Nearly all heatpipes work with water vapor. Water under low pressure is ideal for cooling applications between 25 - 100 °C. This is exactly what most electronics need. If you hear anything about exotic gasses, they are just there to prevent corrosion (ideally you want no O2 or H2 in your heatpipe). (Exotic gasses are not used for cooling unless you're talking about cryogenics.) A heatpipe usually fails due collapse (dents), because they are under low pressure even when hot. Sometimes a heatpipe may fail by leaking due to micro cracks or a pinhole leak. There is very little water in a heatpipe. The one in the video likely only contains a couple of drops of water. Once a hole is formed, the water escapes as a vapor, usually very quickly when the heatpipe is hot. There may be a pinhole leak somewhere that you can find under a microscope. But the leak can easily be so small and so well hidden that you won't see anything on the outside even with high magnification. A properly functioning heatpipe should have a very small temperature difference over its length (, a couple degrees at most). This means it's showing a high heat conductivity, easily exceeding that of a regular copper pipe of the same size.
@@SusanAmberBruceno. Once the vapour evaporates from the cooler it’s toast. It has to remain sealed for it to function as the heat it picks up turns it to a gas then as it cools it turns back into a liquid where it returns to the soups the heat to then turn back into a gas. If that gas escapes then you will have nothing left to turn back into liquid to pick up more heat. When that happens it is literally just a hollow copper block then which isn’t as effective.
Another test is to hold one end of the heatpipe between your fingers, then dunk the other end into a cup of hot (not boiling hot, you'll hurt yourself) water. A working heatpipe will transfer the heat to the other end within a second or two and you won't be able to hold on. One that's busted will take much, much longer to move the heat and you may even be able to hold on indefinitely. The difference is massive.
Thank you for another brilliant video, Graham. I refurbish Dell Inspiron laptops as a hobby. I have found and replaced at least three faulty heat pipes on Inspiron laptops in the past two years. This video is a valuable addition to your growing library of highly educational and helpful content. Cheers!
Another indicator is how hot the air is coming out of the exhaust. If the system is overheating and the air coming out isn't really hot then there's an issue in the heat transfer. If the air is hot then the cooling is working correctly, taking it from the source and dumping it elsewhere. If the air is cold then it's it's not actually taking the heat and dissipating. Good video.
THANK YOU. I have this EXACT same issue on my Insprion 3785 that I bought in December 2019. I used to run Folding at Home every night, and it barely made any fan noise. Then one morning my laptop was shut down and I thought it was a update that forced itself. But soon it ran hotter and hotter and one day I witnessed a thermal shutdown when FaH started. Also like yours, it would porpoise between 380Mhz and it's normal clock speed. I even had it on an external cooling pad(I always use one on my laptops, b/c I prefer the high angle to type). There was a lot of heat under the left side of the keyboard, which I realized coincided with where the heat sink was after looking at teardown videos. I too replaced the thermal paste multiple times, and it still shuts down and runs ~80C under normal web browsing/video watching. I read on Wikipedia that when the pipe stops working, either by a leak of the fluid, or being outside of the designed operating range, the heat conduction drops to the level of just the base metal, which in the case of copper, is 1/80th of the functioning system. I watch a Gamers Nexus video on how heat pipes are made and they are soldered at one end, have fluid injected, then have the air vacuumed out, and crimped/soldered at the other end. They are then formed to shape. What I suspect is happening(especially seeing this happens to Dells a lot) is a manufacturer defect where one end of the pipe gets a microscopic leak, which both breaks the vacuum and also allows fluid vapor to escape, until there is nothing in the pipe to conduct heat, except the pipe itself.
Thanks for the info. How about a closer inspection of the pipe to find the exact cause of failure (Microscope, pressure test under water, cut it open...)
Excellent explanation! I have a Surface Pro that badly throttles similar to what you show. Will definitely look into following this through process to diagnose a potential heat pipe failure. Thank you so much for all your content, I watch every single video and soak in all you knowledge. This is what RUclips is all about in my world.
Great timing I think my Latitude 7400 may well have this very problem. It was reaching a 100C so I changed the thermal paste which was dry but It made very little difference. I had to change the thermal management settings in bios to "cool" in order to stop it still reaching 100C. I shall investigate further now and again many thanks for a great video.
My experience is that the Dell 7400, 7410 and 7420 are a pretty bad set of computers. We have a number of them at work that have failed with various symptoms, so maybe all falls back to a bad batch of heat pipes. My thought was that there was insufficient air flow.
