Kapton tape heated bed

Hi everyone,

I've decided to enter the DIY side of 3d printing and as a long-term project, I've decided to build a 3d printer from scratch. The bed design for that printer requires a special design for the heating pad, which I'm also trying to build myself.

I've followed this video. However, the dimensions of my build plate as well as the width of the copper tape are different.

I was wondering if anyone could tell me whether or not a 24V power supply would be appropriate for my design and how I can calculate the power of the bed. ( It may have nothing to do with the design but I really don't know where to start from).

Thanks

P.S. If you need more information about the design I'll answer as soon as possible

There are too many variables to give an accurate answer but approximately 1.12 Ohms at 24V and will need about about 520 Watts - slightly less when hot. This is based on 1mil (0.025mm) thick rolled copper. My guess is that you are using copper foil of the type made for stained glass. Note that tape is not specified for esistance and comes in 1.25mil and 1.5mil.

Mike

I don’t know if by telling you the thickness of the tape (0,06mm) you would be able to tell me if that is the right one to go with or if I should try buying a different one.

And apart from that, from what I’ve understood it would be better to buy a 520W+ power supply rather than using the one I currently have. Am I right?

The power consumed by a resistor is equal to the voltage squared divided by its resistance. The resistance is dependant on the material (copper) and is proportional to the length of the conductor (wire or foil) and also to 1 divided by the cross sectional area. For the 1mil (0.025mm) by 6mm wide this is 0.15 square mm and the resistance will be about 1.11 ohms so power will be 514W. If the copper foil is 0.06mm then the resistance will be a little less than 0.5 ohms. At 24 volts this will draw a bit more than 48 amps and will use about 1200 Watts. So 24V is a lot more than you can reasonably get.. At 12V the draw would be 24 Amps and power would be 288W. All these figures may change somewhat with changes in copper specification - annealed or hard drawn, pure or alloyed etc.. Edit: You will need to look at the specs for the mosfet on your controller as they may not be good for this current.

Concensus on this forum is that a 300mm square bed will need more than 270W - but much more than twice that will be afire risk if the controling mosfet fails.

The power supply you linked to will not do. Use something that can comfortably deliver at least 50% more than the heater needs so about 400W will be O.K. The wires to the heater for this current should be flexible wire greater than 1.5 square mm and keep them in open air to stop local overheating.

Mike

Edited 1 time(s). Last edit at 08/26/2021 09:43AM by leadinglights.

A heater doesn't care if it's powered by DC or AC, so why spend extra money on a big power supply to give it tightly regulated DC? Switching power to a DC heater requires a DC SSR and they tend to get hot in operation, so you have to cool them, usually with a heatsink (cost) and fan (noise). Line power switching using an AC SSR leaves the SSR as cool as a cucumber without a fan, and often without a heatsink. Low voltage DC power requires a heavier cable- its lack of flexibility can be a problem depending on the printer architecture.

I suspect that variations in the thickness/width of the copper tape will result in hot spots that may burn through the conductor or just make for uneven heating. You can buy 300mm square heaters from competent manufacturers in China for about $70-80. How many hours, generations of prototypes, and how much material cost are you going to need to develop your own heater? Some things (like designing and building a 3D printer) are worth DIY because you learn a lot of useful stuff even if there are commercial alternatives readily and cheaply available. Designing and building a bed heater doesn't seem like one of those things. The heater should be the size of the bed plate- leaving an unheated rim around the plate will lead to warping when it's heated.

Not sure how much power you need for the bed heater? This calculator will tell you how long you'll be waiting for the bed to reach target temperature based on heater power and bed plate material and size.

I don't know what you're planning to use for the bed plate, but you might want to consider a kinematic mount that uses 3 support points instead of putting "leveling" screws at 4 corners (and the center?) of the plate. The hole at the center of the heater will create a cool spot there, right where slicers are going to put the print. If you use the right bed plate (cast aluminum) on a kinematic mount, you won't have to relevel the bed after initial set-up. You won't need auto leveling or flatness compensation, considerably simplifying the electronics, configuration, and Z axis mechanism and reducing the cost of the printer. It will be easier to troubleshoot and maintain the printer, too.

