersized cable? Wouldn't that be against code?
@@EEVblog Yes your correct, but it is absolutely possible to have any size conductor provided it is protected by an appropriately sized breaker . Only in solar do we see cables with ratings far higher than load current due to the need to ensure low voltage drop. Additionally, solar panels generally don't have high short circuit current . Normally fault currents usually always exceed cable ratings hence the need for a breaker.
The issue is with DC and the melting temperature of silver. Plasma itself is low resistance. Highest heat is generated at the transition point between metal and plasma. Copper is also a problem except it conducts heat away so easily. Tungsten variants are used because high melting temperatures.
why are there four "women" who all joined 4 hours ago at the same time sharing the same profile pictures and commenting generically, scammers i bet , watch out for the telegram messages everyone 😂
I don't like mini-isolators just in general. Operation isn't spring-loaded and the ARC isn't pulled off the contacts very quickly, so every operation damages the unit a bit more. Eventually you get enough resistance to cause a fire. But doubly-so up on roofs. Nor should these things ever be put on top of a roof where they can't be easily checked.
That big spring is because these are spring loaded. You have to load the spring up with considerable torque/energy before it releases it's energy and actuates the actual mechanism inside.
@@gnif Ahh, I missed the spring. Well, that's an improvement. But I still avoid these things like the plague. I want something that will actually trip-off thermally so I pretty much universally use real DC breakers instead of disconnects. Not because of the current trip (which will never happen), but because nearly all failure cases for a breaker result in a trip-off. -Matt
@@junkerzn7312 How will a breaker in a DC solar string trip if the switch arcs and allows the continuous (no fault) current to flow into the otherwise working inverter?
@@junkerzn7312 What do you mean by, "nearly all failures"? How many failures are you experiencing? If your answer is more than 1, something in your setup or design is wrong/bad.
@@EEVblog I did say that it doesn't trip normally. The purpose of using a breaker is simply as a manual disconnect. This is why most solar combiner boxes have master breakers. The solar panels can't put enough current into the breaker to trip it normally. But its a better disconnect than a solar disconnect. Under abnormal conditions the internals of the breaker are under tension. A mechanical failure internally WILL trip the breaker in most situations, yes. And a high-resistance failure mode that overheats the breaker internally will also tend to trip it, either from the bimetalic element or from an internal mechanical failure. It doesn't take much inside a breaker to trip it. If any of those plastic pieces distort from the heat, it will almost certainly trip.
Where's PhotonicInduction when you need a huge-ass distribution transformer and mercury arc rectifier to test this thing with high DC current? Ha.
Right? I've read a lot of rumors about what happened to that guy. I'd love to know from a legit source.
Hes still about I believe I had a dig within the last year. @@NETBotic
@@NETBotic He made a video a while ago explaining what happened. In short, he got married, worked his back off to get her to UK, failed. Trying again...
@@zlotvorxIf he'd regularly upload to TH-cam perhaps he wouldn't have this problem... 😅
@@NETBotic he's alive and well, he just doesn't have time to make videos. Last time I saw him was when he was giving a mercury arc rectifier to The Channel Of Random Crap, th-cam.com/video/VRVKiIkuKko/w-d-xo.html
Great video. The 'arc chutes' you mention are unlikely to be intended to help break the arc because once the contacts have rotated to that point the arc will have broken. I believe they are there to release the burned gas, and cool it down as mentioned in the video.
Real arc chutes are metal, and they work differently. The arc connects to them breaking the arc into many small sections of arc in series. Each initiation and termination of the arc has a fixed voltage drop, which all add up. This exceeds the arc voltage and kills the arc almost instantly.
Never ever assume that the way the wires go in at the top matches the way they come out at the bottom. I have an isolator that swaps the conductors and only because I measure everything after connecting stuff together did I catch that. A gander at the datasheet would have helped but how complicated can a 4 terminal DC switch be ? I wonder how many people blew up their inverter after making that mistake. 😬
[edit] Lol, just got to the part where you also noticed the conductors swap over.
If I read the pin numbers on the label it would have been obvious. It would have been nice if they didn't show the switches in the schematic on the label going straight across!
@@EEVblog Yes, exactly. That's what fooled me into not examining the schematic closer.
Many DC breakers are also polarised as the internal arc suppression is on one side only. Trap for young players. Many were installed backward in our distribution area and a large program was initiated to check for compliance after some blow ups when operating.
@@TobyRobb DC breakers are typicaly polarised because they use a magnetic field (can come from a coil or magnets) to push/blow the arc into the chute/chamber and reversing the current would push the arc the wrong way.
