In video I missed to explain that the 4-layer mother-board have 2 other layers with grounded polygon pours.. Also as pointed out in some comments below, the long grounded thermal-track might act as antenna picking up rf noise. This is a trade-off for this self-reflowing concept, but to minimize the noise I added the possibility to short the track to ground from the two ends. This project is open source and you can find its files here - github.com/CarlBugeja/Open-Reflow You can also help me make more projects by joining my patreon community - www.patreon.com/carlbugeja
I build circuit boards for small satellites. Because they operate in space, keeping them warm is extremely important, and we attach external thermal heaters to them. I've thrown around the idea before of using an internal heater wire just like you have here, but never thought of using it for soldering. Very neat, will be interesting to see if this can remain reliable for other circuits!
On a side note, I wonder how solder in micro-gravity would react.... Like, if there was a fault somewhere, and it got zeroed down to a bad joint, would it be fiesible to send the board through a quick reflow cycle. Of course depending on the instruments onboard it could be a bad idea, out gassing or having essentially a big antenna in the board (although....could it be used to pick up external EMI and used for data correction possibly increasing sensitivity or detecting some kind of EM event to protect the rest of the circuit.... hrmm, Im not into RF voodoo enough to know, just a random thought...), but integrating a heater strip in the PCB would certainly cut down on the individual components provided someone doesnt put a backup external heater in the SOP.....
I would assume you use military grade components; arent those rated for functioning in like -55°C? Or does the temperature in space actually get THAT cold?
@@Dozer456123 But theres way less means for temperature to dissipate, as it is void. So i would think for something to get to that temperature it would need whole years.
@@GLnoG420 space can get to 270 °C or −455 °F, it really is that cold. Instruments that operate there need to tolerate an enormous thermal range, because in the sun that temperature can swing the other way by 600 degrees. Space is a whole different ballgame; electronics have to be heated, shielded and extremely robust.
I can't speak for Carl, but perhaps the inspiration was a short on a board. They can get very hot. It's not much of a stretch to go from there. I think the main reason for not doing this is that it's a slow process compared with simply running it through a reflow oven. Or a hot plate for that matter. The person who had the idea of putting tiny balls of solder in paste flux is the real genius. Why mess with an iron and solder wire when you can just smear past and heat it up.
At first I thought this was just a fun whimsical and sort of silly idea but then it turned out to be extremely practical. This is what real innovation looks like.
Absolutely. I think this is a huge leap forward for small/individual board designers. Someone has a cool idea (like @alvaroprieto and his USB cable tester) - a great product but if you don't have SMD gear, pre-built was the only option. Now anyone could use this approach to sell small SMD kits without the need to pre-build boards. Only issue is the cost Carl mentioned. Sometimes I forget I'm not watching a big electronics R&D company channel when I'm watching Carl's videos.
@@kick1ass20 the cost of getting basic equipment for smd soldering is much cheaper than the cost increase of using higher temperature boards. Even a basic PTC hot plate works well enough for smd soldering and you can get those very cheap.
@@conorstewart2214 yeah, but if it's a slightly customizable board and you want to buy a one-off kit on ebay where you can tweak a few components and then SMD solder it "yourself" like this - this is really cool
@@gd.ritter if you are going to the effort of buying a kit where you need to tweak smd components you can at the very least buy a hot plate or something. With this you need a power supply, constant current is probably best unless they can work out what voltage works for that specific board which would take a bit of effort on the designers side and it needs to be quite a powerful power supply since there probably isn’t that much resistance on the heating trace and it will need relatively high current to get it to the required temperature. This isn’t just a one off solution since you will still need a decent power supply to get it to work. Edit: this is more complicated than buying a PTC hot plate and wiring it up, since they are self regulating and that would allow you to use normal solder paste to reflow it, not low temperature stuff.
It's a very inventive idea! I might worry about using a long meandering trace as a ground plane afterwards. If there were no taps to the circuit except in one place, the 'plane' might act like an antenna and pick up RF noise. If there were multiple taps instead, I'd worry that they may offer alternate current paths during soldering.
You will need to create an additional layer with a proper ground plane and yes, a whole layer made of a long coil might create some issues, especially in high frequency applications. That's basically a huge inductor it might even create a large parasitic capacity on the board
I think that TF noise will definately be a broblem but maybe her ccan add some solder bridges und the botomn, that he handsolders after he is done to ground. That should fix the issue
I wonder if you could just use a regular ground plane as the heating element. One connection along each side of the PCB. It would require much higher currents and it might not heat the plane evenly, but since copper is very thermally conductive, the plane could maybe spread the heat well enough. Could be worth a try. A welder might work as the power supply if the current can be adjusted down low enough so the PCB doesn't instantly vaporize.
You could always shorten the trace by adding vias in various points connecting it to the top surface, where you could connect them later via more zero-ohm resistors, perhaps. Not sure how practical that is since the bridges would have to be added after the initial reflow is completed.
This is probably one of the coolest electronics videos I've seen in a long time. I have no idea whether this is a good idea in the long run, but it's a really clever idea.
This is like the early rep-rap project. instead of 3D printers making more 3D printers. Its PCBs making other PCBs. Very Interesting project! I always look forward to seeing what you come up with next.
There was a PCB-based Hotplate that I didn't want to build because I was scared of the SMD parts that make up the control circuitry. I thought to myself "Man, if only I could use the hotplate to assemble the hotplate", and now this concept is real!
The concept of tools making better versions of themselves has me thinking, what is the first ever mother tool with which the entirety of humanity's tech is built upon? Where is it now?
It also makes for an interesting self-destruct if there's any sensitive information on there. Plus i can see this being used in military applications where one might not have all the equipment needed to do the job this would make for a quick and easy solution in the field.