Those so-called "LED strip" power supplies are unreliable junk. Spend a few $ more and get a MeanWell supply that has safety certifications and built-in protective circuits that actually work. Buy a power supply that is rated for at least 30-50% more power than the printer will actually need. Don't operate anything, especially power supplies, at their capacity limits. It's a recipe for disaster.

Edited 1 time(s). Last edit at 08/26/2021 10:22AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Awesome. Thanks a lot!! That really helped

The thing is that my design requires five magnets (3mm thick) (I want to avoid using a magnetic surface to make the heat transmission as efficient as possible) to hold the flexible steel platform. However, the aluminium base that I’m using is 3mm thick and I don’t want it to be any thicker because of the weight. Those magnets will be embeded into the base and in order to prevent them from falling I’m going to install “mounts”. That is why the heating pad has 5 rectangular holes. Don’t know if there is any other way to do it…

Apart from that, are there going to be any changes if I use an AC power supply rather than a DC one apart from the cost and safety aspects? (I might change to an AC one after what you’ve told me)

Edited 1 time(s). Last edit at 08/26/2021 01:49PM by pmd10.

I won't say don't use a base as thin as 3mm as it may work satisfactorily for whatever your tolerances but generally 5mm thickness is a consensus minimum for a 300mm square bed. 6mm is better and cast aluminium tooling plate as recommended by DD is good. For smaller beds the options are much more open - even using 3mm float glass directly on a heater has worked.

There is a potential pitfall when using thin steel on thin aluminium and then heating the whole assembly. The aluminium will expand about twice as much as the steel and the whole assembly will become concave until friction is overcome - which may be with a creak or a loud click.

As I said, I won't advise you that anything is impossible as I have seen too many accurate, reliable and inexpensive devices made using techniques that would have shocked my teachers. One thing I think I should warn you against though is making a DIY mains voltage heater. High current can lead to fires but they announce themselves first with a stink, but high voltage can kill you without warning.

Mike

In that case I'll most probably use a 5mm thick aluminium plate as you say and I'll see what I can do about the steel plate.
Thanks a lot for you time

What kind of printer are you building- is the bed moving back and forth or up and down, or not at all? How fast do you intend to print? How will the aluminum plate be mounted?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Hello pmd10.
I'm VERY new to the 3d printing scene. I have a Flashforge Creator Pro 2, am very dissatisfied with small bed and printable volume. I am a long-established DIY mechanical and electronics maker and have decided to make my own 3d printer with a 500mmx500mmx500mm print volume.
Now, to get to the point, because of that, I've been doing a lot of (re-)searching in the internet and have found a bloke named Tom on youtube who has a bit to say about heated beds, amongst other things. See one of his videos on
[www.youtube.com]
In this video, he says that he thinks you need at least 0.2W, ideally 0.6W and up to 1.0W per square centimetre for bed heating.

Now, you could use DC power supply with a mosFET switching by PWM but as others have said, if the switching transistor fails as short-circuit that will lead to bed overheating with the possibility of burning out your bed heater or even causing a fire.
My design calls for a big heater for the bed, so I've found (on BangGood) a 1400W heater which is specified for 240V AC or DC, I've chosen to power it directly from 50Hz AC mains and control it with a solid-state relay (Schneider Electric 6210ASX-1) of the zero-crossing switching type (ZCSSR).
Using a ZCSSR avoids high-frequency radio noise caused by fast switching significant voltages as happens in DC PWM circuits. However trying to control temperature using PWM means that power will be on and off each an integral number of AC half-cycles, so the PWM cycle time needs to be longer in order to achieve a reasonably low granularity in the temperature control. The PWM controller (I've chosen Marlin running on an Arduino Due with an Ultimaker 1.5.7 shield) uses a little over 7Hz PWM cycle time for all the heaters, I'm going to have to dive in to the software and put in a patch to extend the bed heater cycle time by a factor of 8 or 16 which I'm hoping won't effect the heating accuracy of my large aluminium bedplate.

