Project BFE

This project is to outline building an ender 5 into a large scale 3d printer.

Public Chat
Similar projects worth following
Large scale 3d printing systems often are highly expensive and closed source despite the reality that they are also using from the open source community. In an industry flooded with inexpensive "consumer" level systems this project is going to explore the usability of a popular consumer 3d printer the ender5 in industrial applications.

The ender 5 has a standard cartesian xyz setup and due to its 2020 & 2040 aluminum extrusion design, the flexibility, ease of assembly and modification makes it a optimal starting point for a robust larger scale system.

Areas of focus:
Z axis modification
XY expansion
Heated Enclosure
Building surface
Electronics & software

We kick off the project midway through R&D. Here is a brief what's happened so far. 

Our build is designed specifically for printing a large enclosure which dimensions are 60in x 48in x 5in and a print goal of sub 3 days while still being strong enough to hold 500lbs of dynamic distributed weight. 

ABS was chosen because it is one of the highest thermally conductive plastics and is also engineering grade which is important for high printing speeds with good layer adhesion. The ender5 stock is not a unit which can handle this material at any speed above 40mm/s without warpage. Over 40mm/s you get layer delamination and at 200mm/s the delamination is so bad the layers separate immediately after extrusion due to the high warping of not having a heated enclosure. Please note the stock build surface is not suitable for printing with abs. Source a powder coated build surface or other applicable surfaces for printing abs.

The tests we made to get to that point we're the following:

1. Set part cooling fan speed to zero. 

This should be by default in your slicer of choice but if not the reason is abs is very thermally conductive which is why it warps. This helped but the stock hotend fan was also blowing cold are on the part. 

2. Get out your tape!

Using the highest quality tape you can source for higher temps seal the bottom of that hotend the best you can. Blue painters tape worked just fine without burning. 

3. Increase nozzle diameter 

This was done because print speeds of 0.4mm nozzles with big parts is so slow the chances for failed prints only increases and the theory is that thicker extrusions will increase the bond between layers because there is more volume to retain heat.

4. