Mains Powered 3D Printer Heated Beds

Converting mains voltage down to 12 or 24VDC to drive a heating element makes no sense. To get 120 watts at 12 volts requires thick wires that can handle 10 amps, whereas at 120V, tiny 1A wires will do. If you’ve ever felt the MOSFET that switches your heated bed on and off, you know it’s working hard to pass that much current. [Makertum] is of the opinion this is a dumb idea. He’s creating a 110 / 230 V, mains-powered heated bed.

Creating a PCB heat bed isn’t an art – it’s a science. There are equations and variables to calculate, possibly some empirical measurements by measuring the resistance of a trace, but Ohm’s Law is a law for a reason. If you do things right, you can make a PCB heat bed perfectly suited for the task. You can even design in safety features like overcurrent protection and fuses. It can’t be that hard. After all, your house is full of devices that are plugged into the wall.

However, there’s a reason we use 12V and 24V heated beds – they give us, at the very least, the illusion of safety. Therefore, [Makertum] is looking for a few comments from specialists and people who know what they’re doing.

Although a mains powered heated bed sounds scary for a hobbyist-built 3D printer, there are a number of positives to the design. It would heat up faster, thin down a few parts, and significantly reduce the overall cost of the printer by not requiring another 100 Watts delivered from a 12V power supply. It’s a great idea if it doesn’t burn down the house. Anyone want to help?

160 thoughts on “Mains Powered 3D Printer Heated Beds

      1. It’s not voltage that kills you, it is current acoss the chest ie heart. Hence so called medical coils of the 19th century and high frequency tesla coils and drawing arcs to your skin that do not kill you. I knew a TV engineer that tested high voltage triplers for CRT TV’s by drawing an arc to his finger. The path is the issue and the current available.

        1. The first time I hooked up to HT it was a 32+ inch screen (about 27KV) and it launched me into the air and I landed 4 meters behind where I was sitting. The second time was a 27 inch (about 24KV) and there was a wall about two meters behind me and I was slammed into that so hard that I was bruised in many places.

          These were obviously mistakes. I would never intentionally come into contact with HT from the anode cap.

          Your friend reminds me of a domestic main voltage electrician that would test with the back of his hand. He’s dead now.

          Oh, and I have fixed about 15,000 of the old CRT TV’s so while my mistakes can’t ever be called good, they’re not without precedent.

          In the above, the first was the result of a fault condition making a sub part of the internal chassis float at HT voltage. The second was the result of a “friend” jokingly giving my shoulder a shove when I was working.

          1. Your friend died of what and why? Testing mains voltage across the back of his hand? Please. Mains voltage is not scary if you know what you are about. I dealt with volts up to 132KV and mega amps and megawatts in my career I worked in the UK electrical power industry.

          2. My friend died of electrocution while on the job. Our mains voltage here is 240V (single phase) or 415V (three phase) and not very forgiving.

            I to have worked with high voltage systems. My point was that it needs to be treated with a bit of respect or it WILL kill you even if not immediately. My friend was an electrician of many years. And sure, I hear lots of stories of how people get away with this and that, but that is just that they gotten away with it so far. They could be dead tomorrow.

            Most people here only deal with low voltages – domestic mains and below. Where I live even mains voltage is lethal.

            Then when you step up a notch to the KV stage the safety rules change because air can now be a conductor.

          3. Is it possible, that the discreet triplers (vs. diode split transformer) were in the time of BW-TV? They had lower power. Or the guy tested the AC before the tripler, I think that would make a great difference. Although I also would not do this intentionally. There would also be a nice arc to a grounded screw driver, if you want to test the HT – without any current path through myself, what I really prefer.
            The electrician, whom you mention, did he really die while testing for voltage with his hand, or was his accident unrelated to this?

          4. @ [Martin]

            I don’t know the details of his death apart from knowing it was electrocution and on the job.

            I expect the whatever the details, it was complacency.

            Although we bumped into each other on job sites, we had different employers. I knew his family and I wasn’t about to ask them about the details of course.

          5. 132KV is probably safer if you don’t test it with the back of your hand. It’s not like those little neon screwdrivers are expensive. You’d think in electrician training they’d have taught him about electrocution.

          6. Nigel, while the UK mains is nominally 230V, power stations and the grid haven’t changed at all.

            I can’t remember the exact numbers, but before the change, UK mains voltage was acceptable between something like 210V and 270V, depending on grid load. EU mains had a similar range. All they did, was tighten up the range. So now it’s nominally 230V, with 240V being within acceptable range. Which is good because the voltage stayed exactly the same. They just trimmed the range at the top end to be within the EU standards.

            I suppose a more flexible grid, with more cheap gas plants, which can fire up quickly, made this possible, as well as making the grid itself smarter. There was a rush to build gas power plants a while ago, since of course gas isn’t going to run out any time soon or anything.

            The Death Star post illustrates why we shouldn’t let economists decide how to run things. That and a million other stories from real life.

          7. Did you keep records so that you actually know you repaired 15,000 TVs? I don’t believe it, and if it is not true then I doubt your other statements on this evidence alone. People should classify their statements as to fact, fiction, or opinion, not just leave them in an ambiguous interpretive limbo.

          8. Back in the day the UK before EU harmonisation the nominal single phase domestic voltage was 240V ac and within a range of +10% and -6% ie 264V and 225.6 V Under EU harmonisation it is 230V +10% and -6% ie 256 V and 216.2 V. I can assure you that voltage control at Primary substations in the UK 33KV/11KV was lowered to the new nominal for the nominal voltage at secondary substations 11KV/415V/230V. A good thing really as it saves energy ;)

          9. In cases of energy shortfall, we could apply 3% or 6% voltage reduction from our central control, to shed load. (along with asking manufacturers to reduce load, part of their contract) Before real load shedding, switching off supplies. I know. I was a control engineer. Admittedly a few years ago :)

    1. You don’t need a 3D printer to use a heated print bed for 3D printing? I’d think that having the rest of the 3D printer would more or less be required to use a heated bed to facilitate printing.

  1. The main issue would be shock protection. A bare PCB with solder mask probably won’t cut it for UL or CE. If you have vias in the element traces, it will probably be really dodgy. On a commercial product which you can’t (are not supposed to) open and tinker with, there is probably a way to do this easily. For a DIY printer, not so much.

    1. ead of single layer traces covered by the paint, what about multilayer PCB? So the traces would be embedded between two FR4. This way it would be safer.

      I would prefer, use a circuit (SCR?) to lower the voltage to around a safer value, where we won’t have any problem in the worse case.

      1. The heater from the inner layer of the PCB still needs to exit to the surface using vias or plated through holes some how. So the contacts are still exposed.
        You might use a TRIAC in something like the phase control in a dimmer switch as a PWM to control heater power, but the peaks voltages at the output are still at line voltages levels.

        1. Meh, just use a relay for your “PWM”. It’s much easier, and you have a huge low pass filter. You can easily go up to a few Hz switching with a good realy. Use a solid state relay and it’s even better (you might break your mechanical relay after prolonged use).

      2. Continuing USWaterRockets: PCB could just have the solder mask on, mostly for oxidation/contamination protection, with the heating element trace a single top layer. Put a plate of borosilicate glass (maybe tempered?) tightly over it to spread heat and protect from shock, with overhang on the hazard side. If you want to get fancy, add an “active” indicator LED and a few thermal sensors on the bottom layer. Mind your PCB glass mat material and construction – make sure that the construction can handle the heat (a good [not great, not cheap] construction can handle ~170C before it reaches glass transition temperature)

        UL rule of thumb for shock hazard: 1.5mm / 60mils distance between a hazardous live conductor and something touchable for single insulation level protection (double that for double insulation).

        continuing akismetuser212339447″ Using a mechanical relay for your thermal control _may_ make you crazy with the clicking. If you think you can take the madness, keep an eye on the make/break current rating, and the lifetime switching at the design current. Switching current and voltage is often far less than carrying current, and electrical contact life is short compared to mechanical life (Switching at zero crossing will help). Also note that failure mode of relays is possibly contacts fused closed – design with safety in mind, and have some sort of feedback. Something like a thermal cutoff fuse?

        1. No edit, so addendum: glancing at comments below, one thing to note: go take apart a toaster and examine the safeties in place there. That may give you some insight into commercial (safety certified, most likely) implementation of a similar technology in the consumer market.

        2. I would also use a solid state relay. And I would think about gluing the glass plate with high temp silicone to the PCB. This vastly improves thermal contact and so heat transfer which reduces peak temp and thermal lag in the control loop. Also it prevents exposed live wires even in case the glass breaks.

          But I don’t like the mains powered heater in this case:
          If you take 120W @24V its just 5A. No problem for a switching transistor or wires with 0,5mm² and you keep the mains voltage in the PSU box. Which I prefer in this case. Mains rated cabling (230V here in Europe) are not thinner and need thicker insulation. In the worst case you have to have PE connection to any metal part of the construction.

        3. You don’t need a big contactor to switch 1A at 120v, you can get little “ice cube” relays that switch up to 10A at 120V and they have a plug-in base so easy to change out when they inevitably fail. Most of them, in new perfect condition, make barely any sound. Nothing compared to the noise of fans and stepper motors. The only issue I can see is running AC mains voltage through wiring that is going in parallel with low voltage. Would need to use thick jacketed, shielded cable to prevent too much EMI. Maybe it’d be better suited using a separate transformer just for the heated bed, rectified/regulated but still a higher voltage for better efficiency. Speaking just from my experience as an electrician and hobbyist level electronics/3d printing experience.

        1. Anything below 50 Volts is considered “safe”. As in, you don’t need to be a licensed electrician to build and operate it. That’s the other thing with dealing with mains voltages in a DIY project. If you build it yourself, your insurance company won’t covery you.

          You can use mains voltage if it involves already approved/CE-licensed parts that are simply wired together such as a power supply to a circuit board, but if you design and build your own part with 120 V inside, it’s on your head when the house burns down.

      1. There are many issues around AC safety.

        1) the frequency (50-60 Hz) tends to make muscles spasm repeatedly (DC spasms and stays there)
        2) the frequency can interfere with cardiac electrochemical signalling (your heart goes into fibrillation)
        3) the current can pass through capacitance (shielding) and via inductance
        4) the current can pass through ground return and other single-wire, open circuit States

        Overall though, you’re just talking about increased RMS voltage allowing more current; the more current the more burn, heart arrhythmia and neuromuscular effects.

        P.S. Human body (skin contact) is 1-3 MOhm, and 100mA is well within lethal range for AC appliances.

          1. Well your musle will clench with AC as well. That’s why friend mine is still among the living. He was an oilfield pumper after n over night thunderstorm there was much equip down because of blown fuses and tripped motor over load protection devices my fried got an a hurry and got careless He reached int a motor control box and grab a fuse in each hand to pull them out. Because of other damage in the box they where energized. The jolt threw him on his ass because his hands clenched surround the fuses the act of throwing him back disconnect him from the electric power as the fuses pulled out out of their holders. I tech school one of the other student got bit by HV inside a TV and it threw his hand way. He wasn’t hurt by the electrical shock, but we kept an an eye him for a while because he looked like he could enter a state of physical shock.

          2. All of this, if you suppose that you are taking a full hand of cable. Got shocked a few times with AC (and DC as well) and since I was just thouching, not handling, I had no trouble to get away with it (just swearing A LOT and throwing my screwdriver through the room).