@Peter Fitzpatrick we have hot air gun, a hot plate which is most normally used, but for someone who doesn't have them both and isn't willing to buy those then this is the best solution. But still, it has its own cons tho.
Very "out of the box" thinking, really like it. If components had already solder on the legs or solder on the PCB pads with extra flux in it, you can make easy to distribute (starter) kits that doesn't require a solder iron. The only thing you have to do is placing the components on the PCB and apply a power source to flow pads and pins. The next thing you have to do is cleaning the flux off with alcohol and you are ready to go! Pure magic! Love it!
Hey ! Very cool idea! There is one more thought - you can not use a thermocouple - you can connect the heating layer periodically - first to a current source, and then to a resistance meter - copper has a very linear function: R=f(t°) - so you can find out the temperature knowing the resistance of the heating layer. Good luck to you ! You are a genius !
This is seriously awesome. As an EE Tech, something like this integrated into prototype PCBs would be amazing for rework on high thermal-mass PCBs. Just pull the 0 ohm resistor and the PCB itself could become an integrated pre-heater for rework
@@dfgaJK you don’t need a winding path either, you could simply use more current and a ground plane instead. Then it could do double duty as a heat source for soldering and isolation without the giant vibrating antenna in your board the floating trace leaves.
This is absolutely amazing. For home gamers, it could cost cosf massively not having to have others assemble the pcb or buying all the equipment yourself. I think the biggest application might be for really small order number niche pcbs where just one person is selling kits and wants to reduce the price by having the user do this part of the assembly or by doing a lot of them at once semi automated at home. Its even greater because you could potentially have boards where most comoonents will stay the same but where you leave extra tracks just in case you need to change the configurwtion and you can then do so at home just with component placement on a more hobby friendly pick and place like the LumenPNP
Such a smart idea. I had to gasp when the PCB price jumped from $56 to over $150 for the 5 boards with increased heat tolerance. A solid no-go for me, at least for now.
This is amazing thing that you have presented in this video.... No need to have hot air gun & costly soldering stations... Please make detailed video about this topic... & please share problems & difficulties that you have faced during this process. So that anyone could easily overcome through it....
7:52 the stench of burnt PCBs is incredible. I had a similar issue once, and it took me almost a week until the smell went away. Apart from this: cool idea!
For sure you were not the first person thinking of this idea, but it was the first time I saw it, and I'm currently struggling with a lack of the necessary soldering equipment for my project =] Anyway, this video "plus" the open-source content "plus" the generous share of the advices for minimizing failures, made me so excited for finding you in my UA-cam feed! Please keep up with this amazing work you are doing!😍💚
Man I love it when people do such smart things that seem like it could have been invented eons ago... Love to have found your channel man. Be well and happy, Peter Lunk
This could be a feature on extreme overclocking motherboards, so when overclocking with sub zero temperatures, the board could keep itself warm to prevent condensation forming.
I’m commenting to congratulate you for being wonderfully original. No way there are no practical applications for self soldering pcb, i can essentially skip any type of traditional soldering and just lay out the components and then heat it using itself. Blown. Away.
This idea is phenomenal. Not only did you build a reflow controller it self-propagates. I just bought a Miniware MHP30 for reflowing small PCBs but it doesn't have the same ability to program a heat curve. With your design I can slightly redesign my existing circuits and guarantee better outcomes. You're extremely gifted. Keep it up!
this is really cool. like people have said, this could be really really good for kits. i'm thinking a kit which mixes smd and through hole components and uses this for the smd parts and lets people solder the through hole components by hand would be quite fun as an intermediate soldering teaching kit, to teach people about the smd soldering heat curves in a way that is fun and interactive but doesn't require previous experience with smd stuff
Great Idea. Only problem could be EMI and SI issues, because the gnd plane should not have any cuts, in order to always have a good return path for the signals on the adjacent layers. But for all non high speed designs, this should be no issue. Very innovative, keep up the great work.
Very impressive!! Love the idea so I am going to use this idea to heat up a pcb in very cold enviroments.. I have a design for a car that does not fully boot when power on in below - 4 degrees.. This could be utilized as internal heater for components and the enclosure. Adding a fuse and a simple controller dedicated for this purpose would do nicely I think. Thanks for sharing! 👍👍
This indeed is an innovative idea! I'd only be worried by the interrupted current paths on this kind of ground plane. Maybe adding more layers to include a solid ground plane and leave the heating layer only for that purpose, could make this even better! Love your projects Carl!
@@TheOriginalEviltech It might cause the device layer to pick up RFI, as mentioned in other comments, by virtue of being underneath it, but the video implies that it would be used as *the* gnd plane, which would risk ground loops.
I love that you didn't need to change anything in the manufacture part so the price is the same. Just adding more functionality to something that is already there. If it were more expensive or required certain types of pcb it would be a harder decision, but as it is, i can't ask for more
Great way to innovate: you take something that already exists for one purpose (Eg.: heated bed of 3D printers) and use the same principle in a different application, adding the microcontroller bit to make it awesome and self-replicating. Makers community needs more people like you.
Really cool idea though! One major downside to using this as a primary gnd plane though, would be that youre likely creating nasty inductances that can show up as ringing along your supply lines if there are many pulsing currents on this board. If you have enough board layers, it would be useful to have a primary solid gnd plane and this be the chopped up secondary plane. Great work!
Dude that is the best idea ever. A self soldering circuit board seems so much smarter than blowing surface mount components all over the bench with a hot air gun. And you get the added bonus of the earth plane when your finished
I used thermal paste with low melting point I am surprised that you don't talk about the downsides ! 1° High price because of antimoine or bismuth content 2° low thermal >>> and very low electrical
I could see this being useful for increasing longevity. I don't actually know enough about circuitry to know what I'm saying, but I've also fixed a few DoA electronics by reflowing in the oven. I just imagine having that built into the boars would be amazing
Brilliant work. The heating function could be very useful if the pcb needs to operate in extremely low temperatures. A possible application in Mars rovers that have to keep their circuits at the correct operating temperature.