Yes have met those, they went in and crossed over inside.
You could poke that "residue" with the HV insulation tester, see if its conductive at 5kV. If not its probably not metal.
Hey Dave, try some "high" DC current with a big inductor in series, you might get some interesting arcing. Maybe long exposure (photo or even higher shooter angle, lower frame rate).
I'm guessing about those 4 to 10A with a gapped inductor in series with the switch, and then open it, the faster the better, maybe. You don't need high voltage, or energy fwiw, the magnetic field can store it for you. I just want to see it arcing!!
If it works and you want to get the geometry shining, maybe an acrylic cover can gide the arcs to the channels.
Might be possible. Not sure a big arse inductor anywhere though...
You'd think a charged 24V truck battery would do a decent job?
Not even a microwave oven transformer? They have all the stack from one side, might act as gapped, or are easier make gapped..
At work I have a spare 2kA inductor, capable of 100kV kick back, but I guess it isn't standard office furniture. We do live off making 40kA arcs here 😅.
The Hager SB432PV looks pretty much identical to this disconnector.
The Hager is probably just a Benedikt with a different label.
I have 3 of the SB432PVs in use and haven't had any problems with them yet. My neighbour bought his DC load-break switches from Amazon (China brand) and already had one that melted.
I have seen DC switch designs with magnets to help push/blow the arc away from the switch contacts.
I love it when Dave turns the lights down
So does Mrs EEVblog
This is such an awesome demonstration, never thought of the design of these DC Isolators, super interesting. Love your videos as always EEV, keep it up!
those switch contacts could be beryllium copper. Ex rated tools are usually made from it, its more wear resistant, wont oxidize as quick and has better properties than plain old copper.
Isn't Beryllium highly toxic?
Perhaps that was a part of the "ozone" smell?
Moving parts probably are, from the colour, but the fixed contacts are plain tinned copper, as the wiping ensures any oxide is removed. they do grow whiskers though with time, unless the tinm has at least 5% lead in it as alloying metal, which inhibits this.
@@SusanPearce_H You only need a small amount, well under 1% to get the improved properties. Copper is toxic anyway as a pure metal, so not much added risk. 0.1% in a batch of scrap copper will make that entire melt non malleable, and require the furnace to be relined to remove it.
@@SeanBZA Thank you for a comprehensive reply.
It's more fun when you let her say it.
When you get into hundreds of gigaohms, doesn't it more sense to switch to Siemens? (Yes, she said that too.)
The Arc of the Conduit?
L'Arc de Triomphe!
11:24 seeing similar "staining" on 400v 63 amp 3 phase rcd's if the terminals have been a bit toasty.
Especially on the neutral if it has been loaded a bit much and the terminals aren't tight enough.
I assume that a big part of the operation is in the shaft assembly. It looks like it's designed to snap the contacts around very quickly.
Also would have been interesting to see the contacts around the burnt/corroded switch.
Amusing that such a simple device can be so interesting in design.
That plastic lines are almost certainly just an optimised mechanical support and ventilation.
I'm sure you can muscle up some electro-mechanical switching design engineer as a guest to comment on this and promote the company's products.
I think you missed the fact that the black disk is actually hollow. The plasma doesn't need to sneak around any gaps. Just as the wiper is near the contacts, plasma still can find its way to the chutes through empty space in that disk.
I'm guessing that big spring goes along with some stiff detents, so the switch snaps rapidly from on to off and vice versa.
yes. 3:19 the spring is like a "servo saver" it lets the thing turning build up energy and then snap once the contacts start sliding.
Nice to see something that is built of modular components that are not glued together and come apart easily. Didn't know something like that still existed! :)
Made in Austria, no BS
Like OceanGate Titan?
Those chutes are there as a vent to keep the switch from exploding from gas pressure. . There is an experiment which would be quite interesting to see. I have tested DC breakers with not much more than 100V using a power supply and load resistor. Pretty interesting that on a double breaker the arc doesn't stop till the load reaches about 35V and there is very little difference in that ending voltage between single and double breaker as well as the arc time which is about 400 microseconds. My most accessed reference book is GASEOUS CONDUCTORS by James Dillon Cobine. Usually after the fact to explain what I've seen.
Did a project on cryogenic coolers for the dc link between mainland Japan and the Island of Hokkaido. They have to be able to absorb the inductive current of thousands of amps in miles of dc cable. They used a lot of high energy absorbers that they wanted to get close to superconductivity to absorb the current so it could be slowly dissipated. It was a fun project. The issue is the inductance of the cable from the pv to your inverter carrying 10 amps or more. You can't just turn it off as cable inductance will generate an arc.