            Funny story : one of my friend had a real electrical problem in his house : earth not well connected or whatever. You could get shocked by touching an USB shield while bare foot. Handling something without a case like a raspi was a living hell.

          3. AC is safer than DC. You can let go of two AC terminals but not DC as DC keeps your muscles clenched. AC doesn’t.

            1) the frequency (50-60 Hz) tends to make muscles spasm repeatedly (DC spasms and stays there)

            No, this doesn’t necessarily make it safer. This makes it different.

            Being able to “let go” doesn’t help when your insides are burnt and your heart has ceased meaningful function.

            Remember that, across the same resistance, power goes up with the square of voltage; 120v RMS can deliver 100x the wattage of a 12v supply, and 240v delivers 400x.

            I’m cynical that the non-tetany behaviour of AC results in being able to pull away. Remember that mains is oscillating at >50 x a second, so the pull-away would have to be within that cycle, as well as overwhelming the hyperpolarisation and ignoring the fact that normal neuromuscular signals are essentially pulse-frequency modulated with saturation causing desensitivity (I.e. A continuous signal is ignored)

            No, it’s far more likely a combination of the strength of your flexor muscles (biceps) being stronger than your extensors (triceps), a bit of spinal reflex (sudden stretch of pain signal = withdraw limb) and observer bias; you’ve probably seen people electrocuted with 120v AC more often than 120v DC.

            At the end of the day all power supplies deserve respect, not least because they’re more likely to kill you by burning your house down than through electrocution.

          4. “AC is safer than DC. You can let go of two AC terminals but not DC as DC keeps your muscles clenched. AC doesn’t. ”

            Completely wrong and dangerously so!

            It’s the other way around. Muscles and nerves respond to di/dt and not constant current. You can let go of DC but not AC because the continuous pulsing is how the nerves signal the muscle to contract.

        1. “(DC spasms and stays there)”

          Doesn’t.

          DC burns. It doesn’t cause a muscle lock-up like AC does. It’s more dangerous in that sense, because you don’t feel it beyond the initial shock until your flesh starts to cook from inside out.

          1. I am glad, that finally someone posts this, because this is also my information about dangers of electricity. I just have not been sure enough, if I had missed something.
            I also heard, that a DC accident can kill you hours later even if you are not burnt that heavily, because DC is able to cause electrolysis in your body which upsets the delicate balance of electrolytes in your blood and tissue.

          2. @ [Martin] DC can kill you hours later because it goes through your body like lightning, it takes the path of the lowest resistance (usually the most moist path) and cooks everything in it’s path including vital organs.

          3. idk, anytime I’ve been shocked on the job I tend to immediately pull back – usually any screwdriver I have in my hands goes flying, and hopefully its not towards me or anyone else’s face. Which is why I use all 1000v insulated screwdrivers now. Better sorry than safe, kill two stones with one bird. You can definitely “feel” the alternating current.

        2. I was mainly thinking about the diff between high vs. low RMS AC voltage but the discussion of AC vs DC has been interesting as well.

          A great point below about wattage increasing exponentially with higher voltage the old P = (V^2)/R.

          Overall though, if everything is designed so it can’t hurt you, even in far fetched and unlikely worst case scenarios then many things would just never get made. In this case I think you could make a safe heated bed as long as you take safeguards to ensure a low probability of accidental shock. A simple enclosure around the board and the connections would more or less end the shock safety debate. Fire is another issue I guess, probably a fuse would solve that.

      2. But my toaster uses heating elements contacted to 120 VAC, and so no barriers against ignorant ding stupid things. I agree the lower voltage is use because of the illusion of safet is probably why they use the lower voltages. Only an illusion to those who haven had the opportunity to study electricity or had made the effort to study it.

        1. > those who haven had the opportunity to study electricity or had made the effort to study it.
          Can’t let Darwin wipe them out when they happens to the target market… Even worse is that there are people that want to build printers on their own, but are not qualified to wire stuff to line voltage safely.

          I suppose if you want, you can stick your fingers into the toaster slot to touch the exposed element and push down the slider or try to make toast while taking a bath. Otherwise, you are not making any contacts with the high voltage.

          1. I guess you missed my writing ignorant people doing stupid things People using a metal utensil to clear a jammed toaster isn’t unheard of but still there are no barriers to that. Perhaps there are bread toaster that use an enclosed heating element like many toaster ovens use, but my fairly new UL labeled toaster uses the same style of heating element hat been used for decades. That what you quoted was directed the the DIY individual more likely to come across something like this, not average John or Jane Doe.

          2. People that build or use 3D printers aren’t necessarily qualified electricians or know how to handle high voltage safely. That’s what I mean for ” Can’t let Darwin wipe them out when they happens to the target market… ”

            There are people that play with high voltage stuff like Tesla coils, induction heaters etc. Now those know enough to respect high voltage.

        2. Toasters kindof have to use mains voltage, there’s a kilowatt or so of power in there. They do their best to keep the mains voltage away from hands, you’d have to deliberately put your fingers in to get shocked. Always remember to switch off at the wall before you stick the knife in to dig out your crumpets!

          This is a heater that’s going to be near fingers, or elbows, and is completely accessible to them, completely easy to touch. One day, when he’s distracted thinking about something else (not rare), he’ll end up touching the heater and getting a shock.

          This is very dangerous. Safety regulations are all about keeping mains voltage away from people, inside casings that are earthed or double-insulated. This has the voltage just lying there. Possibly covered with a bit of silk-screened paint. It’s not guaranteed to kill someone, but for the extra money I’d sooner have a non-deadly heater for a 3D printer.

