We do a lot of research and development on 3D printing systems at SD3D in an effort to improve the quality and reliability of our machines and to help select the best 3D printer technology available for our production factory. Recently, we took six off-the-shelf desktop 3D printers and our proprietary 3DGenie FDM printer and compared the quality of the heated build plate (HBP) with respect to surface temperature gradients.
So why do surface temperature gradients on the HBP even matter? Thermoplastics are very sensitive to temperature fluctuations, specifically when they are being rapidly extruded at temperatures well above their glass transition temperature (Tg). When there is a significant difference in temperature on one side of the HBP compared to the other, we call that a temperature gradient. These temperature gradients and environmental fluctuations are the leading factors that lead to part failure due to “warping”. This effect which we have come to describe as warping is little more than uneven shrinkage of the polymer that leads to a portion of the print dislodging itself prematurely from the bed. Temperature gradients above the HBP can also lead to an effect known as “delamination” in the case where the print remains adhered to the bed, but individual layers above the HBP surface begin to weaken and separate from each other. So if you want to have strong parts that print reliably, a good starting point would be to ensure your HBP is as uniform and gradient-free as possible.
1. MK2 (RepRap)
First up on the list to be reviewed is the commonly used MK2 12V heated build plate as seen on many RepRap based open source 3D printers. These images were taken of an Airwolf XL which uses the MK2 heater with a 1/4″ borosilicate glass plate as the top surface. This type of HBP uses a printed circuit board (PCB) to heat up the glass. The traces in the PCB create a resistive circuit which generate a significant amount of heat when ample current is passed through it.
The heat signature in the image below shows higher temperatures on the HBP as warm colors and lower temperatures as cool colors.
As can be seen in this thermal video of the MK2 holding at 60C, the PCB+glass stack is unable to create an even heat signature. There is a significant hot-spot right of center and large temperature drop off zones near the left corners. The differential between the hotspot (bright yellow) and corners (deep blue) is approximately 20C. Due to this asymetric heating, it makes it very difficult to print reliably on these heated build plates with materials like ABS which are very sensitive to these kinds of temperature gradients.
The 60-110C heatup cycle and hold at 110C show some improvement to the temperature variation from left-right, but it would still preclude the user from making warp-free ABS parts that span the full bed area. As such, we would recommend utilizing no more than 40% of the full bed area on any single ABS piece using this style of heated build plate.
Another scenario which is beneficial to look at is the cool down cycle for the HBP. During cool down stresses are increased at the print surfaces as the bottom layers begin to contract. It is very common to hear cracking (delamination) during the cool down phase if that contraction happens unevenly due to temperature gradients on the build plate or within the build environment. The cool-down cycle for the MK2 shows this effect as the hotspot retains while the corners rapidly begin to cool.
2. Flash Forge Creator Pro
Next up on our review is the Flash Forge Creator Pro. The heat-up cycle on this 3D printer was surprisingly even without any noticeable hot spots or gradients. The heat appears to generate slightly off center during this initial heating phase, but it does not appear that it would have any significantly negative effect on the performance.
Once the FF reaches it’s 60C setpoint and holds there for a few minutes, the steady state thermal gradients begin to appear, with an asymmetric hotspot trailing down the center of the HBP. The hotspot is concentrated at the rear of the HBP just right of center.
What’s probably most interesting about the Flash Forge’s heated bed is that the hotspot actually appears to move depending on the temperature. As can be clearly seen in the 60C hold (below left), the hotspot begins just right of center and near the back end of the HBP. However, as the temperatures continue to rise, that hotspot travels horizontally eventually settling in the back left corner of the HBP.
As the FF cools down the hotspot quickly returns to the center with residual spikes in heat signatures scattered throughout a centerline dividing the HBP vertically down the middle. These cool-down temperature gradients are fairly severe and due to the low thermal density of the FF HBP. With these in mind, we would recommend splitting parts up between left and right sides of the HBP, avoiding the hotspots at the centerline.
3. Punchtec Ordbot Hadron
Moving on to our next printer on the list, the Ordbot Hadron by Punchtec. This little 3D printer uses a PCB to heat the bed, similar to the Arwolf XL. One interesting aspect to note on the Punchtec printer is how the HBP was assembled. Punchtec apparently used a high temp epoxy to permanently bond the glass to the HBP. While this does increase the efficiency of thermal transfer compared to using binder clips, it also creates an inverted temperature gradient than what we have seen on the other machines.
You can clearly make out a hot spot on each corner along with the typical centralized hotspot. The four peripheral hotspots actually decrease the temperature gradients out near the corners, but this leaves a significant gradient along the inner perimeters. As the temperature continues to rise, these gradients settle some. This can be seen in the following two clips which show 0-60C heat-up (left) and 0-110C (right).
While the punchtec has seemly performed better during heat-up and hold than the FF or Airwolf; it’s performance during the cooldown cycle (below right) falls short. The relatively even temperature gradient at 110C rapidly fades and begins to migrate left of center, creating an asymmetric temperature gradient. This will lead to delamination on taller ABS parts and difficulty removing some parts from the build plate.
4. Robo 3D R1
Robo3D uses a low cost resistive trace element to provide heat to the build plate. This method, while cost effective, does not provide very good results if your goal is to create a uniformly heated build surface (which should be the goal of any 3D printer manufacturer). As can be seen in the 0-60 heatup cycle (below right), the resistive traces can clearly be visualized through the heat signature, meaning the temperature gradients on the surface are drastic and non-uniform.
These gradients exist both at 60C as well as 110C, meaning that there is a high probability of warping for just about any ABS print. With such extreme gradients, warping with PLA will also occur. It is likely that a sizable portion of Robo failure rates can be attributed to the HBP design. If you own a Robo, we would highly recommend upgrading your HBP.
5. Lulzbot TAZ5
The Lulzbot Taz5 also has a very bizarre thermal pattern on their HBP. Looking at the physical structure of the TAZ heated bed cannot quite describe the presence of the non-uniform cool spots which can be seen during heat-up:
Once the Taz5 has reached its set point, these gradients appear to settle out nicely, leaving a single cool spot where the power and thermistor plug into the silicone heater. Warp free printing should be relatively easy with the Taz5 so long as the outer perimeter and plug cold spot are avoided for part placement.
6. SD3D 3DGenie
Since we have been researching the effects of HBP thermal gradients for quite some time, we decided earlier this year to take a shot at fixing the issue. We incorporated a very special bed stack into our beta 3DGenie (3DG) cloud automated FDM printer with the aim to eliminate significant thermal gradients at the print surface. To our surprise, this bed has actually exceeded our expectations in practice, allowing for completely non-destructive auto ejection of prints. But comparing the thermal shots we took of our bed to the others in the shop tells the full story:
The heat-up cycle is even with minor gradients at the corners which quickly dissipate to become negligible (<5C) before the bed reaches its 60C setpoint. (Note: at some point during these 3DG shots, the vertical line artifact showed up, extending beyond the frame. We will ignore this artifact for the remainder of the article). As the temperatures continue to rise to 110C and hold, the gradients at the corners remain negligible with an extremely stable thermal signature at the HBP surface.
The HBP on 3D Genie continues to perform exceptionally through the cool down phase as well. There are no visible hotspots or coolspots generated as the heated bed gradually decreases temperature until it reaches ambient. We believe it is this extremely even and stable thermal signature that provides the non-destructive auto-ejection capabilities we see from this HPB stack.
So whenever you are in the market for your next 3D printer, just remember that not all heated beds are created equal!
We would like to thank NXT Robotics for providing the Lepton thermal imaging module used to take these shots.