@@CarlBugeja i have a light switch from the 70s with thermal pcb tracks. no idea why it needs to stay warm but it does. i took it from an abandoned canadian prisoner of war camp
hahaha this is really amazing to watch, as soon as you explained your idea I thought damn why did I never thought about this. Very smart and intuitive that the PCBA your testing it with is rightaway the controller board for the project
I can definitely see this technique being standardized for production processes. It seems it would lower costs and maintain or improve reliability on soldering stages of a production.
I just love electrical engineering. everyday I see another wonderful person having a crazy personal project that could be the start of something really cool just itterating into it and putting out there for everyone to make something of it. I can't wait to finish my degree so I can have time for personal projects again. edit: also as someone who really really likes to make keyboards. this could be useful for the diodes. especially if someone is trying to make a lot of custom keyboards on the side to sell or really like to try a lot of different stuff. this is what i thought off. too bad I am terrible with circuits :P well no better way to learn than to fail I guess
Ah a fellow Maltese person, always great to see :) fantastic project! I love soldering my own PCBs but this would be such a revolution to doing things it's awesome dude!
Could definitely see this being used in entry level hobbyist kits for people who aren't too confident with soldering, especially considering surface mount pcbs ain't the most beginner friendly, also the possibility for bootstrapping using this system is promising too
Just use a cheap toaster oven for reflow then you don’t need to pay for additional layers, one run of several boards is actually more expensive. Bonus points for a thermometer.
Dude! Whilst I've enjoyed your videos for a long time; the production qulity, your passion, and clear enegineering talent. I've always been cynical of the applications. This, however, is awesome! Love it! Such an amazing idea.
I can see a realistic use for this in cold weather applications after assembly. It would save on weight and complexity of the build. You literally cannot get closer to direct heating with this idea! 🤙
I am stupid but couldn't he measure the temperature using the resistance of the loop? It probably wouldnt be accurate but it should show something useful.
Whoa. I'm currently working on a LiPo battery discharging circuit - want a super cheap board I can make 50 of and use them to keep my LiPos at storage voltage, and the entire back layer is just a long trace to use as a power sink when discharging the battery. This is literally the *perfect* idea for this project as I already have the trace, and need an easy way to solder up a large number of them!! Crazy, just crazy. I never would have thought of that. This is brilliant.
Cool idea! I wonder if this would benefit a larger scale production? Great for kits, but what about a small business that make circuits by hand or have a small assembly line of employees? I would love to make guitar pedals & this might be the way to produce a small batch :)
Sometimes this may be quite useful if you want to reflow a connector, which does not tolerate the high temp well. Inside PCB heating does not heat up the entire connector as much as a reflow oven would!
When reflowing you have to heat everything up to the same temperature otherwise you get thermal stressing of components and they fail (especially during the cooling phase). That is why there is a soak period in the reflow thermal profile! You need to understand the process and why it is what it is before cutting corners for the sake of simplicity!
@@simonbaxter8001 you do not have to heat everything up to the same temperature to eliminate thermal stressing, just the very rigid parts. ICs are quite rigid, some MLCCs too. But the connectors I am thinking about, which need somewhat lower temps, only introduce low amount of mechanical stress, no matter how you solder them, because of the relatively long and flexible leads (they are designed not to add extra stress, because being connectors already adds external stress). Lower profile parts would likely heat up better with inside PCB heating, compared to connectors. So using PCB heating does not seem an instant death sentence as you seem to imply. You do have to rate each component for this process though, and doing an x-ray on every BGA part after reflow would be highly advisable to catch cracked balls!
@@adamrak7560 If you have to xray bga devices to check their integrity, you would use proper controlled manufacturing processes, not cheap untested self soldering PCBs!
@@Tasty_sand Yeah, should be good. Although I don't know how thermally conductive FR4 is. While the outer temp for the component layer might be inside spec, the inner heating layer might have to be a bit hotter. But since hotplates work fine, I don't really think that this solution should be an issue.
@7:00 "So freaking Cool". yep, you said it. so cool. I can see this in a MI scenario, 'this circuit will self-destruct in 5 4 3 2 ...' gone. Or the last thing you do as you eject is fry the guidance systems.
I would think you'd want multiple attachment points of the soldering/groundplane to circuit ground, otherwise all ground current has to flow through that one resistor and it creates a large capacitor
This is a great idea! And what Joey Murphy said about satellites made me think, why not leave the thermal port intact? I can see that be relevant in the future for example if a pcb needs to be reflowed. In general that can be useful instead of having to put a pcb in the oven but in space even more to remotely reflow a pcb if it fails.
I don't mess around much with soldering and the like, but this was the coolest process to watch and see. As others have said, I predict this could be the solution for kits with SMD components, especially as the paste seems to be sticky meaning that the placement can be done much easier, and not having to deal with learning the special technique of hot airing that can blow components away.
It would be interesting if you could use the resistance-temperature dependence of the trace to skip the external probe. The resistance of the copper should be quite linear up to well over 200℃ and have a temperature coefficient of about 3930ppm/℃ so going from 20℃ to 150℃ would increase the resistance with about 51%. You could either have Altium estimate the trace resistance, measure it yourself or if you have a temperature sensor on the controller PCB you could use that as the initial guess for the temperature of the PCB you want to solder
Nice idea, BUT you really need to appreciate the very real EMI, EMC, RF and signal cross talk problems that those heater traces are going to cause! Maybe you need to learn some PCB layout fundamentals and signal integrity knowledge before pushing this idea too much as your stepping into design integrity minefield here.