      3. https://en.wikipedia.org/wiki/Extra-low_voltage
        >Extra-low voltage (ELV), in electricity supply, is one of several means to protect against electrical shock. The International Electrotechnical Commission and its member organizations define an ELV circuit as one in which the electrical potential of any conductor against earth (ground) is not more than either 25 volts RMS (35 volts peak) for alternating current[inconsistent], or ripple-free 60 volts for direct current under dry conditions.

        It is not an illusion when it is a recognized safety standard.

        1. Point taken. Perhaps it’s time to change of the the mindset, of the at large DIY community to when treated with properly 120/240 VAC can be used safely, as it’s being used day in and say out. I recall a project of Limor Fried made a project more as need be be becase she appered to be overly frearful of line voltage, then again she was taking into consideration the carelessness in the hacker community. One the flip side I personally know electrical engineers with decade in the filed who flippantly say 120 VAC can’t kill you. I know enough to know they are being full of shit In the end because of irritation fear useful resource is being ignored. Then again I have history of being comfortable with working on motor controls with neatly 800 VAC feeding them or working with electronic devices that have more than 30,0000 volts inside the cabinet. Lower voltages bring greater safy to life burt rquire higher current to do the amount of work, potetiall a fire hazrds id something goes wrong.

          1. Not really clear but good point. I met an electrician who told me : ‘230VAC is not dangerous, I’m getting shocked a lot, it really hurts when its 400VAC. Oh by the way, I have to get an appointment for my heart !’

            My testimony : 230VAC doesn’t kill if you’re lucky. Lots of people died on 230VAC (Claude François….)

          2. You completely missed the point. The safety regulatory doesn’t change regardless of the *perception* or the *knowledge* of the person using it. If you are selling a product or publishing a design, you may ended up being liable for the end users. Even if you are the person that design and build your own stuff, plugging in something not certified pretty much voids your home insurance if it happens to be the cause of a fire etc. On the other hand if you are playing with ELV from UL etc approved/certified supplies, you are covered.

            Yes you can use those high voltage safely, BUT your typical DIY people don’t even bother to read or understand datasheets. These people need the proper training, but aren’t likely to be bother to go beyond the typical monkey see monkey do watching youtube and think they know it all. I have seen enough of bad layouts of the average projects that doesn’t even understand that the creepage space between the primary and secondary side of an optoisolator aren’t meant to be for routing!

  2. The reason may be the cost of getting UL approval. You can buy off-the-shelf, UL approved, 12/24VDC power supplies. Once you are working with 12/24VDC most products don’t need further UL approvals. But — if everything is mains power you are going to have to pay for a new round of UL testing. That can be $10,000 or more.

  3. It would require a molded socket attached to an edge with a standard IEC polarized C7 connector. No exposed conductors anywhere. Could embed a green light for powered up and a red one for hot into the socket molding.

    1. That’ what I was thing one simple resign feature could make it as safe as most electrical appliances. Assuming heating element isn’t subject to contact. For me the phonographs weren’t clear in that regard.
      .

  4. Why not model it after ceramic space heaters, waffle irons or my popcorn maker?

    They all have a ceramic rod or coil wrapped with nichrome wire that turn red hot. A vaccum forming table design I looked at used ceramic rods made the same way with reflectors to evenly distribute the heat. I believe all of the above (I did not verify if the table was) were UL marked.

    I also own an electric kiln designed to heat and temper glass so a page can be taken from that design. I haven’t made the effort to investigate the heating elements as of yet.

    Point is, there are already pretty well established and widely accepted methods for uniform heating. Why re-invent the wheel AND sacrifice safety?

    1. Yup. Isolation is important as other commenters have noted. I’d compare electric heating pad design too, and probably require a door interlock to cut power to the bed when the door is open (enclosed only). A simpler way may be to simply put a transformer or “dirty” switcher under the bed and not put the higher voltage into the build plate in the first place. Use either a digital on/off for the switcher or a ZVS SSR for the transformer high side control. Would a 1/100 or 1/120 of a second minimum duty cycle and lsb affect the temperature that much, given the thermal mass of the system?

  5. I do not recommend diy 120/240 volt AC in this application. While you are correct that the higher voltage allows lower current draw, you are not proposing limiting the current, which can be very dangerous. If your hands were wet with sweat, you would act as a lower resistance path to any available ground you might be touching. In much the same way a hot dog will cook with 120 AC poet applied to each end, you will do considerable damage to your skin, muscle, and fat tissues along the current path. Use DC power, a larger wire, and maintain the safety margin. One trip to the ER will buy many hundreds of meters of wire and plenty of DC power supplies. Oh, and by the way, I’m an electrical engineer.

    1. I neither would recommend the mains power approach (230V in Europe), the power of 120W is very well to be handled with 24V, even 36V is necessary. This 3D printers are in many cases still some kind of “hacker toy”, often are modifications or experiments done.

      Of course a safe mains powered heater is possible, but you need a double insulated cable (which is stiffer and heavier) and a reliable strain relief for it. You have to permanently bond the heater PCB to the glass print bed with silicone. You need a fuse and a thermal cut out (thermal fuse) for fire safety, because the available power in the mains circuit is much higher.

  6. I did this to my 3D printer quite awhile ago. I made my printer bed from 1/2 inch thick aluminum for thermal mass. I went to walmart and bought a $6 clothes iron. The iron had a cast aluminum foot with the integral ceramic heating element. I took the iron apart and cut the element out from the foot with a bandsaw. I applied some thermal grease to the cutout heating element and screwed it to the underside of the thick build plate. I attached the thermister sensor a ways away ftom the heating element so i would read the build plate rather than the element. i used a solid state relay for control. For safety i ran a ground wire to the build plate and fused the mains. It works great!

