Testing the world's biggest LED

פורסם בתאריך 13 אפר 2020
The world's brightest LED is SO POWERFUL that just powering it on is a challenge in itself! Also, the first 1000 people to use this link will get a 2 month free trial of Skillshare Premium
Membership: skl.sh/diyperks2
Here's the LED in question: store.yujiintl.com/products/y...

תגובות: 5 911

  • FAQ Time! But first, don't forget that after your first two months of free Skillshare trial goodness it's LESS than $10 per month if you go with an annual plan, which is like a good cup of coffee per week but instead you're building on your lifelong knowledge. Enjoy! skl.sh/diyperks2 Now, question and answer time! The first think I'll say is that there's been a surprising spread of discussion about various aspects of this build in the comments. It's great to see! I actually learn a lot from reading comments that you guys leave, and there are some serious golden nuggets of info here and there which I appreciate a lot for building on my own knowledge. I will first address a few common misunderstandings though. Firstly, some folk are assuming that the LED is 90%+ efficient, which would mean that it would only output 150w as heat and the rest as light. This isn't the case unfortunately, as even some of the best LEDs are only 40% efficient MAXIMUM. This figure changes depending on the colour accuracy, with more colour accurate LEDs being less efficient as they have to emit more of the visible spectrum. Let's assume the LED is roughly 30% efficient (which is probably fairly accurate, and may even be a bit generous) that means that 450w is being emitted as light, and the other 1050w is being wasted as heat. So, I built my cooling system around the assumption of needing to dump 1000w of heat into the air, which is why I had four large radiators and made an extreme cooling block for it. LEDs last longer when they're cooler, so it's super important. I also want the final unit to run quietly, not with maximum fan speed, so this extra headroom allows me to do that. Another thing that crops up is talk of corrosion. In water loops it's very important to use similar metals, otherwise you run in to galvanic corrosion where one metal will literally steal electrons from the other, causing the other one to corrode quickly as they're 'connected through the water'. Brass and copper are similar enough, but the lead in the solder may be susceptible to corrosion. However, adding an anti-corrosive additive to the liquid should solve this in the final build. The radiators are copper, not aluminium, so no worries there. As for the PSU's failing... there have been many theories but nothing solid. They are 12v PSUs, and are rated for 83A on that lane. When in series, they give 24v at 83A... 83*24=1992w. The reason for my wanting a higher voltage is that it literally halves the amps on the voltage booster, meaning they can run easily rather than at their absolute maximum ratings. Hit me up with any more questions! Anyway, I hope that clears things up. If you still see these questions popping up please relay the answers contained here for me please :)

    • You had to Put a Lux Meter in Front.

    • Where did you get those voltage booster circuits from? @DIY Perks

    • Hi Matt. When are you following this video up with one showing it all working outside etc? Can’t your new power bank deliver the necessary power?

    • Hey you might want to talk to "styrophyro" he's vary safe and educational

    • when will we see outdoor tests?

  • These are standard in BMW's headlights when they drive behind you.

    • But *ONLY* when they are behind you

    • 😂

    • @Lean Mean Green Bean Machine true lol I'm telling it because i own a Volkswagen and it uses the Shane light

    • @Blue Blue they make me see hell

    • The CRVs on their lowest brightness blind the hell out of me 😭

  • This is filmed so smoothly and professionally I feel like I'm watching advanced children's TV just for adults

    • Feels like "How it's made" but more personal

    • I feel like I'm watching Mr maker but the producers trust me with a soldering iron

    • kinda perhaps like, advanced adult TV or something?

  • As Abraham Lincoln once said: "If I had 10 minutes to power an LED I'd spend the first 8 cooling it"

  • I think discord's light theme is much bighter than that LED

    • Lol

    • @evan ma sciao :3

    • Lol guarda chi c'è sono evanthehusky

    • @Asaultik Someone is writing this for me, because that made me blind :(

  • "Lighting and LED enthusiast" My brain: "moth man"

  • I'm more impressed by the dynamic range of your camera than by the led

    • I think he shot it in LOG

    • @swilbs no body cares.

    • THIS

    • And that is AFTER yt compression . Imagine the raw footage ! wow

  • Fun Fact: Electricity actually runs on smoke - if the smoke escapes, the electricity stops working

    • Ive never heard this before lol thats great

    • Absolutely. A semiconductor fails when the magic smoke escapes 🙂. Unfortunately this joke will be lost on people without an electronics background.

    • This is true, I've worked for a factory that puts the smoke in.

    • @Peter goes to Hollywood Sometimes even not electrical devices.

    • Every electrical device is a smoke machine if operated wrong enough!

  • I’m always amazed these COBs don’t rip themselves apart with the thermal expansion changes of substrate vs. semiconductor. A lot of material science must have gone into their R&D!

  • The only man I've ever seen wearing 2 pairs of sunglasses, unironically.

    • Markiplier wore 5 or 6 with the 100,000 lumen flashlight

    • @Noodle's Gamebox probably 4 to 5 i guess

    • @Viny Redx i wonder how many pairs he would rate this one?

    • Well... Markiplier kinda did it the same thing, but he was more masochist, he put sunglasses and put right into his eyes

    • @Prat formerly .. styrofoam indeed

  • There was actually a couple of things that went wrong with your power supply solution. And the series topology wasnt one of them, thats totally fine to do, aslong as the output of the supply (internally) isnt earth referenced or negative commoned in some way. The reason why it overheated was of a overcurrent situation. Well how can a 1000W PSU overcurrent with only 750W load?, one might ask. High Power power supplies for servers, or even consumer computers utilize several voltage rails. Yes there are some PSUs that have them commoned together, so that every rail has the same max power/amp output, however this isnt usual. There are also other voltage rails, like 5V and 3.3V. So a 1000W server PSU might have a 100W 5V rail, 50W 3.3V rail and then like 4 12V rails at 250W each. So if you only connect to one of those rails you effectivly only have a 250W 12V PSU. Now I dont know the specs of yours, so they could have been 500W rails for all I know. But that gives the same result, the rail had a overcurrent situation. The second reason it failed, apperently the PSU doesnt have working overcurrent shutdown on a per rail basis. These arent mandatory to have for each single rail and in cheap supplies often are only used for all rails combined. Third DC/DC converters arent perfect, especially the cheapo chinese ones you used. This even gets worse for a Step up topology which you used. Those units may have a effiency at that voltage difference of maybe about 80%-85%. So that would be instead of 1500W power draw from the PSUs, around a 1800W power draw. That could overheat even a single rail PSU if the 12V rail isnt that high powered. So even if these PSUs were single rail at 12V that could still mean that they had like a 250W 5V rail + a maybe 800W 12V rail as an arbitrary example. (yes different voltage rails can (and pretty much always do) sum up to a higher power rating than total).

    • @CoolKoon It doesnt matter that they are server PSUs. All points still stand. Server PSUs can be garbage quality also. Every semi-high Power PSU that converts high voltage AC to low voltage DC utilizes multiple rails. *Every* Server/Desktop PSU. Some may tie the rails together, making it work to the enduser like a single rail, but infact internally they are multirail. Lets give this PSU the benifit of the doubt and say that it is indeed capable of delivering its spec of 83A. Server PSUs almost never tie the rails together (why would they? It decreases safety in overload conditions and increases cost), meaning as an example that a single rail could provide something like 20A max. or 40A, you get my point. Besides that to put it in your words, your argument clashes with the tiny fact that you consider server PSUs to be always of high quality, while in fact this isnt always the case, as with every product on this planet. If this product had any quality whatsoever it wouldnt have let all the magic smoke let loose when an overload condition occured, no matter if it were only overloaded on one rail or whatever the case may be, a proper power supply would have turned itself off long before it started to cook itself.

    • @OP unfortunately most of your arguments clash with the tiny fact that Matt has used modular server power supplies (which he even mentions in the video). Not only are those definitely not cheap supplies (most often than not they don't have a generic form factor, they vary even within the same manufacturer) but they are made exactly for this kind of sustained, heavy load (presented usually by sustained high CPU load and an armload of HDDs). As such, these have a completely different build than what you have in your standard desktop PCs and its rails are a lot wider than the stuff found in standard desktop PSU cabling too. He did mention in the FAQ that the power supply has nominal rating of 83A for the 12-volt rail, which means it must be able to deal with the load.

    • I didn't understand a word you say..but I am gonna give it for you...you know your shit....