Designed such that you use a single plastic part for the switches, with only the actual conductors being different. Single unit for forming the base and the top of the next switch. Then a top and bottom, with the terminal ends providing both arc covers and vents, plus the terminals. The heavy operation only from the spring, the actual contacts being relatively easy moving, though the spring does provide a rapid engagement when closing, to aid in reducing arcing, and same for opening.
Looking at some online sources, 2 full load operations are guaranteed, and at least 4000 mechanical operations. Full load is breaking the rated short circuit current, for most breakers around 2kA, the regular overcurrent trip is included in the mechanical life. For the disconnector they are not rated to break current, or for operation under load, but only for service use.
Damage in the switch is from moisture, probably some that got in the case when assembled, or that wicked through the seals from rain hitting the hot box, and being sucked through the glands and case seals, because nobody will actually use proper IP68 rated glands, with a silicone rubber seal on the gland side, because of the cost. Instead use the plain IP3X cheap glands instead, that do not provide any water protection, unless you use PVC weld to seal the gland into the plastic case. They might even have tightened the nut full,y but often you will find them either loose, or they are so badly made they split with time.
5:15 the Numbers at the Terminals are there for a reason 1/3/5/7 are line in and 2/4/6/8 are line out or throw. The current enters through terminal 1 and exits through terminal 2 and so on
Greeting from a sparky out of Germany
Adding switches in series sure feels like an inappropriate solution for scaling towards higher switching voltages. Since mechanical separation tends to be fairly slow and have fairly large tolerances. Ie one contact will take the bulk of the burden.
Paralleling contacts has the same issue.
High voltage, and especially combined with a decent bit of current makes switching wonderfully non trivial.
The designation DC21B on the side is the standard rating it’s tested to. It will specify the current to be switched as well as the number of cycles. It’s common for switches tested toIEC standards to feature such a designation.
They could have used a solid state device that opened the supply slightly before the mechanical switch
Could have used some magnets to drag the arc away from the contacts. in the same way they do in some high power DC relays.
The Fronius snap inverters use a benedict isolator switch/mating block, the base plate contains the isolator with a set of pins, ac and dc the inverter just slots in. Good kit made in Austria. Our cable fault location kit is Baur also from Austria also good stuff.
Is it possible that the rusty/discolored tabs were due to the plastic "sweating" due to high temps caused by the current flow. See if the wetness smells like diesel fuel.
My guess would be that this was a roof mounted isolator (a very very stupid standard that ONLY Australia enforced), and it had water ingress along one of the cables, high contact resistance, and enough current to cause it to heat up. These things cause many roof fires.
@@gnif Yes, possible, this was my Cobalt Black rooftop DC isolator. Glad my new install doesn't have one on the roof.
Cool demo Dave! I'm glad we've killed the rooftop isolator in the Aussie standards as they caused so many problems for us electricians over the years!
Yeah, my new install doesn't have it. The AC one for the Enphase is still mandatory though so the installers told me.
@@EEVblog Had a little play with the Enphase gear; More of an industrial sparky here, those solar installers are like gym junkies! There's a lot of manpower that goes into them, I'm mostly hired for my knowledge to find faults fast.
Hi Dave, you need much less voltage than you think to get a decent DC arc, switching slowly is the key here.
100V DC, if switched slowly, with a load that draws a few amps can already create arcs of a few centimeters.
But I doubt you would see anything usefull when those "modules" are open on one side.
Try connecting a DC welder for getting a sustained current arcing because that HV tester with its internal tiny current limit is acting as a gap capacitor charger, you can even hear the crackling of the high speed sparks
I learned something new!
For shutting down a PV inverter, you are supposed to isolate the AC side first, so that you can then break the DC without it being under load.
Then the reverse for startup.
This is usually signed as such as well.
Yes, but the DC isolator is capable of it, that's what it's designed for.
@@EEVblog correct, but you'll put unnecessary wear on it. Less wear to break the AC side first if it's a routine shutdown and not an emergency.
@@shaunclarke94 Yes, of course, but that's not the point of the video. In fact the entire point of the video is about what happens when you do break a HV DC load.
10 car batteries should get enough current and voltage to test.
Some high voltage switches use alloys with cadmium for the contacts, you don't wanna lick that
Interesting, i expected more metal parts. I imagined there was a discharge chamber with many metal fins to spread the spark like in DC breakers ... the more you know. ^^
wohoo plasma chutes and ladders!