Dave very good presentation on a failed heat pipe! Most heat pipes have a coolant, never air (just the vapor of the coolant) as you need the vacuum. Otherwise it’s a solid rod of copper (old school) If you have a micro scope camera (fiber optic) you could have cut the end cap off carefully not allowing the metal debris to enter the tube and sent down the scope into the tube. In this case this was defective from the factory as metal filings managed to clog the pipe. What I’ve mostly seen is holes in the pipe either caused by corrosion or a weak spot in the wall, so over time it fails. Before taking the heat sink off or even cleaning I often inspect both the logic board and bottom cover looking for the signs of a drip where the heat pipe/s are located as well as clumps of dust and signs of wet on the pipes surface. Lastly, just having even a laser IR thermometer is enough to make a quick scan like you did with your newer heat camera (I wish I had back in the day) BTW - who’s do you have as it looks quite nice!
Heatpipes contain several essential elements that allow them to efficiently transfer heat: Liquid: A small amount of liquid (usually water or a mixture of water and additives) is present inside the heatpipe Wick: A porous structure within the heatpipe that facilitates the movement of the liquid Evaporation Zone: At the hot end of the heatpipe, where the liquid evaporates due to the heat input Condensation Zone: At the cold end of the heatpipe, where the vapor condenses back into liquid Vacuum: The interior of the heatpipe is typically evacuated to create a vacuum, allowing for more efficient heat transfer So this failed one could have multiple faults.
Very interesting video! I don't remember seeing a heat pipe fail before, but then again, that's not something I'd normally look for. I have a Dell in my shop right now that has some severe performance issues that I can't seem to narrow down, so I think I'll check the heat pipe. Thanks for the video and keep up the good work!
Done watching, thank you very much for the informative repair video. I have learned significantly more troubleshooting & repair lessons in this tutorial video and to your other repair videos as well compared to my ENTIRE 4 YEARS OF COLLEGE due to the rotten & outdated standards of education here in the Philippines. I hope you will soon have a mini-series for Schematic & Boardview-free Voltage/Power Rail Tracing[12V/18-20V Main Voltage Rail, 5V, 3.3V, CPU/GPU Core Voltage Rail, DRAM Voltage Rail, IGPU Voltage Rail, System Agent/Northbridge Voltage Rail, PCH Voltage Rail, BIOS Voltage Rail, Battery Power Rail], Proper method of testing/checking of potentially faulty MOSFETs & ICs/Controller Chips, CPU/GPU/PCH Reballing and BIOS Bin File Editing.
Great demonstration. Wish I'd had a thermal camera to quickly diagnose my GPU when the heat pipe decided to die. (Reference 2080Ti, 1 Pipe failed. The other was fine. So the card was working just very poorly.) Replacing the cooler fixed it like a charm of course. But it was difficult to diagnose due to the fault.
Smart vid there. Not seen anyone show that. I bet that's whats wrong with my old laptop. Redone the paste on it twice, and cleaned the sinks out spotless, but it still gets hot and eventually shuts off. I don't have a heat cam, but I could give it the finger test to check. Nice one
I would never have found that! A part from an obvious kink in the pipe, I wouldn't have thought they could fail. And being metal, I would have thought they were pretty much fool-proof (unless bent!). I wonder if dust has blown back inside, causing a blockage? Maybe a strong air compressor used to clean the vents/fan?
My current work laptop is a similar-looking Dell Precision, and the i7-13850HX in it does get it quite hot when I'm running a build, but that's a genuine workload, plus I've got 15 minor things running in the background, not including the base Windows processes. Good to know I need to look out for failures other than CPU degradation with this one.
Thank you, very informative. This I will remember forever. I have had overheating issues before and went through hell to repair them not realizing that when I replaced the heat pipe. That was the problem. Now after seeing this vid. I understand what I did was the correct thing Afterall. Thank you.
1:45 - I love HWMonitor - it allows me to show my current CPU & GPU temps in the taskbar [at least, it does on Win 10...]. So I know that my CPU is currently idling at 33.
On my 10th-Gen Toshiba Portege the heatpipe is working but it would not boost more than x32 (supposed to go up to x44 at 4.95GHz). I added a 3mm thermal pad on top of the heatpipe and underneath on the PCB so that some of the heat transfers to the bottom of the keyboard and lower case and it now boosts to x42... a massive improvement. I also added thermal pads to the smps chips on the motherboard (6mm was required for it to touch the bottom case) and also on the NVME m.2 HDD. Every litrtle bit of themal pad helps to reduce heat and thus the laptop will last longer instead of chips overheating and failing.
amazing demonstration of a failing heatpipe, i had something similar in a desktop with an aftermarket cooler, it was not working at all, cpu high temp, after replacement all ok. I had no infrared camera at that time.
Not really. The vacuum has failed and the few drops of deionised water will have gone. Cutting it open won't show anything, there literally will be nothing to see 😉 Of course there may be a steam burn somewhere that will show where the pin hole appeared, but probably not enough to cause liquid damage.
i have had the same problem with my asus p52f a few years ago. was wondering why it kept overheating when no hot air was blowing out and the fan ran at full speed
I press the like button before watching the video, because as always I love the content! Graham, could you let me know which desoldering wick you use please? Thank you
You’d have thought the copper itself would be a good enough conductor of heat, regardless of what was inside? Just to mention a lot of modern multimeters now come with a temperature probe.