I hope this gives you an idea or two and good luck with your build.
Peter.

... if you're new to 3D-printing, then better start with 300x300mm max -- bigger brings a lot more problems! ... even 300x300mm is not so easy - many experienced DIY-ers resigned with bigger volumes for heating and warping issues ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

As VDX says, even 300mm cube is a stretch. The reason that the RepRap movement took off was that at 200mm cube, or even better, at 150mm cube, simple construction and materials like float glass, could give astonishing accuracy and speed. this was sort of a magic size for the DIY world.

One way a DIY maker could make a large 3D printer is to allow very undemanding tolerances which can be used to make things where the positional tolerance could be plus or minus 2mm or even more.

Things that a 3D printer with reduced accuracy might be used for include making an armature or framework for a human-size sculpture. Another example is visually accurate representations of engined for early aircraft - this allows museums to replace the engine in an aircraft hung from the roof of a museum - and the heavy engine displayed safely on the floor.

For larger sizes with accuracy in the 0.1mm region, the DIY maker who does not have degrees in mechanical, electronic and thermal engineering, should concentrate on making everything really really really solid.

Mike

Thanks for all the responses but there is still one thing I don’t get.

As Mike said at the beginning, 300x300 beds will need between 270-540W of power. Then Petters’ video ( very helpful by the way ) showed that some beds can even use 700W. And, apart from that, he has bought a 1400W heater.

Now, my question is, what is the point of all this power if most motherboards can’t even get to it? To set an example none of the bightreetech 6 motors motherboards ( the octopus ( 300W) and the skr pro ( 360W )) can?

Apart from that, how would a, let's say 360+W, bed perform with any of the previous motherboards?

To get to higher powers a solid-state relay would be used for DC and either a solid-state relay or an SCR with a zero-crossing driver for an AC circuit. I can't recommend any particular one but I am sure I have seen such recommendations on this forum in the past.

Mike

... there are several "big" 3D-printers with four 400x400mm-heaters, cso overall 800x800mm heated area ... but they are all with enclosure and air-heating too ... and with the heat-tolerant setup and mechanics pretty pricey too (search them in the range between 10k to 20k USD)

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
pmd10
In that case I'll most probably use a 5mm thick aluminium plate as you say and I'll see what I can do about the steel plate.
Thanks a lot for you time

You shouldn't have a problem with different expansion of your steel bed and aluminium plate as your magnetic fixing method will allow the two to slide against each other.

Quote
pmd10
Thanks for all the responses but there is still one thing I don’t get.

As Mike said at the beginning, 300x300 beds will need between 270-540W of power. Then Petters’ video ( very helpful by the way ) showed that some beds can even use 700W. And, apart from that, he has bought a 1400W heater.

Now, my question is, what is the point of all this power if most motherboards can’t even get to it? To set an example none of the bightreetech 6 motors motherboards ( the octopus ( 300W) and the skr pro ( 360W )) can?

Apart from that, how would a, let's say 360+W, bed perform with any of the previous motherboards?

Others have mentioned solid state relays as the final switch in a high-power PWM temperature control system. In my case, my motherboard delivers a 12V-20V (depending on what power supply is used) signal meant to directly power a lower-power heater element. I intend to use that signal to switch a solid-state relay which can accept an input as low as 3V up to as high as 32V.
I've had a quick look at both of the motherboards you mention and the hardware of both is capable of driving a solid-state relay in exactly the same way. I expect that my board and both the boards you mention drive using an open-collector or open drain from a power transistor to drive the heater and might need a resistor across the outputs so that its switches off cleanly.

BTW, Tom also has a video on youtube in which he conducts a very professional, technical comparison of the BLtouch and 3Dtouch bed-levelling sensors. His performance in this testing earned my respect.

Peter.

Edited 1 time(s). Last edit at 08/29/2021 08:27PM by OldBikerPete.