    1. Now that’s something :D… And for other commenters: of course exposed mains voltage is dangerous, lethal even. But why would it need to be exposed? It could be made like Davic’s hack, or in countless other ways that are safe. Hide heating element under thermal mass, use secure connectors and proper grounding.

    2. At least you thought about fuse and earth ground wire. Hopefully you also kept the thermal fuse of the iron – normally they have one to prevent fire in case of a welded thermostat contact (or short circuit in the solid state relay).
      And hopefully you kept or built a good case for all this (the open part of the iron).

    1. No gain at all!

      Of course you can use an extra 50Hz transformer, you do not need voltage regulation. But you need power (DC) for motors and electronics. If you have already a switching PSU (which are often even cheaper than big heavy iron/copper transformers of the same power) the extra power capability for the heater does not cost very much. Lets say you will not get the 120VA transformer for the price difference of lets say a 80W PSU vs. a 200W PSU.
      Also you can switch the DC with a single MOSFET.

  7. Why not use a silicone pad block heater for cars? You glue them on, they run on 120v and work good. It’s robust enough to sit on the bottom of your engine, in elements and gravel spray from the road, for years. They even come with a little tube of aluminum adhesive. Seems like a pretty good fit.

    http://www.canadiantire.ca/en/automotive/interior-auto-accessories/winter-accessories/automotive-accessories/polar-pad-engine-heater-125w-0300004p.html

    I mean really this shouldn’t be a big deal. Resistance heating that you plug into the wall has been done in millions and millions of devices for basically a century. When you really think about it, a toaster is probably WAY more dangerous than what is being proposed here.

    I’ve thought about converting my Solidoodle 2 to 120v bed heating for ages now. I think it’s a good idea with a lot of benefits. I’ve been far to lazy to actually do it, however. Maybe one day I’ll get my act together. In the mean time I’ll just sit here and envy other people who have had the initiative to actually do it.

  8. two suggestions…
    1. bridge it and use as a 230v rectified/ cleaned up DC as per your choice
    2. as with pop up toasters and hairdriers, use convection and radiation instead of conduction. use a hot coil and duct the hot air for circulation under the bed.

    1. 230 volt DC is deadly and needs to be avoided if possible.

      On my printer I closed the sides, took the heated bed off and fitted a relay on the driving wires connected to a travel (230 volt AC) hair dryer screwed to the back. ABS prints like PLA now, no warping or other problems other than the normal ABS shrinkage factor.

  9. “To get 120 watts at 12 volts requires thick wires that can handle 10 amps, whereas at 120V, tiny 1A wires will do”.

    This makes no sense when you’re designing a heater. The higher the current, the hotter the wire will be. That’s why power companies transmit power at high voltage in the first place: they want to minimize heating. 120 watts flowing through a wire does NOT mean that 120 watts of heat are dissipated in it. You’ll get a much more efficient heater at lower voltages and higher currents.

    1. You may be confusing voltage drop ALONG a power line with the voltage BETWEEN lines. Maximizing the latter will transmit same power with less current, minimizing the first and hence Ohm’s Law power loss through transmission.

      1. He’s saying heating is a function of current, not voltage, or even overall watts. Which is true.

        Of course in this case, there’s less heat per length of track, but the tracks are longer. So it cancels itself out in the end.

        It’s still dangerous though. Cost of a 12V power supply vs a small chance of killing myself, I’d spend the money. 3D printers aren’t cheap anyway, why economise on safety?

        1. Heating is a function of power loss, which varies by the square of the voltage drop and inversely with the resistance. Current is a fallout of the voltage drop across the resistance. While typically stated as I*I*R losses, getting that “I” to be a big number with a low input voltage means that R has to also be low, which decreases power. Requiring a low R for the heating element magnifies the power losses in the rest of the circuit, where heat isn’t desired.

          tl;dr Higher voltage delivers more heat where it is desired and less where it shouldn’t be.

    2. You are not understanding something correctly with these statements. You do not want a “hot wire” until you get to where the heat is needed so you have to use thick wires if the voltage is lower between your source and destination if you use a lower voltage.

  10. With 12/24V, the resistance is low enough that it can be done using copper traces on PCB. P=V^2/R, so the resistance needed to be increase by 100 times for 12 to 120V. The traces gets a lot thinner and longer.

    The control circuit for the 12V/24V can be as simple as a MOSFET and don’t even need to be isolated if you know what you are doing. Because you are using a regulated supply, the same heated bed can be used at different countries with different voltages. The safety regulatory stuff get you covered as 12/24V are considered to be low voltage.

  11. This is a fantastic idea: a regular PCB, that gummy conformal coat over any solder joints, solder mask is fine for insulation if you space your traces correctly, and an SSR to turn it on or off. If you have a problem with the thing getting too hot, set a maximum with a variac on the input and you’ll get slower ramps at lower voltage but lower maxima. This is not unsafe, I would think it a fine idea, and I may have a different bed material in mind too.

    1. Agreed. I work in electronics as a profession. High voltage electronics no less. Industrial size high voltage power supplies. Everything about them up to maintaining them down to component level. The comments here from people are scaring me shitless from the lack of knowledge to the over the top over complication of the problem.

      Electricity can and will kill you. dead. it doesn’t care if it’s dumping into you from an AC or DC source or at what frequency. A little tickle to one person may be the correct amount at the right time to drop another who has a heart defect they didn’t and will never know they had. “Freezing on circuit” and all the other bullshit people wax on about is meaningless when you are dead. And electricity will grab you and tear out your soul before you have a chance to fucking realize you are dead. If you honestly don’t know what you are doing, then shut the fuck up and stay the hell back. A dead “know it all” knows nothing. Turn off your fucking ego because it’s going to get you killed. Period. Use common sense to keep yourself alive, if you have none, stay the fuck away from electricity.

      That said. We routinely use run of the mill power resistors in the aluminum heatsinked package, you know the ones, they start with the letter “D”, as heaters. Ohms law doesn’t fucking care if it’s DC or AC, but a little RMS will explain that DC will do the same amount of work at a lower voltage. It doesn’t fucking matter because you can choose another resistor value and get the same thing with AC. A cheap temperature controller and a relay will suffice if you have enough thermal mass to average the temperature.

      They make silicone pad heaters for automotive stuff like someone else mentioned, that would probably work too.

      They make “plate” heaters that are metal plate looking thing with a couple of stud terminals sticking out of them as well.

      This is a rather simplistic problem that a little gooogling around would solve in about 5 minutes.

        1. Usually called a Residual Current Device (RCD) or Earth Leakage Circuit Breaker (ELCB). Usually set on 30mA, but they take a small time to trip, which is dependent on how much current is leaking into the poor unfortunate 3D printerer. Much better than nothing, but not 100% guaranteed to save your life. I’d sooner leave mains voltage alone for things I’m tinkering with. I only mess with the mains for simple jobs, and then I’m very cautious.

          Once got electrocuted from a computer monitor though, that a relative thoughfully gave us without a case. I was reaching underneath to adjust the brightness or something. Touched something live. The monitor was on the floor, with the 8-bit computer (a Spectrum). My hand touched the live PCB, jerked away, bounced off the floor, and back up to the PCB again. Eventually my sister switched it off at the socket. Bit stupid really, I wouldn’t let kids near something like that, but…

          You can get plugin RCDs. For the cost, you could probably just get a bigger 12V power supply.

  12. No. Just no.

    Creepage to the mounting screws is one issue and that fuse crowed by traces at mains potential is just bad design.

    I’ve never used a 3D printer but I imagine the printer nozzle is conductive – a small misalignment, the nozzle breaks through the solder-mask, your whole printer then becomes live, blowing up the PC connected to it at the same time.

    Just. No.

    1. The printer does not print onto the heater. The heater is normally under the bed which can be a variety of materials. I use polycarbonate, a lot of people use glass, I think most printers come with aluminium with Kapton tape on top. The heater for the bed is usually safely tucked away where you would need to be somewhat of a contortionist to actually touch it.

      1. Kapton tape is nice but I know it peels and needs replacing eventually. You wouldn’t legally be able to sell something using just tape as a mains insulator. Glass can break. It’s a small chance, but risk should be proportionate to the danger when things fail. You might sometimes have metal screws or whatnot on the bed if you’re assembling a couple of pieces there. Best not to mess.

  13. Guys, checkout Alirubber.

    http://www.aliexpress.com/store/product/110v-220v-Silicone-Rubber-Heater-Bed-600MM-x-600MM-24-X24-For-3D-Printer-With-100k/709519_1847472742.html

    They made me a 110v silicone rubber heat bed with adhesive backing at the size I wanted with 2 integrated thermistors and appropriate cabling for like $60 delivered to my door. Add a 110v relay and you’re good to go. The Eustathios and Herculien builders group have been using these for a while.

  14. My fairly large (print volume 450x250x350mm) home-made 3D printer uses a mains-powered heat bed made from a junk-pile sourced electric pancake griddle which I ground down to more or less flat and glued a 24×12″ ceramic floor tile down to its top with high temperature silicone. It has a 10 ohm resistance and I use the original temperature control knob as a safety / sanity interlock against thermal runaway and then it is driven by 120v mains power switched by a 25A solid state relay controlled by the normal RAMPS heatbed output. The mains input runs through a breaker box before going to the 24v power supply for the NEMA23 motors and the ATX power supply for the Arduino, RAMPS, extruders, hotends, and Odroid U3 that make up the rest of the printer.

  15. I’ve built a 24 V heated bed for my DIY (as I call it) Mendel 90 XXL printer (400 x 400 mm print area). I use a 24 V 400 W LED power source and – since I already had a 12 V RAMPS – put a 24->12 V DC-DC converter in line with the RAMPS and driving a 24 V heating mat (for cars) via a DC solid state relay. As long as it’s not too cold in the room I get up to 60°C which is helpful for PLA but of course not sufficient for ABS (but I don’t want to breathe these ABS fumes anyway)…
    I’d rather use a thick cable (in combination with an energy chain) than playing with mains voltage.

  16. If you use 24V for 120W, you only need 5 amps. That’s not much. A good mosfet shouldn’t even get warm at 5A, and you don’t need very fat wires.

    And you could still move to 48V if you needed more power while staying safe.

  17. Well, no mention of some things that I thought you would be discussing –

    “Normal” mains voltage heating elements use nichrome (nickel chromium) some time called “resistance wire” because it has a higher resistance for a given surface area.

    Nichrome expands a lot when it’s heated which is why it’s often in spirals or loops. Copper also expands significantly when it heats so what ‘relief’ could there be for this expansion – and at a high enough temperature the copper traces would de-laminate from the FR4.

    The other issues aren’t as significant. The copper can be the inner layer like you see in multi layer (3+) boards. The only exposed part can be for the connector and the connector can be made water resistant and closed off.

    But can this actually be done with the expansion and de lamination issues.

    1. You are right, FR-4 has a glass transition temperature of around 120°C, so if you heat up your traces above that, and take into account the mechanical stress from expanding copper traces, delamination is something waiting to happen sooner or later (especially if you design corners in your traces). Add to that the fact that you usually don’t measure the temperature directly on the traces to regulate input power, you could go way beyond 120°C in your copper traces. I don’t think a lot of DIY guys actually know that there would be high-TG materials on the market better suited for PCB based heater applications…

    1. Good call, light bulbs and heating elements don’t need perfect smooth precise 12vDC, as long as it’s in the ball park you’re good.