It's a good testament to the design of a modern heatpipe where the condensation and evaporation can take place in such a tiny space, and one successfully built can drastically outperform a basic copper pipe.
inside the copper heatpipe there should be copper dust and a couple drops of water, that evaporate and condensate moving heat from one side to the other that is how it works on desktops, but these copper heatpipes are thinner so there must be nothing apart the drops of water in there surely it was not properly sealed and water escaped, on one of the sides or near the heat finns there must be a hole or a fracture
Desktops with decent coolers certainly have heatpipes. My Noctua cooler uses 6 heatpipes that guide the heat from the hotplate into the finstack where it is dispersed by a fan duo placed in push pull configuration. Most stock coolers do use only a finstack with fan though.
Honestly you could probably get away with not burning ur fingers by using a generic thermometer for your kitchen or whatever and get the same results as the camera for a case like this it appears to be be pretty red or blue lol I it’s working or it’s not. I’m definitely going to take a look at my old dell laptops to check for this now that I’ve seen this video.
1. Probably there might be corrosion on the inner side which blocked the heat conductivity further till radiator. 2. There might be dust that get gathered from the hole at the CPU side. 3. Even a small bug might enter through the hole and died. These are just an assumption.
I literally have this exact problem now on a iffy GPU motherboard which i reflowed and got going again but THEN i looking at the temp after a put fresh paste on and its 75-80c idle and bit above?? I had the board out and the pipe was really hot but the cooler was stone cold so i thought WTF?? Maybe this explain the GPU failure overheated. I ASS sumed the copper pipe was just metal so cant fail but obviously something is very wrong and theres black along the side of the tube where the fins are SO... something has leaked out and now it wont transfer heat? i might put it in water and see if bubbles come out indicating a failure. I could fill it or partially but with what? Too much would explode? any rate i could solder the leak. Any ideas on fluid? Presario CQ62, i use for testing and experimenting. Might have to find another tube online and replace
Something I have never seen! I know they can't be bent etc but never seen a defect like this. Modern chips still amaze me, we're both from the era when a hot CPU would get hot enough to cook itself... kids these days don't even know. I once saw a demonstration of an Nvidia GPU running a load and the heatsink removed mid-benchmark, it just clocked down so far and eventually stopped executing... and then started back up as soon as the heatsink was re-applied, didn't even crash. Modern CPUs are basically the same as far as I know.
I might finally know why my old Dell started booting up to 600 MHz. All this time, I thought it was the OS. I might dig out that dinosaur and take a look.
I just replaced the heatsink on my Dell with a New Old Stock one I bought on the 'Bay, and for $14.00 and the time to take it apart, it's pretty much back to it's old self. I'm currently running Folding at Home on both the GPU and CPU at full, and the temps are consistently in the low 70's vs. before when the laptop would have had a thermal shut down even after throttling back to 300mhz.
Heat pipes are constructed of metal pipes made of copper, aluminum or other metals with high thermal conductivity, sealed inside which is a small amount of liquid called a working fluid (e.g., pure water) and a capillary structure (wick). When heat is applied to one end of the pipe, the liquid evaporates, absorbing heat in the process. This vapor travels to the cooler end where it condenses back into a liquid, releasing the absorbed heat. Similar to your refrigerator or air conditioner with no compressor needed. All these devices move heat from one place to another.
8:33 there's no blockage. Heat pipes are hallow with a small amount of "liquid" amonia mixed with alcohol and other stuff that i can't remember, that helps absorbing the heat which turns into vapor and the result of that process is heat being "transported from one end to the other" like a conveyor belt. In time that liquid can evaporate if the seal of the heat pipe is compromised do to manufacturing defects or plain old corrosion and the result of that is what we see on thermal camera.
I have never seen this problem before. Weird! Surely this laptop would always of had this problem? It's just a bit of copper. Like the tape fix for 'glare' on thermo. Cheers. Edit: I'm still trying to get my head round how this can happen. It's just a bit of metal.
Heat pipes are much more complex than just a piece of copper. They're hollow and contain sintered copper wick, which allows a small amount of fluid inside to move back and forth rapidly, continuously vaporizing and condensing.
For a few months now I've been grilling ever since i was able to acquire a Weber charcoal grill. Delicious stuff! Anyways, one fo the issues that comes with grilling and i guess any type of cooking is knowing when the meat is done. Internal temperature. Got myself a little digital meat thermometer to check temperatures. It might also help checking the temperatures on components as opposed to searing your finger on them. Funnily rnough, ive used it to help diagnose the AC when my wife tells me it's not blowing cold or whatever.
You saving that heat pipe for posterity? Post mortem that sucker. Show everyone what's inside so they understand that is just not a hollow tube. Those things are really hard to make, they are not smooth inside. Those things are more like ear canals than pipes. It's much more likely to be thermal paste than this failure. Good catch. I had one of these the other day. Also a Dell Inspiron but a 7573. Dog ass slow. Worked on it so long a message actually popped up that the CPU fan had failed. Hadn't seen that one. The fan was indeed dead though it was spinning when I started. This machine was painfully slow. Minutes per click. His were they using this thing even? So $13 later there's a new fan installed and it made zero difference. So yes thermal paste was the correct answer. They had been using like that forever so that paste was cooked and conducting nothing. I was out of mx4 but Fennel makes some g107 silver that works just fine and is under$20 for 100g. 100g is a lot, I've been working on this little tub for many years now and there isn't even a dent in it. When you buy Liquid Metal they give you 1.5g to do a PS5 if that helps put 100g in perspective. This G rating system is gold. The higher the g number the stickier the compound. If you buy any if those little modules to do projects get some g109 and upgrade or install heatsinks on your boards and save yourself some grief. Anyway new paste...like awakening a sleeping dragon in this case and that's something's I see pretty often. Splay that pipe open for us. Let's have a look. Got a friend with the right laser to cut copper? I'm thinking slice it like bread so we see good working sections and then the difference. Those pipes are crazy complicated and most people have no idea. No laser then cut hallway through then do the metal fatigue bend it back and forth thing. Peek inside maybe cut from the other side. See what you get.
Surely that fan should be cutting in at a lower temp, fan settings in bios could be wrong.I wonder if the original heat pipe got damaged by continued heat cycling. ?
@@Adamant_IT And Dell often mess with the fan curves through the laptops life... I remember the first firmware my laptop came with was much more aggressive which led to a bunch of silly complaints about it being too noisy, later EC firmwares seem to have made it too lazy to kick in fans IMO.
@@Adamant_IT I have the 17 inch version of this laptop, the 3785, and have the same issue, and people always suggest the BIOS, thermal paste, or the fins are blocked by dust. Like you said, there's no thermal settings in the bios, I replaced the paste 2x and the fins barely had any dust on them.
for a U-series Ryzen for laptops? its acceptable but when you're crossing the terittory of H and HX Ryzen CPU's , multiple heat pipes is definitely needed and a standard ❤👍
The far bigger problem in my 2500U was that it seemed to be hard coded to throttle the iGPU power after 10 minutes. Extremely annoying as it was sufficient for some light games at 60fps for 10 minutes, then would tank to under 30fps. Temperature never went above 60C.
@@Adamant_IT Like in the early t14 gen the Intel ones come with a single heat pipe but the AMD ones have a dual pipe config and the Intel one are notorious for heat issues
Nearly all heatpipes work with water vapor. Water under low pressure is ideal for cooling applications between 25 - 100 °C. This is exactly what most electronics need. If you hear anything about exotic gasses, they are just there to prevent corrosion (ideally you want no O2 or H2 in your heatpipe). (Exotic gasses are not used for cooling unless you're talking about cryogenics.)
A heatpipe usually fails due collapse (dents), because they are under low pressure even when hot. Sometimes a heatpipe may fail by leaking due to micro cracks or a pinhole leak. There is very little water in a heatpipe. The one in the video likely only contains a couple of drops of water. Once a hole is formed, the water escapes as a vapor, usually very quickly when the heatpipe is hot.
There may be a pinhole leak somewhere that you can find under a microscope. But the leak can easily be so small and so well hidden that you won't see anything on the outside even with high magnification. A properly functioning heatpipe should have a very small temperature difference over its length (, a couple degrees at most). This means it's showing a high heat conductivity, easily exceeding that of a regular copper pipe of the same size.
Can you heat up a heat pipe and put underwater and look for bubbles then maybe solder for repair
Thanks - Excellent info!
@@SusanAmberBruceno. Once the vapour evaporates from the cooler it’s toast. It has to remain sealed for it to function as the heat it picks up turns it to a gas then as it cools it turns back into a liquid where it returns to the soups the heat to then turn back into a gas. If that gas escapes then you will have nothing left to turn back into liquid to pick up more heat. When that happens it is literally just a hollow copper block then which isn’t as effective.
I've never seen a heatpipe fail you have opened my eyes and thank you for a very well make video, instant thumbs up.
I don't think i've seen a heatpipe fail before. Cool video
I can forsee the algorithm picking this video up two years from now.