I'm pretty sure anyone who is making a PCB capable of self heating and including a USB-programmable MCU and thermocouple probably knows what a wire antenna is... Relax a little.
@@ryanchojnacki4421 Not sure they do. And I'm pretty sure all those that think it's a good idea on here are being taught a really bad idea. This needs a followup with some real electronic principles and a demonstration of the issues it causes. Definitely not one for the average maker, but definitely one for those that will do electronics as a career.
It is definitely going to be a problem when the circuit on top is operating at high speeds. The examples I have given are all operating at slow speeds, expect for the clock source which I still think it's safe considering it's very close to the mcu and there are other grounded polygon pours on other layers.. But I intended this idea to be used for maker projects, and not something that can pass esd/emc/emi .. Maybe I can explore that in the future but as you pointed out rf is not my field of expertise
@@CarlBugeja Thanks for the comment. Anything above 10Mhz which any digital signal will produce (edge frequencies in fast rise times), the micro itself if it has a PLL, usb is above 10MHz, any SPI bus to peripherals port displays, etc. Robert Ferenac demonstrates in his videos the effects of these issues and the frequencies at which they become an issue. You idea is pretty innovative, but in my experience it would cause anyone who takes up electronics, as a serious career or business, some headaches. But I would hope that before applying this idea to a design, that they would understand the implications and rule it out as a very bad idea!
This is so much more interesting from a rework after a party fails perspective. I would love to have a selective heater under a bcg component or some 0201 resistors
What a great idea, just beautiful! Thank you so much for sharing and making this open source. How has noone ever thought of that before? Or maybe someone has, but didn't show off the results? Gotta try this at some point. It boggles my mind how far home brewn electronics have come along in the last years. I recall when the only option was to etch your own pcb and what kind of a mess that was. Nowadays you can just order multilayer boards with all kinds of fancy customization for pocket change, it's every hackers dream come true.
Depending on how thin you can make those traces you might be able to design the ground plane like the mesh on a microwave, this would only work for square PCBs and has its own issues, but provides better grounding and wouldn't have issues with RF. Just using a hotplate is still much easier and faster though.
In video I missed to explain that the 4-layer mother-board have 2 other layers with grounded polygon pours.. Also as pointed out in some comments below, the long grounded thermal-track might act as antenna picking up rf noise. This is a trade-off for this self-reflowing concept, but to minimize the noise I added the possibility to short the track to ground from the two ends.
This project is open source and you can find its files here - github.com/CarlBugeja/Open-Reflow
You can also help me make more projects by joining my patreon community - www.patreon.com/carlbugeja
Great, Thanks
This is amazing.
This is next level Genius, well done 👍
Just looked at your videos list, you're doing some INCREDIBLE work 😲
Maybe you can use it as feature, not a bug
Self-soldering antenna included board... Freaking awesome!)
It’s isolated and grounded… any rf caught would discharge.
I like the idea of using it as an additional wifi/cell/Bluetooth antenna
I build circuit boards for small satellites. Because they operate in space, keeping them warm is extremely important, and we attach external thermal heaters to them. I've thrown around the idea before of using an internal heater wire just like you have here, but never thought of using it for soldering. Very neat, will be interesting to see if this can remain reliable for other circuits!
On a side note, I wonder how solder in micro-gravity would react.... Like, if there was a fault somewhere, and it got zeroed down to a bad joint, would it be fiesible to send the board through a quick reflow cycle. Of course depending on the instruments onboard it could be a bad idea, out gassing or having essentially a big antenna in the board (although....could it be used to pick up external EMI and used for data correction possibly increasing sensitivity or detecting some kind of EM event to protect the rest of the circuit.... hrmm, Im not into RF voodoo enough to know, just a random thought...), but integrating a heater strip in the PCB would certainly cut down on the individual components provided someone doesnt put a backup external heater in the SOP.....
I would assume you use military grade components; arent those rated for functioning in like -55°C?
Or does the temperature in space actually get THAT cold?
@@GLnoG420 space is neg 270c
@@Dozer456123 But theres way less means for temperature to dissipate, as it is void. So i would think for something to get to that temperature it would need whole years.
@@GLnoG420 space can get to 270 °C or −455 °F, it really is that cold.
Instruments that operate there need to tolerate an enormous thermal range, because in the sun that temperature can swing the other way by 600 degrees.
Space is a whole different ballgame; electronics have to be heated, shielded and extremely robust.
i started soldering more than 25 years ago, and this never in a million years occurred to me. Outstanding.
I can't speak for Carl, but perhaps the inspiration was a short on a board. They can get very hot. It's not much of a stretch to go from there. I think the main reason for not doing this is that it's a slow process compared with simply running it through a reflow oven. Or a hot plate for that matter. The person who had the idea of putting tiny balls of solder in paste flux is the real genius. Why mess with an iron and solder wire when you can just smear past and heat it up.
bro imagine saying saying solder yourself to your pcb lol
@@evoluckievo Don't try it. It hurts like hell and the solder doesn't stick to skin anyway. It just balls up. Maybe they need to make some skin flux.
This is SO cool Carl. Gonna have to try this at some point.
Thanks Scott! It would be fun watching you try this out
Carl Bugeja x Stange Parts crossover episode when?
Wow praying for speedy recovery and more vlog of xichen market and more waiting strange
This is non very cool, this is very hot! ;)
Of course Scotty has been here before me! Looking forward to it!
At first I thought this was just a fun whimsical and sort of silly idea but then it turned out to be extremely practical. This is what real innovation looks like.
This is amazing, i could definitely see this being incorporated into peoples kits
Absolutely. I think this is a huge leap forward for small/individual board designers. Someone has a cool idea (like @alvaroprieto and his USB cable tester) - a great product but if you don't have SMD gear, pre-built was the only option. Now anyone could use this approach to sell small SMD kits without the need to pre-build boards. Only issue is the cost Carl mentioned. Sometimes I forget I'm not watching a big electronics R&D company channel when I'm watching Carl's videos.
@@kick1ass20 the cost of getting basic equipment for smd soldering is much cheaper than the cost increase of using higher temperature boards. Even a basic PTC hot plate works well enough for smd soldering and you can get those very cheap.
@@conorstewart2214 Maybe it is just a novel idea at this stage. Perhaps with material substitution it could become more practical.
@@conorstewart2214 yeah, but if it's a slightly customizable board and you want to buy a one-off kit on ebay where you can tweak a few components and then SMD solder it "yourself" like this - this is really cool
@@gd.ritter if you are going to the effort of buying a kit where you need to tweak smd components you can at the very least buy a hot plate or something. With this you need a power supply, constant current is probably best unless they can work out what voltage works for that specific board which would take a bit of effort on the designers side and it needs to be quite a powerful power supply since there probably isn’t that much resistance on the heating trace and it will need relatively high current to get it to the required temperature.
This isn’t just a one off solution since you will still need a decent power supply to get it to work.
Edit: this is more complicated than buying a PTC hot plate and wiring it up, since they are self regulating and that would allow you to use normal solder paste to reflow it, not low temperature stuff.
It's a very inventive idea!
I might worry about using a long meandering trace as a ground plane afterwards. If there were no taps to the circuit except in one place, the 'plane' might act like an antenna and pick up RF noise. If there were multiple taps instead, I'd worry that they may offer alternate current paths during soldering.
You will need to create an additional layer with a proper ground plane and yes, a whole layer made of a long coil might create some issues, especially in high frequency applications. That's basically a huge inductor it might even create a large parasitic capacity on the board
Will it be possible to recreate this by picking two GND points and letting current across?
I think that TF noise will definately be a broblem but maybe her ccan add some solder bridges und the botomn, that he handsolders after he is done to ground. That should fix the issue
I wonder if you could just use a regular ground plane as the heating element. One connection along each side of the PCB. It would require much higher currents and it might not heat the plane evenly, but since copper is very thermally conductive, the plane could maybe spread the heat well enough. Could be worth a try. A welder might work as the power supply if the current can be adjusted down low enough so the PCB doesn't instantly vaporize.
You could always shorten the trace by adding vias in various points connecting it to the top surface, where you could connect them later via more zero-ohm resistors, perhaps. Not sure how practical that is since the bridges would have to be added after the initial reflow is completed.
This is probably one of the coolest electronics videos I've seen in a long time. I have no idea whether this is a good idea in the long run, but it's a really clever idea.
I'd say it's one of the hottest electronics videos 😂
This is like the early rep-rap project. instead of 3D printers making more 3D printers. Its PCBs making other PCBs. Very Interesting project! I always look forward to seeing what you come up with next.
There was a PCB-based Hotplate that I didn't want to build because I was scared of the SMD parts that make up the control circuitry. I thought to myself "Man, if only I could use the hotplate to assemble the hotplate", and now this concept is real!
The concept of tools making better versions of themselves has me thinking, what is the first ever mother tool with which the entirety of humanity's tech is built upon? Where is it now?
It's almost like the hardware engineering version of self-interpreter software.
you think the machine that makes usual pcb doesnt have a pcb in it?
@@mfaizsyahmi a rock
It also makes for an interesting self-destruct if there's any sensitive information on there. Plus i can see this being used in military applications where one might not have all the equipment needed to do the job this would make for a quick and easy solution in the field.
Woah! This is super cool. A lot of people struggle with soldering SMD components. I see this as a great solution for kits
Whats wrong with using a hot plate ?
The solder mask and component-placement is the same process in both cases anyway... ? 🤔
@Peter Fitzpatrick we have hot air gun, a hot plate which is most normally used, but for someone who doesn't have them both and isn't willing to buy those then this is the best solution. But still, it has its own cons tho.
@@WarriorRev6300 I've used my stove more than once 🥴
@@tacitus_ That's also a choice, I used it once to desolder smd components lol.
@@tacitus_ my stove is gas, it didn't pan out too well.
Very "out of the box" thinking, really like it. If components had already solder on the legs or solder on the PCB pads with extra flux in it, you can make easy to distribute (starter) kits that doesn't require a solder iron. The only thing you have to do is placing the components on the PCB and apply a power source to flow pads and pins. The next thing you have to do is cleaning the flux off with alcohol and you are ready to go! Pure magic! Love it!
Cool idea! It even has a hidden feature, on-board rapid intentional disassembly function!
whoa @_@
Hey ! Very cool idea! There is one more thought - you can not use a thermocouple - you can connect the heating layer periodically - first to a current source, and then to a resistance meter - copper has a very linear function: R=f(t°) - so you can find out the temperature knowing the resistance of the heating layer. Good luck to you ! You are a genius !
This is seriously awesome. As an EE Tech, something like this integrated into prototype PCBs would be amazing for rework on high thermal-mass PCBs. Just pull the 0 ohm resistor and the PCB itself could become an integrated pre-heater for rework
The large ground plane would immediately turn form bane to blessing!
@@dfgaJK you don’t need a winding path either, you could simply use more current and a ground plane instead. Then it could do double duty as a heat source for soldering and isolation without the giant vibrating antenna in your board the floating trace leaves.
This is absolutely amazing. For home gamers, it could cost cosf massively not having to have others assemble the pcb or buying all the equipment yourself.
