Making a heated build platform (HBP) for Type A Machine, or any 3D Printer

First of all, let’s see the video of the heated print bed in action! This table is built for Type A Machine 3D printer, but it should works for all other printers.

I’ve been playing with my Type A Machine series 1 3D printer for a while. I like its volume – 9″ x 9″ x 9″, which is a lot bigger than most of its competitors on the market, and, of course, its affordable price. All magic, however, comes with a price. Its lacking of a heated print table makes my life really hard to print anything big with ABS. Comparing to my smaller UP!3D, Type A Machine usually ends up with very serious corner curling problems, which sometimes turns my print into a completely useless piece of part.

This problem has been annoying me ever since I had it. You might not be familiar with this problem if you print with PLA filament, or you print small parts with ABS. Otherwise, curling and warping are becoming dominant issues for 3D printing. Unfortunately, many of my models are quite big (more than 100mm x 100mm). And PLA doesn’t work for me. Attaching a HBP seems to be an ultimate solution, but the manufacturer doesn’t seem to provide a HBP for this model anytime soon. I decided to make one on my own.

3D_Print_Corner_Lifting_Curling

Figure 1. Example prints from Internet with problems of corner lifting, curling or warping.

I am using 1/8″ thick aluminum sheet as my table material since I am lucky enough to have access to a CNC machine. The table dimension is all you need to know before milling the sheet into a build plate. The company, however, is releasing the drawings for anything else except for the table. I ended up measuring everything with my caliper. The overall size for the table is about 260mm x 310mm. The big screw hole is 0.31″ in diameter, and the small one is 0.158″. The center of these two holes are about 0.415″ apart to make a long slot. There are four of those slots sitting on the corns of a 6.05″ x 11.07″ rectangle. All the measurements don’t need to be crazily accurate, and they work just fine for me.

Aluminum_Sheet_For_HBP

Figure 2. The aluminum sheet is prepared for CNC mill.

Mill_Slot_On_Corner

Figure 3. Not a CNC expert, the hole was milled quite ugly. Countersink-ed the edge a little bit to make it “smoother”.

I talked about temperature control circuit using micro-controller, thermocouple and power relay in my last post about “hacking a toaster oven into SMT reflow oven“. To control the temperature of the print table, it is basically the same thing. I am going to using the same circuit except for the heating source and switch, for which I choose metal clad resistor (TE 10ohm, 16W, screw mount) and N-channel power MOSFET (Fairchild FQPF33N10L). This power MOSFET has a low gate threshold voltage (2V), so the micro-controller can directly control it.

Metal_Clad_Resistor_HBP

Figure 4. The resistor is mounted on the back of the aluminum sheet with screws. I recommend using thermal-conductive tape for better heat transfering.

8_Metal_Clad_Resistor

Figure 5. Eight resistors are all mounted and wired in a pattern of 2 parallel groups of 4 series. I am choosing this pattern for the least amount of wiring. Four 10ohm/16W resistors in series will require 48V to obtain the best performance. Each resistor consumes 1.2A current, roughly 14.4W of power. So a 48V AC DC power supply with load capacity of at least 2.4A  is a must.  The actual test show that the heated build platform can heat up to over 100°C in room temperate with power consumption at about 150W. The thin cable in brown color is the thermocouple, which senses the temperature back to the micro0-controller.

N_MOSFET_Switch

Figure 6. The N-Channel MOSFET is wired to the micro-controller. I suggest using a 100ohm in series between the MOSFET gate and micro-controller output, especially in this case where a big current needs to be switched on/off. The rest of the wiring is pretty simple. Since the N MOSFET is a low-side switch, +48V from power supply goes directly to the load, the other side of the load connects to the drain of MOSFET, and source goes to ground.

Print_Flat_Base_No_Corner_Lifting

Figure 7. Finally, here is the photo of a new print finished on the HBP! The corners are completely stuck to the plate flat!

Big 3D Print Result

Figure 8. The biggest part I’ve ever print! It took quite long, about one day and half, but turned out to be a big success. The heated print bed really helped keep the bottom flat. As you can see, the supportive raft is peel-able now. I will talk about the tricks to 3D print peel-able base next time.

Advertisements

9 thoughts on “Making a heated build platform (HBP) for Type A Machine, or any 3D Printer

  1. Hi XueMing,

    How long do you keep the plate heated? Do you stop after the first layer or you let it on past that? What is the longest amount of time your heated bed was on continuously? Thank you so much for your help.

    Regards,

    Val

  2. I’m planning to build this table heater for my Mbot 3D. Do you have the schematics and code you used? Thank you for your help.

    1. Attached a hand-drawn schematics for your quick reference as I didn’t have one. Code part is basically running a PID loop in the MCU. Depending on different MCU, PID implementation may change a lot, but the idea is simple. You should be able to find it anywhere online.

      DIP control

      1. Thank you for clarifying the circuit. I will be using a Parallax Propeller microprocessor for control. It looks like it should be fairly easy. I will have an aluminum plate bonded to a borosilicate glass plate. I may use temperature hysteresis instead of the PID loop. It seems like it should produce similar results without having to calibrate the PID.

        Two more quick questions:
        1. Is that an ABS slurry on your plate?
        2. Did you have to calibrate the temperature, or does it come out correctly from the control chip?

        Thank you for your help. I appreciate it.

        Regards,

        Jim

      2. 1. Yes, I dissolve ABS scraps in acetone to make “slurry”.
        2. No, temperature readout from MAX6675 is already calibrated.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s