Another test is to hold one end of the heatpipe between your fingers, then dunk the other end into a cup of hot (not boiling hot, you'll hurt yourself) water. A working heatpipe will transfer the heat to the other end within a second or two and you won't be able to hold on. One that's busted will take much, much longer to move the heat and you may even be able to hold on indefinitely. The difference is massive.
Great tip, ty
Great idea using the masking tape for the reflections! I'll have to remember this one. Cheers.
Thank you for another brilliant video, Graham. I refurbish Dell Inspiron laptops as a hobby. I have found and replaced at least three faulty heat pipes on Inspiron laptops in the past two years. This video is a valuable addition to your growing library of highly educational and helpful content. Cheers!
Another indicator is how hot the air is coming out of the exhaust. If the system is overheating and the air coming out isn't really hot then there's an issue in the heat transfer. If the air is hot then the cooling is working correctly, taking it from the source and dumping it elsewhere. If the air is cold then it's it's not actually taking the heat and dissipating. Good video.
I’ve been in this business for 12 years now and still learning from your video’s. Thanks Graham
THANK YOU. I have this EXACT same issue on my Insprion 3785 that I bought in December 2019. I used to run Folding at Home every night, and it barely made any fan noise. Then one morning my laptop was shut down and I thought it was a update that forced itself. But soon it ran hotter and hotter and one day I witnessed a thermal shutdown when FaH started. Also like yours, it would porpoise between 380Mhz and it's normal clock speed. I even had it on an external cooling pad(I always use one on my laptops, b/c I prefer the high angle to type). There was a lot of heat under the left side of the keyboard, which I realized coincided with where the heat sink was after looking at teardown videos. I too replaced the thermal paste multiple times, and it still shuts down and runs ~80C under normal web browsing/video watching.
I read on Wikipedia that when the pipe stops working, either by a leak of the fluid, or being outside of the designed operating range, the heat conduction drops to the level of just the base metal, which in the case of copper, is 1/80th of the functioning system. I watch a Gamers Nexus video on how heat pipes are made and they are soldered at one end, have fluid injected, then have the air vacuumed out, and crimped/soldered at the other end. They are then formed to shape. What I suspect is happening(especially seeing this happens to Dells a lot) is a manufacturer defect where one end of the pipe gets a microscopic leak, which both breaks the vacuum and also allows fluid vapor to escape, until there is nothing in the pipe to conduct heat, except the pipe itself.
Thanks for the info. How about a closer inspection of the pipe to find the exact cause of failure (Microscope, pressure test under water, cut it open...)
Excellent explanation! I have a Surface Pro that badly throttles similar to what you show. Will definitely look into following this through process to diagnose a potential heat pipe failure. Thank you so much for all your content, I watch every single video and soak in all you knowledge. This is what RUclips is all about in my world.
Great timing I think my Latitude 7400 may well have this very problem. It was reaching a 100C so I changed the thermal paste which was dry but It made very little difference. I had to change the thermal management settings in bios to "cool" in order to stop it still reaching 100C. I shall investigate further now and again many thanks for a great video.
My experience is that the Dell 7400, 7410 and 7420 are a pretty bad set of computers. We have a number of them at work that have failed with various symptoms, so maybe all falls back to a bad batch of heat pipes. My thought was that there was insufficient air flow.
Thanks Graham, I always thought they were solid copper and thinking, "how can this possible fail" always learning ! 🤠😎
Dave very good presentation on a failed heat pipe!
Most heat pipes have a coolant, never air (just the vapor of the coolant) as you need the vacuum. Otherwise it’s a solid rod of copper (old school)
If you have a micro scope camera (fiber optic) you could have cut the end cap off carefully not allowing the metal debris to enter the tube and sent down the scope into the tube. In this case this was defective from the factory as metal filings managed to clog the pipe. What I’ve mostly seen is holes in the pipe either caused by corrosion or a weak spot in the wall, so over time it fails. Before taking the heat sink off or even cleaning I often inspect both the logic board and bottom cover looking for the signs of a drip where the heat pipe/s are located as well as clumps of dust and signs of wet on the pipes surface.
Lastly, just having even a laser IR thermometer is enough to make a quick scan like you did with your newer heat camera (I wish I had back in the day) BTW - who’s do you have as it looks quite nice!
Heatpipes contain several essential elements that allow them to efficiently transfer heat:
Liquid: A small amount of liquid (usually water or a mixture of water and additives) is present inside the heatpipe
Wick: A porous structure within the heatpipe that facilitates the movement of the liquid
Evaporation Zone: At the hot end of the heatpipe, where the liquid evaporates due to the heat input
Condensation Zone: At the cold end of the heatpipe, where the vapor condenses back into liquid
Vacuum: The interior of the heatpipe is typically evacuated to create a vacuum, allowing for more efficient heat transfer
So this failed one could have multiple faults.
@@I_Am_Your_Problem what is your point then? Is the information incorrect?