I think the biggest application might be for really small order number niche pcbs where just one person is selling kits and wants to reduce the price by having the user do this part of the assembly or by doing a lot of them at once semi automated at home. Its even greater because you could potentially have boards where most comoonents will stay the same but where you leave extra tracks just in case you need to change the configurwtion and you can then do so at home just with component placement on a more hobby friendly pick and place like the LumenPNP
Such a smart idea. I had to gasp when the PCB price jumped from $56 to over $150 for the 5 boards with increased heat tolerance. A solid no-go for me, at least for now.
This is amazing thing that you have presented in this video....
No need to have hot air gun & costly soldering stations...
Please make detailed video about this topic... & please share problems & difficulties that you have faced during this process. So that anyone could easily overcome through it....
So simple yet so functional that I'm astonished how I've never heard of this method before, and I also never thought about. Great concept!
7:52 the stench of burnt PCBs is incredible. I had a similar issue once, and it took me almost a week until the smell went away.
Apart from this: cool idea!
For sure you were not the first person thinking of this idea, but it was the first time I saw it, and I'm currently struggling with a lack of the necessary soldering equipment for my project =]
Anyway, this video "plus" the open-source content "plus" the generous share of the advices for minimizing failures, made me so excited for finding you in my UA-cam feed! Please keep up with this amazing work you are doing!😍💚
Man I love it when people do such smart things that seem like it could have been invented eons ago...
Love to have found your channel man.
Be well and happy,
Peter Lunk
This could be a feature on extreme overclocking motherboards, so when overclocking with sub zero temperatures, the board could keep itself warm to prevent condensation forming.
I’m commenting to congratulate you for being wonderfully original. No way there are no practical applications for self soldering pcb, i can essentially skip any type of traditional soldering and just lay out the components and then heat it using itself. Blown. Away.
I don't know how you come up with ideas, and frankly, at this point, I'm too afraid to ask. This is amazing. Well done, once again!
I mentioned the idea in the comments of a GreatScott video about reflow PCB heaters a couple of months ago. No idea if this guy saw it though.
Fantastic video. I especially liked seeing the mistakes, fixes, and the failure case where the PCB temperature is exceeded.
This idea is phenomenal. Not only did you build a reflow controller it self-propagates.
I just bought a Miniware MHP30 for reflowing small PCBs but it doesn't have the same ability to program a heat curve. With your design I can slightly redesign my existing circuits and guarantee better outcomes.
You're extremely gifted. Keep it up!
6:43
Reminds me about the concept of universal assembler; while on a different scale, it's a step in the same direction. Very cool!
this is really cool. like people have said, this could be really really good for kits. i'm thinking a kit which mixes smd and through hole components and uses this for the smd parts and lets people solder the through hole components by hand would be quite fun as an intermediate soldering teaching kit, to teach people about the smd soldering heat curves in a way that is fun and interactive but doesn't require previous experience with smd stuff
what a cool idea. I'm also thinking about this as a tool for repair, like reflowing the solder of the components when the device is dropped
Dropped? How about when it's overused or drilled by a competing bot?
Coming from 3D Printers, and being fascinated by PCB designs, i'm confident to say that this idea has earned you a place in maker history. 🥳
This is one of the coolest things I’ve seen. Absolutely brilliant.
This is such a cool idea it's borderline art
Congratulations you just made history.
Great Idea. Only problem could be EMI and SI issues, because the gnd plane should not have any cuts, in order to always have a good return path for the signals on the adjacent layers. But for all non high speed designs, this should be no issue. Very innovative, keep up the great work.
Very impressive!! Love the idea so I am going to use this idea to heat up a pcb in very cold enviroments.. I have a design for a car that does not fully boot when power on in below - 4 degrees.. This could be utilized as internal heater for components and the enclosure. Adding a fuse and a simple controller dedicated for this purpose would do nicely I think. Thanks for sharing! 👍👍
This indeed is an innovative idea! I'd only be worried by the interrupted current paths on this kind of ground plane. Maybe adding more layers to include a solid ground plane and leave the heating layer only for that purpose, could make this even better!
Love your projects Carl!
Yep, signal integrity on those PCBs won’t be good. Better only use it for low-speed analog stuff and so.
A meandering, resistive ground circuit might also lead to ground loops, which would not be fun to debug
@@abeta201 If it is sandwiched between ground and power plane i don't see why leaving it as a heater or totally disconnected would be a problem...
@@TheOriginalEviltech It might cause the device layer to pick up RFI, as mentioned in other comments, by virtue of being underneath it, but the video implies that it would be used as *the* gnd plane, which would risk ground loops.
@@abeta201 How else would you implement #rst_doublefree ?
I love that you didn't need to change anything in the manufacture part so the price is the same.
Just adding more functionality to something that is already there.
If it were more expensive or required certain types of pcb it would be a harder decision, but as it is, i can't ask for more
Great way to innovate: you take something that already exists for one purpose (Eg.: heated bed of 3D printers) and use the same principle in a different application, adding the microcontroller bit to make it awesome and self-replicating. Makers community needs more people like you.
This is awesome. I immediately thought that this system would be useful in a space exploration sense.
Great idea and a very nicely put together video 👍
Really cool idea though! One major downside to using this as a primary gnd plane though, would be that youre likely creating nasty inductances that can show up as ringing along your supply lines if there are many pulsing currents on this board. If you have enough board layers, it would be useful to have a primary solid gnd plane and this be the chopped up secondary plane. Great work!
Super great idea! Amazing that it works so nicely! Opens up a whole range of new possibilities! Might even work on double sided PCB's for small parts!
Dude that is the best idea ever. A self soldering circuit board seems so much smarter than blowing surface mount components all over the bench with a hot air gun. And you get the added bonus of the earth plane when your finished
I used thermal paste with low melting point I am surprised that you don't talk about the downsides !