Quote
VDX
... if you're new to 3D-printing, then better start with 300x300mm max -- bigger brings a lot more problems! ... even 300x300mm is not so easy - many experienced DIY-ers resigned with bigger volumes for heating and warping issues ...

Thank you Victor. In building my 500mm printer, I am very aware of the effects of differential expansion in the bed and of shrinkage occurring in a large print. I intend only that my printer will always be larger than any object I wish to print (the largest I've previously wanted to print was 250mm) and that it will be big enough to print multiple much smaller objects in one session if the design I'm building calls for it.

Peter

Quote
pmd10
The thing is that my design requires five magnets (3mm thick) (I want to avoid using a magnetic surface to make the heat transmission as efficient as possible) to hold the flexible steel platform. However, the aluminium base that I’m using is 3mm thick and I don’t want it to be any thicker because of the weight. Those magnets will be embeded into the base and in order to prevent them from falling I’m going to install “mounts”. That is why the heating pad has 5 rectangular holes. Don’t know if there is any other way to do it…

Apart from that, are there going to be any changes if I use an AC power supply rather than a DC one apart from the cost and safety aspects? (I might change to an AC one after what you’ve told me)

I like your idea of a steel bed held in place with magnets. I'm curious as to why you are using rectangular magnets when cylindrical neodymium magnets are readily available and so much easier to mount.
Using magnets like that will allow the two plates to slide on each other and avoid problems caused by the steel and aluminium having different rates of expansion.

I would suggest applying a heat-conducting grease as used in electronics heat sinks between the aluminium and steel. Contact between the two plates is unlikely to be absolutely uniform possibly causing irregular temperature across the steel plate.

I'm thinking (but I might be wrong - It's been known to happen. I remember once back in 1970........) that a 3d printer doesn't put any transverse force on a print so one can get away with doing this. If that is wrong, then a couple of steel pins along one edge should overcome that problem.

I had been thinking of placing glass on top of my aluminium plate but now perhaps I'l try steel first.

Peter.

Quote
the_digital_dentist
A heater doesn't care if it's powered by DC or AC, so why spend extra money on a big power supply to give it tightly regulated DC? Switching power to a DC heater requires a DC SSR and they tend to get hot in operation, so you have to cool them, usually with a heatsink (cost) and fan (noise). Line power switching using an AC SSR leaves the SSR as cool as a cucumber without a fan, and often without a heatsink. Low voltage DC power requires a heavier cable- its lack of flexibility can be a problem depending on the printer architecture.

I suspect that variations in the thickness/width of the copper tape will result in hot spots that may burn through the conductor or just make for uneven heating. You can buy 300mm square heaters from competent manufacturers in China for about $70-80. How many hours, generations of prototypes, and how much material cost are you going to need to develop your own heater? Some things (like designing and building a 3D printer) are worth DIY because you learn a lot of useful stuff even if there are commercial alternatives readily and cheaply available. Designing and building a bed heater doesn't seem like one of those things. The heater should be the size of the bed plate- leaving an unheated rim around the plate will lead to warping when it's heated.

Not sure how much power you need for the bed heater? This calculator will tell you how long you'll be waiting for the bed to reach target temperature based on heater power and bed plate material and size.

I don't know what you're planning to use for the bed plate, but you might want to consider a kinematic mount that uses 3 support points instead of putting "leveling" screws at 4 corners (and the center?) of the plate. The hole at the center of the heater will create a cool spot there, right where slicers are going to put the print. If you use the right bed plate (cast aluminum) on a kinematic mount, you won't have to relevel the bed after initial set-up. You won't need auto leveling or flatness compensation, considerably simplifying the electronics, configuration, and Z axis mechanism and reducing the cost of the printer. It will be easier to troubleshoot and maintain the printer, too.

Those so-called "LED strip" power supplies are unreliable junk. Spend a few $ more and get a MeanWell supply that has safety certifications and built-in protective circuits that actually work. Buy a power supply that is rated for at least 30-50% more power than the printer will actually need. Don't operate anything, especially power supplies, at their capacity limits. It's a recipe for disaster.