Very interesting video! I don't remember seeing a heat pipe fail before, but then again, that's not something I'd normally look for. I have a Dell in my shop right now that has some severe performance issues that I can't seem to narrow down, so I think I'll check the heat pipe. Thanks for the video and keep up the good work!
Great use of the thermal camera and software to illustrate the issue with the heat pipe, great for people who are not familiar with how they work.
Done watching, thank you very much for the informative repair video. I have learned significantly more troubleshooting & repair lessons in this tutorial video and to your other repair videos as well compared to my ENTIRE 4 YEARS OF COLLEGE due to the rotten & outdated standards of education here in the Philippines. I hope you will soon have a mini-series for Schematic & Boardview-free Voltage/Power Rail Tracing[12V/18-20V Main Voltage Rail, 5V, 3.3V, CPU/GPU Core Voltage Rail, DRAM Voltage Rail, IGPU Voltage Rail, System Agent/Northbridge Voltage Rail, PCH Voltage Rail, BIOS Voltage Rail, Battery Power Rail], Proper method of testing/checking of potentially faulty MOSFETs & ICs/Controller Chips, CPU/GPU/PCH Reballing and BIOS Bin File Editing.
Intersting didnt know a heatpipe could fail. Thank you!
Very interesting, thank you. I now have a better understanding what a heat pipe does in action, and what to look for.
Great demonstration.
Wish I'd had a thermal camera to quickly diagnose my GPU when the heat pipe decided to die. (Reference 2080Ti, 1 Pipe failed. The other was fine. So the card was working just very poorly.)
Replacing the cooler fixed it like a charm of course. But it was difficult to diagnose due to the fault.
Smart vid there.
Not seen anyone show that.
I bet that's whats wrong with my old laptop. Redone the paste on it twice, and cleaned the sinks out spotless, but it still gets hot and eventually shuts off.
I don't have a heat cam, but I could give it the finger test to check. Nice one
What an eye opening video, very informative, thanks so much you're the boss 👍
An interesting source of failure... never thought about those pipes. Thankz! 👍
Interesting. Probably an illusion, the faulty pipe appears to be damaged close to the CPU mount (6:03) compared with the replacement (0.56)
I would never have found that! A part from an obvious kink in the pipe, I wouldn't have thought they could fail. And being metal, I would have thought they were pretty much fool-proof (unless bent!). I wonder if dust has blown back inside, causing a blockage? Maybe a strong air compressor used to clean the vents/fan?
My current work laptop is a similar-looking Dell Precision, and the i7-13850HX in it does get it quite hot when I'm running a build, but that's a genuine workload, plus I've got 15 minor things running in the background, not including the base Windows processes. Good to know I need to look out for failures other than CPU degradation with this one.
Thank you, very informative. This I will remember forever. I have had overheating issues before and went through hell to repair them not realizing that when I replaced the heat pipe. That was the problem. Now after seeing this vid. I understand what I did was the correct thing Afterall. Thank you.
I never fail to learn something watching your channel.
Keep up the incredible content!
How did the Lan party go hope you had fun Graham.
1:45 - I love HWMonitor - it allows me to show my current CPU & GPU temps in the taskbar [at least, it does on Win 10...]. So I know that my CPU is currently idling at 33.
I never thought of this. Great video! Thanks for sharing 🤩
"It's straight as an arrow", lol )))
Thanks alot for that interesting info and the great demonstration.
On my 10th-Gen Toshiba Portege the heatpipe is working but it would not boost more than x32 (supposed to go up to x44 at 4.95GHz). I added a 3mm thermal pad on top of the heatpipe and underneath on the PCB so that some of the heat transfers to the bottom of the keyboard and lower case and it now boosts to x42... a massive improvement. I also added thermal pads to the smps chips on the motherboard (6mm was required for it to touch the bottom case) and also on the NVME m.2 HDD. Every litrtle bit of themal pad helps to reduce heat and thus the laptop will last longer instead of chips overheating and failing.
amazing demonstration of a failing heatpipe, i had something similar in a desktop with an aftermarket cooler, it was not working at all, cpu high temp, after replacement all ok. I had no infrared camera at that time.
Very interesting! I didn't know heatpipes could fail without visible physical damage. Thanks for showing us!
good analysis 👏👏👏
Very nice! Thanks for all the great videos.
It would be intriguing to perform an autopsy on the thermal pipe.
Thank you.
Not really. The vacuum has failed and the few drops of deionised water will have gone. Cutting it open won't show anything, there literally will be nothing to see 😉
Of course there may be a steam burn somewhere that will show where the pin hole appeared, but probably not enough to cause liquid damage.
Interesting and informative. Thanks.