  • This guy is basically a safer electroboom

    • Ok boomers

    • Electroboom knows exactly how electricity will behave and blows stuff up exactly how he intends to for teh lulz

    • Liked simply for best jolteon as avatar :3

    • OwO UwU i see Sam as a jolteon~ 😏

  • This looks wonderful, loving the experiment! Regarding the PSU's, it looks like you disconnected them via the wall plug. I would guess that this caused the fans to stop before the PSU's had time to cool down, causing components to stay at high heat for a while. Good luck with the next project!

  • 11:11 gotta say, i'm impressed with the camera's dynamic range and ability to actually capture the brightness of the light, and simultaneously expose its surroundings well enough to comfortable see.

  • "Even an air cooler this size can only handle one fifth of the heat" Water cooler: is the size of a small country

  • "The heatsink is cold, that's really good" - not the most obvious reaction to have, I'd be getting temperature sensors out and double checking exactly what I had put between the LED and heatsink at that point :)

  • As someone that works with high powered LED's regularly, unfortunately I know for a fact that nothing through the video can do the insane brightness this thing probably is creating justice. Really fun and cool build though.

  • 911 what’s your emergency? My neighbor is trying to recreate the Big Bang in his attic

    • what an original way to tell everyone that you don't understand the Big Bang at all.

    • Judging from the thumb nail i first saw a cup of tea worth $1000...

    • I guess in UK it's 999

  • I was looking for LED light for my bicycle

    • Yeah that would be a nice flood light for night rides although you need to bring a lot of equipment even if your just using 1 strip to power the LED

    • Flash? is that you?

    • I think you'll need a motorcycle to bring for you all the equipments necessary for running this led😂🤦🏻‍♂️

    • yeah, that one should work out... at least if you would like to brighten up your neighbourhood while driving through the streets :D

  • Thanks for the video. It was fun to see lots of overkill parts together. Here at home we enjoy your builds. Please continue the good work.

  • This is such a lovely engineering moment. He puts tons of effort into designing the casing for heat dispersion and it is the power supply that forms the bottleneck.

  • Amazing work you've done there! I think the problem with the power supply was that in addition to what you said, you taped it to each other and the heat can no longer dissipate from the side that was tapped to and that contribute to over heat and burnt smell.

  • Your joy when crafting all the things you craft is literally miasmic by watching. Keep up that spirit and good work!

  • being a retired electronic and software engineer, i must say that your projects are very well explained and right, well done!

  • We totally need a projector with this LED light :D

    • Output of LED movie projectors for theatres is even brighter than this

  • I would love to see a light meter reading on this at full power! How many foot candles does it produce?

  • Your videos are the only ones I watch fully without skipping even a second. Your intros are also uniquely short, unlike most other videos.

  • I could come up with some uses for this, such as a light source for a lens type light source (i.e., lighthouse beacon or spotlight). But, what is its typical use?

  • We need this for an artificial sun 2.0 build :) Also, I suggest using 2 consumer 1000 watt PSUs not in series. To my knowledge, you can just connect half of the voltage booster boards to one PSU, and the rest to the other.

  • I once had to fix a tankless water heater I find using a 2500w heatgun in combination with the torch greatly improves the ease of soldering (and in my case makes it possible at all) Perhaps I could have used a oxy set but i don't have one.

  • Those power supplies might be running in a redundant configuration, so one of the power supplies is handling all the load while the other one is a backup. It is also possible that the power supply has multiple rails, and the 12/24v rail is only rated for less than the total rating.

  • That LED have the same power as my waterboiler. That makes me realize about how much warmth that must generate.

  • "I don't know why these failed" He says with a frying pan over the lamp :D I really enjoy watching you come up with these crazy and non-relatable builds! Keep em coming

    • Non relatable? or u mean to say, un affordable. Cause I would do this all day. But I'm broke, comparativly speaking.

    • He was trying to cook something.

  • This could possibly be my favourite channel. Keep up the great work!

  • Ok, the 6:00 part convinced me! :D I always thought LED lamps where very small and energy economical The latest generation of nuclear reactors are cooled with a lead(not L.E.D., but Pb) and bismuth eutectic alloy (sodium + potassium may also do the job) instead of water. He might catch on on that ;)

  • Awesome! Theoretically this could compete with an HMI light. If they brought out any LEDs with even half this power, it would be incredible.