1° High price because of antimoine or bismuth content
2° low thermal >>> and very low electrical
My experience with ChipQuick-s low-temp paste is "beads, beads under components, beads beside component, beads between pads, beads everywhere".
This is the kind of guy that invent things that we dream of inventing when we become engineer.
Just brilliant idea. You are so creative !
I could see this being useful for increasing longevity. I don't actually know enough about circuitry to know what I'm saying, but I've also fixed a few DoA electronics by reflowing in the oven. I just imagine having that built into the boars would be amazing
Brilliant work. The heating function could be very useful if the pcb needs to operate in extremely low temperatures. A possible application in Mars rovers that have to keep their circuits at the correct operating temperature.
Thanks! Yeap that's definitely another possible application for this - not sure if they are already using thermal pcb tracks for the Mars rover etc
@@CarlBugeja i have a light switch from the 70s with thermal pcb tracks. no idea why it needs to stay warm but it does.
i took it from an abandoned canadian prisoner of war camp
hahaha this is really amazing to watch, as soon as you explained your idea I thought damn why did I never thought about this. Very smart and intuitive that the PCBA your testing it with is rightaway the controller board for the project
I can definitely see this technique being standardized for production processes. It seems it would lower costs and maintain or improve reliability on soldering stages of a production.
This is the coolest thing I've seen in a while!!! So much creativity and professional knowledge went in, and i absolutely applaud you :))
I just love electrical engineering. everyday I see another wonderful person having a crazy personal project that could be the start of something really cool just itterating into it and putting out there for everyone to make something of it.
I can't wait to finish my degree so I can have time for personal projects again.
edit: also as someone who really really likes to make keyboards. this could be useful for the diodes. especially if someone is trying to make a lot of custom keyboards on the side to sell or really like to try a lot of different stuff. this is what i thought off. too bad I am terrible with circuits :P well no better way to learn than to fail I guess
SMH for ohmic heating in electric car accessories; works at -35°F, but inefficient. Add a 1 T layer, get heat pump functions?
Ah a fellow Maltese person, always great to see :) fantastic project! I love soldering my own PCBs but this would be such a revolution to doing things it's awesome dude!
Could definitely see this being used in entry level hobbyist kits for people who aren't too confident with soldering, especially considering surface mount pcbs ain't the most beginner friendly, also the possibility for bootstrapping using this system is promising too
Just use a cheap toaster oven for reflow then you don’t need to pay for additional layers, one run of several boards is actually more expensive. Bonus points for a thermometer.
Dude! Whilst I've enjoyed your videos for a long time; the production qulity, your passion, and clear enegineering talent. I've always been cynical of the applications. This, however, is awesome! Love it! Such an amazing idea.
Very nice idea(as always), could be used in not too complex projects and in general for open source kits. 👍
I can see a realistic use for this in cold weather applications after assembly. It would save on weight and complexity of the build. You literally cannot get closer to direct heating with this idea!
🤙
Great idea 🥳 Just wondering if the zig zag ground plane is not having impact on the HF spectrum...
RFSpice and the very many USB3_US and USB2_HS warnings.
Very very smart! Thank you for making it open to the world!
You should combine this with an in-situ thermal sensing pcb trace/pad so you don't even need to attach a sensor.
I am stupid but couldn't he measure the temperature using the resistance of the loop? It probably wouldnt be accurate but it should show something useful.
Whoa. I'm currently working on a LiPo battery discharging circuit - want a super cheap board I can make 50 of and use them to keep my LiPos at storage voltage, and the entire back layer is just a long trace to use as a power sink when discharging the battery.
This is literally the *perfect* idea for this project as I already have the trace, and need an easy way to solder up a large number of them!!
Crazy, just crazy. I never would have thought of that. This is brilliant.
Cool idea! I wonder if this would benefit a larger scale production? Great for kits, but what about a small business that make circuits by hand or have a small assembly line of employees? I would love to make guitar pedals & this might be the way to produce a small batch :)
I understood nothing but still the coolest video I've seen this week
Sometimes this may be quite useful if you want to reflow a connector, which does not tolerate the high temp well.
Inside PCB heating does not heat up the entire connector as much as a reflow oven would!
When reflowing you have to heat everything up to the same temperature otherwise you get thermal stressing of components and they fail (especially during the cooling phase). That is why there is a soak period in the reflow thermal profile! You need to understand the process and why it is what it is before cutting corners for the sake of simplicity!
@@simonbaxter8001 you do not have to heat everything up to the same temperature to eliminate thermal stressing, just the very rigid parts.
ICs are quite rigid, some MLCCs too. But the connectors I am thinking about, which need somewhat lower temps, only introduce low amount of mechanical stress, no matter how you solder them, because of the relatively long and flexible leads (they are designed not to add extra stress, because being connectors already adds external stress).
Lower profile parts would likely heat up better with inside PCB heating, compared to connectors. So using PCB heating does not seem an instant death sentence as you seem to imply.
You do have to rate each component for this process though, and doing an x-ray on every BGA part after reflow would be highly advisable to catch cracked balls!
@@adamrak7560 If you have to xray bga devices to check their integrity, you would use proper controlled manufacturing processes, not cheap untested self soldering PCBs!
This is pretty wild. It never seizes to amaze me what some people can come up with
Wow! This is such a cool idea. I wonder if it will have any long term effects on the lamination.
The temps were within specs so it shouldn't be a problem - is what I'm hoping.
@@Tasty_sand Yeah, should be good. Although I don't know how thermally conductive FR4 is. While the outer temp for the component layer might be inside spec, the inner heating layer might have to be a bit hotter.
But since hotplates work fine, I don't really think that this solution should be an issue.
Dude this is incredible. I love using it as the ground plane after. Genius.