Great demo, I enjoyed that thanks.
i have had the same problem with my asus p52f a few years ago. was wondering why it kept overheating when no hot air was blowing out and the fan ran at full speed
Man i started blowing at the screen just to cool this one lol
I press the like button before watching the video, because as always I love the content! Graham, could you let me know which desoldering wick you use please? Thank you
You’d have thought the copper itself would be a good enough conductor of heat, regardless of what was inside? Just to mention a lot of modern multimeters now come with a temperature probe.
It's a good testament to the design of a modern heatpipe where the condensation and evaporation can take place in such a tiny space, and one successfully built can drastically outperform a basic copper pipe.
According to wikipedia, base copper has 1/80th of the heat conductivity of a working heat pipe.
That is a lot off heat going through there wow.
wow never seen a heatpipe failure like that, thats gotta be 1 in a million incident.
inside the copper heatpipe there should be copper dust and a couple drops of water, that evaporate and condensate moving heat from one side to the other
that is how it works on desktops, but these copper heatpipes are thinner so there must be nothing apart the drops of water in there
surely it was not properly sealed and water escaped, on one of the sides or near the heat finns there must be a hole or a fracture
Interesting (and disturbing) at the same time! Thank you
Damn, now even heatpipes can fail , another list of things to worry about in a laptop, or desktop for that matter 😂😂
Desktops have no heat pipe, just a large fan cooled finned chunk of aluminum.
Desktops with decent coolers certainly have heatpipes. My Noctua cooler uses 6 heatpipes that guide the heat from the hotplate into the finstack where it is dispersed by a fan duo placed in push pull configuration. Most stock coolers do use only a finstack with fan though.
Impressive! Thank you!
I would never consider this! Thanks!
Good tip with the masking tape
It’s all pipes !
I've seen dented ones, obviously that's a issue.
I had a overheating problem with my legion 5 pro and it isent the heat pipe
I watched this with my morning tea... been bothering me all day. What is in a heat pipe and what could cause it to clog up at a specific place?
Honestly you could probably get away with not burning ur fingers by using a generic thermometer for your kitchen or whatever and get the same results as the camera for a case like this it appears to be be pretty red or blue lol I it’s working or it’s not. I’m definitely going to take a look at my old dell laptops to check for this now that I’ve seen this video.
1. Probably there might be corrosion on the inner side which blocked the heat conductivity further till radiator.
2. There might be dust that get gathered from the hole at the CPU side.
3. Even a small bug might enter through the hole and died.
These are just an assumption.
I literally have this exact problem now on a iffy GPU motherboard which i reflowed and got going again but THEN i looking at the temp after a put fresh paste on and its 75-80c idle and bit above?? I had the board out and the pipe was really hot but the cooler was stone cold so i thought WTF?? Maybe this explain the GPU failure overheated. I ASS sumed the copper pipe was just metal so cant fail but obviously something is very wrong and theres black along the side of the tube where the fins are SO... something has leaked out and now it wont transfer heat? i might put it in water and see if bubbles come out indicating a failure. I could fill it or partially but with what? Too much would explode? any rate i could solder the leak. Any ideas on fluid? Presario CQ62, i use for testing and experimenting. Might have to find another tube online and replace
Ha, was about to ask what had happened to the pipe
Something I have never seen! I know they can't be bent etc but never seen a defect like this.
Modern chips still amaze me, we're both from the era when a hot CPU would get hot enough to cook itself... kids these days don't even know. I once saw a demonstration of an Nvidia GPU running a load and the heatsink removed mid-benchmark, it just clocked down so far and eventually stopped executing... and then started back up as soon as the heatsink was re-applied, didn't even crash. Modern CPUs are basically the same as far as I know.
1:20 The integrated gpu has seen some high temperatures 💀
HWmonitor is very buggy, HWINFO is better replacement @Adamant IT
LOL..."going like the clappers", what the hell does that mean, Graham? 😊
Interesting 👍🏻👍🏻
I never seen a heatpipe fail before
so simple when you know how.
good fix, makes a change from power rail shorts
I might finally know why my old Dell started booting up to 600 MHz. All this time, I thought it was the OS. I might dig out that dinosaur and take a look.
I just replaced the heatsink on my Dell with a New Old Stock one I bought on the 'Bay, and for $14.00 and the time to take it apart, it's pretty much back to it's old self. I'm currently running Folding at Home on both the GPU and CPU at full, and the temps are consistently in the low 70's vs. before when the laptop would have had a thermal shut down even after throttling back to 300mhz.
Heat pipes are constructed of metal pipes made of copper, aluminum or other metals with high thermal conductivity, sealed inside which is a small amount of liquid called a working fluid (e.g., pure water) and a capillary structure (wick). When heat is applied to one end of the pipe, the liquid evaporates, absorbing heat in the process. This vapor travels to the cooler end where it condenses back into a liquid, releasing the absorbed heat.
