PCB Heater (Diy Joule Heating)

Intro: PCB Heater (Diy Joule Heating)

I found a website whilst searching online which documented some experiments done with printed circuit boards with a serpentine pattern etched onto them, To use them as a heater.

Uses for a heater like this could be:
• To submerge in a liquid & heat it.
• Coffee Cup warmer
• Possible heat bed for a 3D Printer ?


It is resistance heating in its simplest form, Its essentially the same as putting a high current through a resistor, The resistor would heat up. This is doing the same but on a larger scale, on a Printed Circuit Board.

*I do apoligise for the bad pictures*

STEP 1: Design of the Board

Firstly I looked on the internet to see if there were any available Vector format images of a serpentine pattern, But I was unable to find one. I quickly designed one in Eagle
(A design program to design printed circuit boards & schematics)
It’s a simple pattern with a trace in a serpentine pattern, which would provide resistance. Once a current is applied to this, it will heat up.
I have included the Eagle files below, for those of you who don’t have eagle on your computer I have supplied a picture in .PNG format.

STEP 2: Creating the Board

I then printed out this pattern onto a piece of glossy photo paper and used the common ‘Toner Transfer Method’ to transfer my design onto a piece of copper clad board.


There are a few simple steps in this method :

1.Print out your design on a LASER Printer, An inkjet will not work, Glossy paper is the best to use (or use a page out of a magazine)

2. Place your Board design on your copper clad board & iron it on the hottest temperature for about five minutes, just to be sure all the toner has been transferred.

3. Place your board in cool water for a few minutes, Then rub the paper off. You should be left with the design on the copper clad board.

4. Using your etchant of choice, Etch the board. I use ferric chloride & the method as described in 'TechShop Jim's' Instructable https://www.instructables.com/id/Sponge-Ferric-Chloride-Method-Etch-Circuit-Bo/ I use this method as its quicker & uses less etchant.

5. Once your board is etched remove the left over toner, There are a few methods to do this. I use a mixture of steel wool (To get the toner off to a rough standard) then clean the board up with acetone to ensure there are no bits of toner on the board.

STEP 3: Drilling & Soldering Your Board.

Once your board has been etched & cleaned, You can drill and solder it.

This design is extremely simple and only requires two holes to be drilled. 

I used a 1.0mm drill bit which i bought from Maplin Electronics in a set of 10 for £2.50.

To drill the hole i used a standard cordless screwdriver, But you can use a dremel or the tool of your choice. It's extremely easy to snap these small drill bits so use with caution.

Once my holes were drilled i used my multimeter to check for continuity, To ensure there were no breaks in the board. I measure its resistance, with my Fluke multimeter & it measured 8Ohms, When heated to full temperature it measured at 13 Ohms.

STEP 4: Finding an Appropriate Power Source.

To get the most out of this heater, You need a power source with a high current. I have tested 3 so far:

An ATX Power Supply, It heated up quickly & got to a good temperature. Easily available, And cheap.

A 'Wall-Wart' - Rated at 12 Volts DC, 1.2 Amps - It heated up quickly & got to a moderate temperature, Too hot to touch.

A Xbox 360 Power Supply - 12 Volts 16.5 Amps (Or so the label says), I would rate this the overall best out of the other two.

I have also tested it using a 12 Volt Car battery, But it didn't run very well.

STEP 5: Future Uses & Updates

I intend on building another one of these with a similar pattern, but larger. I will also be building a high current power supply for it.

The design of this is incredibly flexible & quite powerful for its size, My one was only 10cm by 6cm, But would make water incredibly hot. I intend on using a better power supply & I believe this could be used to boil water.

Thanks for reading...

Comments

[ScottR133]

What is the trace width? Is the copper 1 or 1/2 oz? What is resistance per inch?

Nice job!

[rajhlinux1]

How hot this gets?