      That said, why not use 12v halogens (or even mains voltage halogens) to provide the heat? They are ~99% efficient at that, cheap, plentiful, universally available…

      1. Halogen lamps hate being run at substantially less than full power! The additive (bromine or iodine) is used to keep the inside of the bulb clean and regenerate the filament, but it corrodes the filament if it’s not hot enough.

        1. This is true in a certain temperature range, I think between 70 and 90% of their nominal power. And in spite of that there are so many lamps with halogen bulbs, which are fitted with dimmer circuits. And you know why? Because the bulbs are very cheap, so nobody cares :-)
          And they are often used as heating elements: in a professional reflow oven (3000,- euro class) and in laser printers.
          But in the end I think for the heated print bed you do NOT want a radiative heat source, it complicates the things

    1. Yeah really though. It’s kind of embarrassing. I’m an engineer and work in big industry (usually minerals, bulk material handling, oil and gas, doing a big robotics project right now for a wood products plant, etc.) and this would be an afterthought. No one would even question it. I mean really a no brainer. People use electric blankets on their bed. I really don’t get it. It really shows how amateur hour most of the people who comment on this site are.

      Actually what they WOULD question is why you’d throw a big power brick on something if you’re just doing resistance heating. Bizarre…

      1. Electric blankets have lots of testing to ensure they’re safely insulated from the user. They also don’t get nearly as hot. They also use too much power to economically run from a low voltage PSU.

        This plate is too likely to touch the user. It’s sensible to have a certain fear of mains voltages, makes you pay attention when you’re working with it. The DC power suggested would come from the printer’s existing PSU, pay a little bit extra for a few more amps, the overall wattage isn’t high. Much less than I presume you’d use in industry. Thanks to power-hungry graphics cards, high-power PC PSUs are cheap.

        It’s just sensible to avoid risks where you can. If it doesn’t cost much extra, why not bother? Tinkering with a 3D printer is likely to take concentration away from other areas. Low voltage is intrinsically safe from electrocution. Don’t make things dangerous for yourself.

        1. The point is it can be done safely and cheaply and is proven to be safe; so we can’t work out why everyone was so dubious.

          I’m a bit disappointed, I thought this was a safe place for dangerous ideas. If you want to diy an ELV heat bed go ahead, but that isn’t going to suit everyone.

          1. The risk / reward ratio just doesn’t seem worth it to me, when it’s so easy to have an intrinsically safe low-voltage one.

            I approve of much greater risks if it’s something awesome. But for this, why bother?

  18. Mains powered heated glass plates are common place in catering, same deal as a car back window, the heating wires are on the other side from the “working surface” these things can tolerate touching and even fluid spills, why reinvent the wheel, the solution is already commercially avaliable.

  19. Why reinvent what is already on the market? There are heating panels beeing sold as accessory for 3d printers already. I just saw a 300x200mm 200W 230VAC silicone heating pad for around 90$. The vendor (multec.de) however clearly states that the installation has to be done by a certified electrician. I’m pretty shure these pads could be sourced a lot cheaper on other sources too. If you think you can just DIY around with 230VAC, make a mistake and hurt/kill somebody or burn down a house, you will be f”cked for life, because no insurance company will cover you. And if you kill yourself and have a family, well, thats not that great for them either. Keep that in mind.
    I’d rather use a bed made for 48VDC. Powersupplys are abundant for that voltage (standard for a lot of telecom/industrial equipment), reasonably safe to work with compared to 230VAC, and you still get 4 times lower currents (or more power at same current) compared to a 12V bed.

  20. LOL– I stepped into doo-doo deep enough to over top chest waders, I’m over six feet tall, so that’s deep doo-doo. First I was surprise to learn that the heated beds weren’t heated by line voltage, IMO that’s a no brainer. Appliances across the word use line voltage to produce heat with no shock hazard present when use properly. Getting back to the deep doo-doo. Hackaday does present instruction posts. Line voltage is useful resource that is avoided because of irrational fear. I’m calling irrational because the fear is due to ignorance. I suggest Hackaday create instructional post on how to use line voltage in a way that’s safe to life and doesn’t cause property damage when something goes wrong. Peer reviewed studies and guidelines information is preferred.

    1. You do in fact need to be a licensed engineer/electrician to design stuff with mains voltage, or you need to have your work inspected and signed for by one. It’s not just a matter of good instruction – it’s illegal for the regular Joe to even repair mains powered circuits beyond swapping a frayed power cord.

      For example, you can assemble a computer and do anything you want with the low-voltage parts, but opening the PSU and fiddling with that is illegal in many parts of the world. IF the device later catches fire and the insurance company finds out it was modified by yours truly, they can deny coverage, and if your neighbor’s house burns along with yours they can sue you etc. etc.

      1. That depends on the country. I live in Europe, and people here are allowed to do basic repairs and modification with mains voltage, like adding extra outlets. All the materials such as PVC conduits, wires, outlets, lights, switches and dimmers are readily available in the hardware store. It is however not allowed to mess with anything before the meter or fuses. Some equipment like electric stoves don’t come with power cords. Instead, you open up some compartment, and attach the wires on some screw terminals.

        Of course, you’re still responsible for your work, but you don’t automatically forfeit your insurance just because you added an extra lamp and it caught on fire.

        1. Far as I’m aware, you’re supposed to get a qualified person to fit an electric cooker. Of course, doesn’t stop me doing it. Actually more complicated than you think, it’s not just 3 terminals, there’s a dozen bits of plastic and metal in there that all need putting back in the right order.

          But even so, you’re not likely to be arrested for adding your own electrics, I dunno that it’s actually illegal. Charging money for it would be different. But I can imagine insurance companies would hold it against you. Any excuse, of course. You’ll have to check the teeny-weeny print in your policy, and the extra documents they didn’t send you. But it’s not impossible for someone to fuck up mains wiring and burn their house down, and that would be their fault.