Interesting! I have never seen plastic arc chutes before. I would have expected the heat of an arc to carbonize the plastic and maybe form a conductive path across the switch
These are not rated for on load disconnection, only to isolate the panels for service, so switching should not involve current flow, only the arc from the capacitance and inductance in the wire. Plastic arc chutes are common though, most breakers use a combination of thermoplastic and metal arc chutes these days, at least using plastic parts for the channels and any arc diverters in the switch.
I wonder if that burning is just from the plastic slowly melting or possibly some heated dielectric grease that leaked in. Looks kinda sticky.
I have played with 130 voc solar array at about 9 amps and was surprised by the size of arks I could make at that lower voltage point
A maze for plasma. Nice!
Yep, and it gets so tired doing the maze that it just loses all its energy and gives up.
@@EEVblog There's also wild magnetism forces at work, the current rips it's own plasma rope apart if forced into a funny shape.
@@leggysoft Yes, at the physics level that's what's happening.
If someone wonders if they can use a regular fuse or one of those din mounted automatic AC circuit breakers as a DC fuse/ isolator because it's cheaper and "works" the answer is absolutely not. DC arc is nasty and much harder to extinguish compared to AC
ok, exception being low amps and voltages, then use whatever
Some fuses come with suitable DC ratings; most energy-limiting generic MCCBs also come with a (though less suitable) DC rating.
Most bigger ceramic fuses can handle it, 10x38mm "midget fuses" and bigger. Of course read the datasheet!
Most inverters should offer a software shutdown that should be used for maintenance before switching the actual DC switch. Helps with contact wear-out.
solar DC isolation switches are not recommended here as it increases the risk of installation failure. Most inverters have this built in so you dont need extra failure point.
"...that's where the shaft from the big knob goes is" said the actress to the bishop.... 😅
A question to you please, how would you prevent Switching bouncing in a high voltage DC? For low voltage DC it is easy but for high DC voltage (120V and up ) it wouldn't be easy .. taking in consideration that you need to have a ON voltage instantly .. i.e. you should be able to take transient characteristic sof a DUT using that DC voltage.
Why does it need to do fancy denouncing when it makes contact?
At nominal current and typical arc voltage of ~50V it's not that much power, and only for a brief moment.
Besides that, it shouldn't be hard to design the contact pressure delivery setup such that the ringing from the impact is dissipated somewhere else other than the mating surface.
@@namibjDerEchte
If you are testing a device or a board, and the switching ON is captured by an Oscilloscope, i.e. the transient. That is the situation I am talking about.
You need to have a pure, quick high voltage ON to the circuit.
I thought about Solid state relay, but they are not without delay.
It is real problem with arcing. I wonder why not to short PV input with a IGBT first, and then switch it off, and then release IGBT short? Instead of dealing with full PV voltage, shorted PV will have voltage equal to Vce which is about 3V
Because electrical standard committees are allergic to solid state _disconnects._
Note how you'd have to use two switches (but one need not be capable of breaking operating voltage/current, only carrying fault current while mated) with the IGBT in series with the weaker one, and sequence "break main; wait for arc to extinguish; turn off IGBT; when current has subsided, break auxiliary".
In practice for DC situations (with >=400V) you would want to use SiC JFETs that are always in the current path.
The reason is that you can trip them extremely quickly before a large fault current has built up (like, literally in single-digit microseconds), which also conveniently allows selective breaker tripping to isolate the fault without interrupting the rest of the local DC grid (e.g. panels with power optimizers feeding a DC bus to which inverter, battery, and straight DC loads (inverter HVAC units and the PSUs for modern electronics (computer, laptop, TV) could run off of a couple hundred volt DC with minor or even no modifications) are connected).
With all due respect Dave, it looks like some artist created some mold designs that look the part and eventually one such design passed some specified tests. For one energy for sustaining the plasma can only come from the external circuit, not the rotating contact which has all the elaborate designs for plasma control in the off position. Maybe I'm completely wrong but I have not seen enough to be convinced otherwise.
Huh.... I wouldn't be able to tell, only have a Civil Engineering and Ocean Engineering background. At least in that you can pretty quickly tell where architecture meets the actual Engineering design principles... I am unsure how one would be able to analyze a design for electrical engineering. Since a huge portion of designs go unseen in our day to day lives I don't know how one would tell good design from bs.
This is a well studied field of high voltage switch design. Choose not to believe it if you wish.