    • The spec of this LED unit definitely could compete with an HMI without needing 3-phase and much cheaper electrical costs (or HT 85 gels if you wanted 3200K), which means big savings in studios/sound stages and smaller, more portable generators on location. I think it would be far cooler and quieter once you are able to optimise the cooling solution, too. In theory you could have the cooling and power units on a long thick cable which you can sound and further temperature isolate also, in the same way you do with ballasts for HMIs. The overall cost for a proper industrial-manufactured unit would likely be much lower. The reliability of the colour rendering between units would noticeably be better, too, I think.

  • Please show us the outdoor test! Don’t forget about this project!

  • Would have loved to see this thing light up a forrest or a lake.. some tangible real world testing basically..

    • Andrew -Lithium Ion Battery bank.

    • I'd love to have this mounted in my car for idiots that have those gawd awful blue headlights, or can't turn off their high beams.

    • Part 2 mate...

    • This thing looks like it can light up an entire city on its own. lol

    • Hes gunna take that all outside how?

  • Which voltage rail were you pulling power from? The PSUs are rated for their total output, but each voltage rail can only output a certain amount, so if you were drawing everything from the +12V rail for instance then they wouldn't be able to keep up.

  • I never would've thought you basically need a whole water cooled pc to power an led

  • A phenomenal amount of light (from 10:00 in) and heat generated - ideal for those cold dark winter days.

  • I read the title as “Testing a $1000 LSD”. Would have been a pretty interesting video, but LEDs are cool too.

    • DIY lsd ain't all that great, but a 1000$ version = :-))))!

  • "So I'm gonna have to up the ante and use watercooling." Linus: "... go on ..."

    • @Alexandru Marzenco thats simply not true

    • And with all linus's tools, he still managed to make a better waterblock then LTT ever did

    • Linus: RGB?

    • @Corned Beef Gaming A/C is a bad idea for this. There's no benefit to having the LED colder than ambient, unlike a CPU. This is all about removing as much heat as possible.

    • Water cooled RGB

  • Great video! However, LED are not supposed to be driven by voltage, but by current. So, your LED drivers should have some max voltage set not to damage the LED chips, but you should regulate the output by current

  • It would be really cool if the power supplies could also be watercooled with the same loop, seeing as the LED didn't seem to be stressing it too much alone.

  • This guy is just so genuine with his passion

  • Server power supplies are meant to receive additional cooling! That's why it is loud and freezing cold in server rooms.

    • I'm surprised he didnt also route the water through and across all the voltage boosters' and power supplies' cases or heatsinks Even if just to remove noise, but as soon as he said they were hot to touch he gave himself something to kick himself over in hindsight

  • Given there are 6 separate powerable sections you could easily run the PSUs in totally separate circuits, be it 2 or 3 sections per PSU.

    • @CME 3 phase isn't common in a domestic setting here in the UK (where DIYPerks is from) it's only really used and available in business/commercial buildings. Some very old houses particularly in military towns may have functioning 3 phase still, but it's not the norm

    • @Dave Fayram If it is built to EU code, you should have 3 live's, 1 neutral and one for ground. You'd typically have a few triple phase plugs around the house, at 480V, usually used for stuff like electric stoves etc. Also hurts a lot to get shocked by. Then the rest have just 1 phase and 1 neutral for 240V. All of those are accompanied by the grounding cable.

    • @Viddiecent Wait, UK houses have separate neutrals but not the same hot? 🤔 Tell me more.

    • @Friso Smit You're probably right actually after watching again the LED seems to be powered in sections so input side probably wouldn't matter at all. probably should have paid more attention to that haha

    • Isn't that way LED on cars flicker on high speed shooting? To preserve longevity and power?

  • love your work!.. did you get around to lighting it up outdoors?

  • Hey man, I can almost guarantee that you would have had better performance from an off the shelf water block. The type of water block you created was essentially the first water block type. Then they changed to non restrictive pins, then they went to restrictive micro channels, now a days restrictive micro channels still exist but there are elsewhere restriction designed with high performance. The gist of things is that the new designs can remove a lot more heat than non restrictive pin or fin designs. So even with a reduction in surface area on the LED it would have still out performed it. What you built is probably more than the might needs anyways.