Really interesting and innovating idea. Great job Carl!
Wow! 🤯 You took soldering PCBs to the next level.
This was incredible! What a brilliant idea.
Simply...Brilliant!!
Thanks you for sharing all your excellent ideas.
This is genius!! Always a treat to see your innovations. This would be a great workflow for prototyping boards that can self solder. Keep them coming!
@7:00 "So freaking Cool". yep, you said it. so cool. I can see this in a MI scenario, 'this circuit will self-destruct in 5 4 3 2 ...' gone. Or the last thing you do as you eject is fry the guidance systems.
I would think you'd want multiple attachment points of the soldering/groundplane to circuit ground, otherwise all ground current has to flow through that one resistor and it creates a large capacitor
Yeah, this probably won't work for higher frequency boards
This is a great idea! And what Joey Murphy said about satellites made me think, why not leave the thermal port intact? I can see that be relevant in the future for example if a pcb needs to be reflowed. In general that can be useful instead of having to put a pcb in the oven but in space even more to remotely reflow a pcb if it fails.
I don't mess around much with soldering and the like, but this was the coolest process to watch and see.
As others have said, I predict this could be the solution for kits with SMD components, especially as the paste seems to be sticky meaning that the placement can be done much easier, and not having to deal with learning the special technique of hot airing that can blow components away.
This is the most brilliant and innovative idea I've seen in ages. Great work.
Cool stuff :) I'm trying to think if this is actually useful... Definitely cool either way!
for small projects yeah. For big things that are mass-produced probably not
Quality like this, and refferences to proper margherita = Instant subscribe!
It would be interesting if you could use the resistance-temperature dependence of the trace to skip the external probe. The resistance of the copper should be quite linear up to well over 200℃ and have a temperature coefficient of about 3930ppm/℃ so going from 20℃ to 150℃ would increase the resistance with about 51%. You could either have Altium estimate the trace resistance, measure it yourself or if you have a temperature sensor on the controller PCB you could use that as the initial guess for the temperature of the PCB you want to solder
So innovative! Keep up the great work! You've quickly become one of my favorite youtube channels.
Nice idea, BUT you really need to appreciate the very real EMI, EMC, RF and signal cross talk problems that those heater traces are going to cause! Maybe you need to learn some PCB layout fundamentals and signal integrity knowledge before pushing this idea too much as your stepping into design integrity minefield here.
I'm pretty sure anyone who is making a PCB capable of self heating and including a USB-programmable MCU and thermocouple probably knows what a wire antenna is... Relax a little.
@@ryanchojnacki4421 Not sure they do. And I'm pretty sure all those that think it's a good idea on here are being taught a really bad idea. This needs a followup with some real electronic principles and a demonstration of the issues it causes. Definitely not one for the average maker, but definitely one for those that will do electronics as a career.
It is definitely going to be a problem when the circuit on top is operating at high speeds. The examples I have given are all operating at slow speeds, expect for the clock source which I still think it's safe considering it's very close to the mcu and there are other grounded polygon pours on other layers.. But I intended this idea to be used for maker projects, and not something that can pass esd/emc/emi .. Maybe I can explore that in the future but as you pointed out rf is not my field of expertise
@@CarlBugeja Thanks for the comment. Anything above 10Mhz which any digital signal will produce (edge frequencies in fast rise times), the micro itself if it has a PLL, usb is above 10MHz, any SPI bus to peripherals port displays, etc. Robert Ferenac demonstrates in his videos the effects of these issues and the frequencies at which they become an issue. You idea is pretty innovative, but in my experience it would cause anyone who takes up electronics, as a serious career or business, some headaches. But I would hope that before applying this idea to a design, that they would understand the implications and rule it out as a very bad idea!
Cockbite. See if anyone lets you implement waveplates in PCB layers going forward.
Magic smoke just got a completely new meaning!
I really like the idea. Might tinker with some of my designs soon.
That's so freakin cool Carl! Very creative!
THIS IS nothing short of what we all been thinking AND wanting for so long :) pure genius
Love this! Genuinely one of the most innovative channels on UA-cam.
What a weird, ingenious idea!
You're like a microelectronics artist!!
This is so much more interesting from a rework after a party fails perspective. I would love to have a selective heater under a bcg component or some 0201 resistors
the future of diy kits is gonna be amazing
That’s the kind of vendor sponsorship I wanna see, nice deal
What a great idea, just beautiful! Thank you so much for sharing and making this open source. How has noone ever thought of that before? Or maybe someone has, but didn't show off the results? Gotta try this at some point. It boggles my mind how far home brewn electronics have come along in the last years. I recall when the only option was to etch your own pcb and what kind of a mess that was. Nowadays you can just order multilayer boards with all kinds of fancy customization for pocket change, it's every hackers dream come true.
I suggested this in a GreatScott video comment a couple of months ago. No idea if this guy saw it though.
This is an excellent idea.. May be problematic for RF sensitive stuff but for 99% maker stuff this should work
Anything with a micro on it running over 10MHz has RF properties and is sensitive, especially if you have SPI and i2C busses on the board too!
This is such a clever idea! Thanks for sharing it and publishing it as open-source.
Incredible idea and design Carl, I'll test one!
Wow very cool. Good idea using the the thermal track as extra ground plane.
I like how at the end the pcb is charred and destroyed but the copper tracks inside look good as new, just looked cool.
Depending on how thin you can make those traces you might be able to design the ground plane like the mesh on a microwave, this would only work for square PCBs and has its own issues, but provides better grounding and wouldn't have issues with RF.
Just using a hotplate is still much easier and faster though.
I would definitely love to experiment with this.
With my projects this would definitely make it more interesting.
What a clever concept! You always bring us the best of electronics diy!!!