          1. You can burn your house down in many different ways. Knocking over a poorly placed candle, or dropping a napkin on the electric stove are much more likely than a fault in the wiring, and the insurance company will pay out just the same. They would have to show that you were grossly negligent.

          2. Knocking over a candle would be an accident. Keeping barrels of petrol in your living room would be negligence. Doing your own mains wiring without the proper qualifications, I dunno which side that falls on. Real accidents can’t be foreseen, but you have to take reasonable precautions.

            Can’t say I know for sure though, I might have a poke round the web.

          3. I had my poke. Mostly people saying “may” or “could” invalidate your insurance and advising caution. Which is sensible enough. Also mention of laws mandating a professional installing things like new wiring for cookers, and other kitchen / bathroom stuff. Makes sense since there’s water about. This is the UK that I looked up, since that’s where I live. Our electrical standards are pretty high, mandatory ring mains, limits on spurs and wiring length, and even our mains plugs have half a dozen safety features built in (protection against the agony of accidentally stepping on one not included).

            I suppose it’d be a case by case thing. But I wouldn’t want to give an insurance company another excuse not to pay out.

  21. When building my printer, I also didn’t like the then state-of-the-art heated PCBs. Thus, I ordered a bespoke silicone heater mat from China, mains voltage, 400 Watts of heater power. Add an aluminium plate to the top and ground it, and safety will be sufficient. I use an SSR embedded into a wall-wart form factor project case with plenty of snubber circuits.
    Cons: More expensive than a PCB
    Pros: Heats up incredibly quickly, reasonably priced, won’t get damaged by prolonged high temperature, easy to make non-life-threatening.

  22. I don’t know, disassemble an induction heater, use the encased induction coil, feed it with mains, put a metal plate on it, and you got a safe but cumbersome heated bed using mains.

    1. You can not do it much more complicated!
      If you want to heat your print bed by induction, it has to be ferromagnetic to work with a commercial induction cookers coil. And using the copper coil as resistive heater is also no good idea.

  23. I don’t think we would save that much money by swapping the heater to a mains powered.
    You would need something that is better insulated, and made of other materials. I highly doubt it would be possible to make a PCB with high enough resistance to work at 230V. That needs to be better insulated, more material, higher cost.
    Then you need to change the driver from a simple MOSFET to something isolated.
    Based on ebay prices of the kinds of supplies used, i would say there is a $10 saved by using a smaller power one. This is easy to see disappearing in better insulation and more complex driver.

  24. FYI: The bare minimum requirement is AWG #20 wires for chassis wiring 11A which is what’s needed for 100W at 12V. I wouldn’t exactly call that a thick wire.
    While you can get away with using thinner wires fopr the AC heater, but you’ll need the wire to be able to carry the current defined by your fuse and have thick enough insulation for the voltage rating. So at the end of the day, the overall diameter for the AC heater still ended up being thicker.

  25. My grandmother had a set of glass topped buffet serving tables where the glass top had a foil element sandwiched between the glass layers. They plugged right into the wall. No components other than the heating element, line cord, and a neon bulb to remind you it was plugged in. The element terminated with a couple of fused in copper rods threaded to take nuts. Got too hot to comfortably touch, but warm enough to keep your food hot. It should be within the capabilities of a glass/ceramics hobbiest with a kiln to sandwich a foil element between two pieces of glass and fuse the sandwich together.

  26. My thoughts on this:

    First, some background. I am a design engineer working on safety of life critical hardware design and technology development for a military application. I noticed this project a while back, and I invested some thought into how I would design the system.

    First off, I would set the system up more like a safety-critical system. I would have an independent, segregated processor running the heated bed. I would have checks in the system processor and re-checks built into an independent circuit built from discrete analog/digital circuits on the system. I would use two, independent temperature sensors on the board, plus at least one thermal fuse, as well as one overload protection mechanism (most likely a fuse).

    I would build a custom circuit board, and not use soldermask on the heater side. Instead, I would bond a sheet of pyralux on the face of the heater. Pyralux is Kapton with integrated thermoset acrylic adhesive on one or both sides. 2 mil Kapton has a voltage standoff of 10KV, so electric shock would not be a problem. Bonding can be done in a standard oven using metal plates and clamps (to apply pressure), or with a special heated press.

    The rest of the circuit would be built on the backside, with judicious use of surface mount electronics rated for miltemp operation (to 125C). Standard connectors from Samtec or Molex can handle the current drive easily, with no problems from the heat. Add an acrylic conformal coat on the backside, or perhaps potting the electronics, and you can make the entire module one single brick module.

    Perhaps I should flesh out the design more than I have thus far. Perhaps if I can make the time.

  27. This is overkill. You don’t need a safety-ECU for this. It is much safer to use the thermal fuse(s) you mention and they are sufficient. If the safety can done without CPU intervention, then do it.
    2mil Kapton may have this standoff voltage, but it is scratched easily. I would bond the (high-tg FR4) PCB heater with silicone to a glassplate (pyrex or something).

    If I really have to use mains AC.

    1. Hey [Martin] I think you deserve some recognition for your contribution here tonight (or whatever time it is for you).

      I haven’t seen you around the traps here before and you have put in a good effort so welcome aboard.

    1. With the large kossel build a mains heated bed is used. The mini kossel build used a reflective surface under the heated plate, a bit heath robinson, that it is used on the new larger build is good, heat is reflected up into the bed, lt worked well on the Mini Kossel.

  28. OK … back to the task at hand. Anyone care to comment on how a toaster oven is wired? How about a thermal blanket? None of these use a 12V or 24V switching power supply brick, so how do they do it???

Leave a Reply to Alex Rich Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.