The mode of operation is to lengthen and cool the arc.
Used also in substations for AC they used to use (SF6) gas to blow the arc away to lengthen it and also cool it at the same time.
Not sure what they use now as it is not good for climate change.
@@FireballXL55 Yes there are other methods like gas and copper bar chutes.
@@FireballXL55 ive never heard of sulfur hexafloride being used to "blow away" arcs, I've only ever heard of it being used as an insulating medium inside sealed enclosures.
just plain old rust. see if a magnet sticks to the contacts.
What can interrupt fault current? A breaker, that's its job! Always best to operate the breaker first then the isolator IMHO.
The solar array DC side does not have a breaker, only the AC side.
@@EEVblog The new update to AS5033 has the rules around whether a DC panel string requires over current protection, If the cable sizing is below the max fault current including any chance of battery backfeed etc. and some other conditions. Luckily most solar is low-ish current but the voltage gets up there and that's where the problem's have been. With the more modern grid connected inverters the voltages get well over 500 volts making them extremely dangerous. Many fires have started as a result.
@@TobyRobb I've never seen a typical home solar install with a DC string side breaker. Why would you use undersized cable? Wouldn't that be against code?
@@EEVblog Yes your correct, but it is absolutely possible to have any size conductor provided it is protected by an appropriately sized breaker . Only in solar do we see cables with ratings far higher than load current due to the need to ensure low voltage drop. Additionally, solar panels generally don't have high short circuit current . Normally fault currents usually always exceed cable ratings hence the need for a breaker.
Solar panels are inherently current sources. Overcurrent is not a thing.
Plain copper contact, not silver? Weaksauce! Yeah that definitely doesn't seem like a high cycle rating deal 😅
There is possibly nothing worse than silver to make a high voltage DC switch with.
@@opera5714 Really? I was always told the opposite. Is silver really only for high amperage?
The issue is with DC and the melting temperature of silver. Plasma itself is low resistance. Highest heat is generated at the transition point between metal and plasma. Copper is also a problem except it conducts heat away so easily. Tungsten variants are used because high melting temperatures.
call it a 4cake switch?
Hayche
A maze ing
If it's DC why does it have 4 current paths? Just a coincidence?
Do you mean the 4 poles? That's for flexibility in configuration. You only need two poles for a single DC solar string.
Made in Austria, NOT Australia. 😉
These plastic ribs are indeed to make thing rigid and also to make mold shape symmetrical. It has nothing to do with arc arresting.
why are there four "women" who all joined 4 hours ago at the same time sharing the same profile pictures and commenting generically, scammers i bet , watch out for the telegram messages everyone 😂
Smelling ozone is bad for you, mkay?
It'll put hair on your chest.
My ears.... wtf
lol .. unfortunately no she didn't.
DC has a tendency to arc, the switches need to be specifically designed for DC
I don't like mini-isolators just in general. Operation isn't spring-loaded and the ARC isn't pulled off the contacts very quickly, so every operation damages the unit a bit more. Eventually you get enough resistance to cause a fire. But doubly-so up on roofs. Nor should these things ever be put on top of a roof where they can't be easily checked.
That big spring is because these are spring loaded. You have to load the spring up with considerable torque/energy before it releases it's energy and actuates the actual mechanism inside.
@@gnif Ahh, I missed the spring. Well, that's an improvement. But I still avoid these things like the plague.
I want something that will actually trip-off thermally so I pretty much universally use real DC breakers instead of disconnects. Not because of the current trip (which will never happen), but because nearly all failure cases for a breaker result in a trip-off.
-Matt
@@junkerzn7312 How will a breaker in a DC solar string trip if the switch arcs and allows the continuous (no fault) current to flow into the otherwise working inverter?
@@junkerzn7312 What do you mean by, "nearly all failures"? How many failures are you experiencing? If your answer is more than 1, something in your setup or design is wrong/bad.
@@EEVblog I did say that it doesn't trip normally. The purpose of using a breaker is simply as a manual disconnect. This is why most solar combiner boxes have master breakers. The solar panels can't put enough current into the breaker to trip it normally. But its a better disconnect than a solar disconnect.
Under abnormal conditions the internals of the breaker are under tension. A mechanical failure internally WILL trip the breaker in most situations, yes. And a high-resistance failure mode that overheats the breaker internally will also tend to trip it, either from the bimetalic element or from an internal mechanical failure.
It doesn't take much inside a breaker to trip it. If any of those plastic pieces distort from the heat, it will almost certainly trip.