  • Just a couple minutes in, but a Threadripper AIO would have been large enough I bet, out of the box. Ah well.

  • You are awesome dude👍. All that so you could just light up the LED perfectly. Perfection 👍

  • This dude constantly looks like he's heard a dirty joke before turning the camera on - and it's great!

  • Amazing! I am surprised you did not damage your CMOS image sensor on your camcorder.

  • It’s more impressive that he does that, even though leds have the least heat production

  • What’s the power factor on those voltage boosters? You’re probably drawing far more Volt-Amps than 750w non-inductive power. I suspect those switch mode supplies are not designed to have highly inductive loads either

    • Yeah but you'd think those power supplies would have some kind of over-load protection.

  • Regarding the overloading of the power supplies, did you account for the energy loss of the voltage boosters?

    • I've read a lot of dumb theories of what went wrong before I got to yours. Good thinking

  • I think you missed something here. The total wattage of the power supplies you used is 2000W however that number is the combined load of 12v, 5v, and 3, Rails. As you are only using the 12v rail to power the LED. You need to use power supplies that have at least 1000w each on the 12v rails alone. I think this is the reason why your one failed. I hope this helps, best of luck!

    • Oh yeah, I just wrote the same. Also, I would not go with the kind of "Server PSUs" he used. They look pretty cheap with their metal cases and tiny fans. Why not just get two regular fat PC PSUs, there are plenty with more than enough power. For example the Seasonic Prime 1300W Platinum/Gold or the BeQuiet Straight Power 11 Platinum etc. The beefier fan will also offer way better cooling.

    • These look like they might be hot swappable power supplies from a server. In some cases these units output 12v only, then any reduction is done on the motherboard. So it's quite possible these PSUs are ample for the task (12v, 1000w x2), but as with anything server related, noise output is usually a secondary consideration.

    • best explanation ever! That's perfect!

    • Think you are over thinking it. As they are in SERIES then you will double the voltage but only at the rated current of 1 supply (ie 750w) To double the current/power you wire them in parallel but would only have 12v output.

    • His real issue is cooling.. cooling the power supplies. Theyre meant to operate at the back of the server chassis which is filled with fans moving tons of air over the components, power delivery components included. These poor things were left in an open air environment without any assistance outside their tiny little 40mm fans

  • now i want to see the slow mo guys to use this imagine the kind of shutter speed they can afford with that much of floodlight

  • This would be insane as a led light bar get like 20 of these with reflectors at about 70 degrees so it points super far

  • I can not express how much I am impressed with the custom water block!

  • The reason I believe power supplies failed is because when you were running them at max power the components inside power supplies got quite hot and as long as the fans were running the temperature remained under the limits but as soon as you may have pulled the power plug the fans, and the power supply turned off but the heat inside the supplies was accumulating as there was no way for it to wick away quickly as the fans are off at that point and the components got super hot super quick which resulted into failure of the supplies.

  • I think the PSUs failed because of inrush currents, since converters ususally have a ton of inrush current. Turning them all on at once is a bad idea. (If you have done that) If you want some references, check out Great Scotts Video on inrush current and soft starters. If you want some help designing your own psu, let me know, I am a hobbyist who does a lot of high power stuff (no degree yet, but I am working on that).

    • ​@TheBl4ckOne They look to be HP DL380G5 series powersupplies ! 12.15V / 82.3A is on the label ... One single rail.

    • @SukoSeiti But the heat can't increase one the power is off, so it's a moot point.

    • In 12:02 he just turned everything off instantly, and the heat on the power supply don't just magically vanish. He probably fried components with residual heat.

    • NTC in series. The same trick that PSUs use at their input.

    • Maybe a coil and capacitor as a simple LC filter would have protected them. Also they have different output voltages and usually can‘t give their maximum power at a single rail.

  • Seeing him make that heat sink makes me want to start building things out of copper and solder lol that shit did look good once polished up

  • Personally, I think you should have waited until after testing the block for leaks before attaching the LED, as water can damage it.

  • Your issue may have been where you pulled power from your power supplies. I believe they typically break up their full wattage delivery into multiple contact pads. Make sure you tie into all of the power supplying pads to get the full rated wattage out of them. Sounds like you may have only tapped into half of their supply points since it was under its rated output.