Similar to your refrigerator or air conditioner with no compressor needed. All these devices move heat from one place to another.
8:33 there's no blockage. Heat pipes are hallow with a small amount of "liquid" amonia mixed with alcohol and other stuff that i can't remember, that helps absorbing the heat which turns into vapor and the result of that process is heat being "transported from one end to the other" like a conveyor belt. In time that liquid can evaporate if the seal of the heat pipe is compromised do to manufacturing defects or plain old corrosion and the result of that is what we see on thermal camera.
I have never seen this problem before. Weird! Surely this laptop would always of had this problem? It's just a bit of copper. Like the tape fix for 'glare' on thermo. Cheers.
Edit: I'm still trying to get my head round how this can happen. It's just a bit of metal.
Heat pipes are much more complex than just a piece of copper. They're hollow and contain sintered copper wick, which allows a small amount of fluid inside to move back and forth rapidly, continuously vaporizing and condensing.
Thermal camera details please
I use the InfiRay P2 Pro
a coldpipe
A mid-week video? Hell yeah!
I ve seen it on Samsung
For a few months now I've been grilling ever since i was able to acquire a Weber charcoal grill. Delicious stuff!
Anyways, one fo the issues that comes with grilling and i guess any type of cooking is knowing when the meat is done. Internal temperature. Got myself a little digital meat thermometer to check temperatures. It might also help checking the temperatures on components as opposed to searing your finger on them.
Funnily rnough, ive used it to help diagnose the AC when my wife tells me it's not blowing cold or whatever.
You saving that heat pipe for posterity? Post mortem that sucker. Show everyone what's inside so they understand that is just not a hollow tube. Those things are really hard to make, they are not smooth inside. Those things are more like ear canals than pipes.
It's much more likely to be thermal paste than this failure. Good catch.
I had one of these the other day. Also a Dell Inspiron but a 7573. Dog ass slow. Worked on it so long a message actually popped up that the CPU fan had failed. Hadn't seen that one. The fan was indeed dead though it was spinning when I started. This machine was painfully slow. Minutes per click. His were they using this thing even? So $13 later there's a new fan installed and it made zero difference. So yes thermal paste was the correct answer. They had been using like that forever so that paste was cooked and conducting nothing. I was out of mx4 but Fennel makes some g107 silver that works just fine and is under$20 for 100g. 100g is a lot, I've been working on this little tub for many years now and there isn't even a dent in it. When you buy Liquid Metal they give you 1.5g to do a PS5 if that helps put 100g in perspective. This G rating system is gold. The higher the g number the stickier the compound. If you buy any if those little modules to do projects get some g109 and upgrade or install heatsinks on your boards and save yourself some grief.
Anyway new paste...like awakening a sleeping dragon in this case and that's something's I see pretty often.
Splay that pipe open for us. Let's have a look. Got a friend with the right laser to cut copper? I'm thinking slice it like bread so we see good working sections and then the difference. Those pipes are crazy complicated and most people have no idea.
No laser then cut hallway through then do the metal fatigue bend it back and forth thing. Peek inside maybe cut from the other side. See what you get.
Surely that fan should be cutting in at a lower temp, fan settings in bios could be wrong.I wonder if the original heat pipe got damaged by continued heat cycling. ?
Yea the fan curve on this was kinda wonky. But there's no settings on consumer laptops for this kind of thing, it's all baked into the EC.
@@Adamant_IT And Dell often mess with the fan curves through the laptops life... I remember the first firmware my laptop came with was much more aggressive which led to a bunch of silly complaints about it being too noisy, later EC firmwares seem to have made it too lazy to kick in fans IMO.
@@Adamant_IT I have the 17 inch version of this laptop, the 3785, and have the same issue, and people always suggest the BIOS, thermal paste, or the fins are blocked by dust. Like you said, there's no thermal settings in the bios, I replaced the paste 2x and the fins barely had any dust on them.
Cool...or nice this case Hot..
A single heat pipe like that should be criminal especially for older Ryzens
Standard issue in cheap consumer laptops 😔
It's only like, a 12w CPU, but still, laptops always have a 'minimum viable' cooling solution.
for a U-series Ryzen for laptops? its acceptable
but when you're crossing the terittory of H and HX Ryzen CPU's , multiple heat pipes is definitely needed and a standard ❤👍
@JoshuaG Considering it's an older Ryzen Zen+ part and crappy cooling. A second heat pipe would've cost literally cents
The far bigger problem in my 2500U was that it seemed to be hard coded to throttle the iGPU power after 10 minutes. Extremely annoying as it was sufficient for some light games at 60fps for 10 minutes, then would tank to under 30fps. Temperature never went above 60C.
@@Adamant_IT Like in the early t14 gen the Intel ones come with a single heat pipe but the AMD ones have a dual pipe config and the Intel one are notorious for heat issues