I agree totally with all you say. That calculator says that my 500 x 500 mm aluminium 6mm plate with a 500 x 500 mm 1400Watt heater will take 284 seconds to heat from 20C to 110C. I'll need a bigger heater - or perhaps insulate the underside?

Edited 1 time(s). Last edit at 08/29/2021 08:39PM by OldBikerPete.

... here one of my heated beds with 460x295mm size (without the isulation on the underside) -- there are 7 rod-heaters embedded in the 34mm thick Aluminium plate (could be more - 13 through-bores from the long side, 9 across).

OK - this is not so meant for the DIY-range, more for comercial heaters ... was a prototype to test/measure for uniform heating -- targeted was 1x1m, but never built

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
pmd10
I don’t know if by telling you the thickness of the tape (0,06mm) you would be able to tell me if that is the right one to go with or if I should try buying a different one.

And apart from that, from what I’ve understood it would be better to buy a 520W+ power supply rather than using the one I currently have. Am I right?

That is a good power supply you already have but even so, I would not use that to power anything bigger than a 200W or so load if you don't want it to fail 7 hours into an 8-hour print!

That's an interesting bed, Victor.
Can you tell me if insulating the underside of the bed is commonly done? What material do you use the insulate under yours?
Peter.

...cotton or mineral wool ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

... for the max. temps -- we used cotton for up to 200degC ... mineral wool if higher temps or eventual "overheating" possible, when testing controllers (had an "accident" with melted heated bed )

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
OldBikerPete
...................You shouldn't have a problem with the different expansion of your steel bed and aluminium plate as your magnetic fixing method will allow the two to slide against each other..........

True as long as the bed is fairly rigid and the friction between magnets and steel plate is not too high. Thin beds and differential expansion is a recipe for stick-slip behavior with the middle of the bed rising and falling.

Quote
OldBikerPete
..................Can you tell me if insulating the underside of the bed is commonly done? What material do you use the insulate under yours?.................

Insulating the underside of the bed gives a small but useful reduction in needed heating power. Preventing breezes from the side also gives a small but significant improvement - more if it is a strong breeze. I use metalized bubble film for this but only to a little over 100°C

Quote
OldBikerPete
.................a 3d printer doesn't put any transverse force on a print so one can get away with doing this

Officially no, but one in a while a bit of plastic will rise above where it was printed and snag on the nozzle - so some transverse retention is needed.

For any DIYers, a couple of pictures of possible alternative ways of making your own heater



The top picture shows an etched PCB bed - short video at [www.youtube.com]
and the bottom two pictures show a heater using a nichrome wire in PTFE sleeving, bare plate, and assembled heater. A video showing the milling of the other side at [www.youtube.com] This is actually the other side of the bed but the milling is the same and is being done on a 3d printer frame - as is the marking of the etch mask for the first heater.

Mike

Quote
leadinglights
To get to higher powers a solid-state relay would be used for DC and either a solid-state relay or an SCR with a zero-crossing driver for an AC circuit. I can't recommend any particular one but I am sure I have seen such recommendations on this forum in the past.

Mike

I've settled on a Schneider Electric 6210SZS-DC3 zero-crossing type SSR. I've attached a link to the data sheet and a capture of a retailer's listing. Note that a significant heat sink is recommended. The entire base of my printer will be a custom aluminium box, so I will bolt the relay to that.

Data Sheet: [docs.rs-online.com]

PS. The data sheet is about 1.9MB and I tried to attach it, since the instructions associated with attachments said that 8MB is OK - but it spat me out saying the file is too big. Instructions need updating.

Edited 1 time(s). Last edit at 08/30/2021 07:28PM by OldBikerPete.

It seems like you're utilizing copper foil specifically designed for stained glass. Please note that our tape options come in 1.25mil and 1.5mil thicknesses, however, resistance specifications are not provided.