  • I am concerned for corrosion in your loop, dissimilar metals corrode, the solder and zinc in particular. The PSU spreads it load across multiple voltages.

  • "Hello, yes, police? I have reason to believe my neighbour is building an illegal arc reactor... yes like Iron Man, exactly"

    • LOL !

    • Next episode, Using nuclear fusion to make a light brighter than the sun.

    • We see exactly how Tony Stark did it all now...sponsors. :D

    • @Nathan Hoeltzle no

    • Owning a self sustained fusion reactor isnt illegal, thats what powers his suit

  • Fantastic channel!!! Power supplies failed for several reasons! One being they were pretty much maxed out at continuous load, those power supplies are not designed for that.( RE: to heat dissipation, electronic design of board components, etc.) Just as a general example look at a landscape lighting transformer/ power supply, If you used a general purpose transformer/power supply rated at the proper wattage it would burn up every time! The proper one is made specifically for that type of load! Not to even begin to talk about losses due to heat involved with creating the pwm needed for led's! A proper power supply/driver needed in this case will be totally different then what you used! A fyi tip on sizing the power supply,driver or transformer, always base it on continuous load and never size it equal to the load. The max load should never exceed 80% of the power supply, driver, transformer load! when these components are manufactured they are designed to operated at there ratings within temp. limitations based on over rating just for temp!

  • Is there going to be a follow up video to this? Would love to see the solution you came up with

  • You would have thought of this already but is there an issue with using the PSUs separately - rather than in series - so each can power up one half of the LED?

  • Perhaps the water cooling solution could be expanded to cover the heat exertion of the PSUs as well.

  • When wrapping PLA, add copper tape, then kapton tape, then copper tape, then aluminium tape. Also, PSU's have maximum draw on each rail - even a 1000W PSU will not deliver that on one rail.

  • Brother you are doing a great job thanks to you . I learned from you some engineering with my tech with your information with simplicity. Thanks you brother..

  • Well I'm glad it's not just me that this stuff happens to lol. When I saw all those power supplies and transformers I said to myself "there is no way I could get away with powering all of that up and not have something go up in smoke"... Then lo and behold, even for Matt, stuff went up in smoke in the end LOL

  • You should install a pressure/flow sensor/switch inline with the LED power supply. This way if the coolant stops flowing for any reason, it immediately cuts power to the LED to help protect it. No flow = No power.

  • From someone who works in a data center: Server PSUs never run in series. They run in n+1 redundancy. So they aren’t very good at working together at the same time. They also get a good bit of cooling from the fans within a server chassis as well as their own so it was probably inadequate cooling. They also never push their full wattage capacity for extended periods of time or reach their full capacity ever really.

    • @mikimomo97 No. PC power supplies are designed to run continuous, as well as have headroom (graphics cards spike power relatively often), as long as operated in spec temperature range.

    • @Midnight DracoFelis most PC PSUs are actually rated for full load on just the 12v rail, as well as any combination of rails (not exceeding their individual limits). So 1000W on 12v or 900w on 12v and 100w on 5v, for example

    • Also, the fact he was only using the 24V Rails, even tandem PSU's have other voltages, unless he was using CPU boosters, and you know those things aren't supposed to be run full load like that... or really at all as most of them are pretty cheaply built.

    • Maybe killing the power (fans) while the PSUs were still hot did not let them cool down well... hence need better cooling / monitoring for the PSUs themselves.

    • @Wieland Bach True, you would think that the self protection circuit would have saved the PS. could have had old ass cooked capacitors in it 🤷‍♂️ EDIT: he also showed the power draw from the wall, which was lower than the combined the PSU's are rated for. so yeah, they should not have popped either way, unless they were faulty or im missing something.

  • I loved when you got to see all the fans on the back of the radiators! Its all the old PC fans of the past come back for one last hurrah.

  • Have a similar setup using lipos and a bucky board it stays lit for a few hours at least 4 hours haven't left it on longer yet, and it's quite portable, interesting Project but total overkill

  • Absolutely Brilliant! Very much look fwd to the next run. Besides the $1300 cost, I could find uses for this; it would take 15 of the 100 watt units, but it looks like this won't be exactly a portable solution.