[warhawk8080]

Suggestion is to get some silicon thin with mineral spirits, paint a few layers over the traces to prevent shorts...silicon can resist over 200-300C and in thin layers really don't block heat, but offer plenty of protection

[offtherails2010]

Wow Awesome !!

Thanks for sharing and keeping your design open-source, you ROCK !!

.

Just curious, how do they make the current hotbeds for 3D Printers that use 12v AND 24v - DUAL voltages but only meant to use ONE at any given time, do you think you would be able to show me a rough diagram of what the PCB traces would look like please please ? if youre curious too hehe :)

.

im asking because im thinking of attempting to etch copper foil thats been directly glued onto Borosilicate glass sheet for a 3D Printer, just as an experiment to see how well it holds up as a heatbed replacement for the aluminum, using high temperature epoxy for attaching the copper foil to the Glass, 213mm x 200mm x 3mm thickness

.

i just cant work it out how they have the option to use either 12v or 24v, the latter for faster heating of the heatbed.

.

No rush, still acquiring the parts, have the glass and foil, just waiting on the high heat glue to arrive lol

.

thanks loads in advance for any light shed on this experiment :)

[eefmf]

Very Interesting.
What is the resistance of the board?
What voltage and current did you use?

[stephenniall]

It says above, the boards resistance is 8ohms. But increases with heat, I've been using 12vdc at 16amps

[eefmf]

Do you know the width, thickness and length of the track?
There is a website that gives current capacity= htp://pcbinfo.blogspot.co.uk/2005/04/current-carrying-capacity-of-tracks.html

12vDC at 16 amps gives 0.75ohms and 192watts.
Dont run that out of liquid or it will melt the copper.
If it is 8ohms 16volts would give 2amps and 32watts.

Another problem is that copper corrodes in liquid.

[stephenniall]

No, Will measure them with a set of calipers tomorrow. It has been fine so far.

I was thinking of trying a simple electroplating & Electroplating it with tin to see what happens. Its all still in the 'thinking' stage yet.

[warhawk8080]

Electroplating would add metal to the traces, which would lower resistance and increase current flow, you want a smooth dielectric, like clear coat high temp paint or something, nail polish, polyerethane

[warhawk8080]

Conformal coat with nail polish or polyerethane ;)

[warhawk8080]

Math is wrong

12vdc at 8 ohms would only pull approx 1.5A at around 18watts...and only drop as he says when the heat of the board increases

[nadav]

I would change the design to be a large spiral (like on an electric stove top). Because, right now, the heat will cause stress on the 90 degree bends on the PCB.

Also, the corners (90 degree turns) would heat more then the straight sections, so having a spiral layout, would give you a much more evenly heated surface.

Still though, its an interesting project, it definitely gets me thinking what it could be used for.

Cheers,
Nadav

[stephenniall]

That will be in my second version, I will use bended corners in my straight sections aswell in future revisions.

[Lectric Wizard]

There's nothing wrong with your design as is!!

[Lectric Wizard]

Why would DC cause the bends to heat more than the straight runs ?? I can see high freq doing that but DC ?

[Janusmarine]

Imagine the tiny electrons driving in their electron cars, metaphorically speaking, along the straight runs, then they hit a turn. Not one of those spiraling turn you see on the freeway, but like those indy car race track 90° turns. Well since electrons don't really slow down for turns, the 90° turn takes a beating. DC is more harsh than AC or PWM, because the tiny electrons are going balls to the walls all the time, unlike AC or PWM which crosses the X-axis giving the copper time to rest every cycle time.

[AustinC42]

Please, don't comment unless you actually know what you're talking about. What you've said here is complete garbage.

In reality, the corners aren't much of a problem. They will get slightly warmer, but not by much. This is mostly because there's more copper being heated at the corners than along the straight tracks.

[Lectric Wizard]

Thanks Austin42. After 50 years in power electronics and broadcasting I knew what it was ...

[Xellers]

I love this comment.

[MostafaA4]

i want reduce the area of the heat bed to 5 cm x 5 cm can you help me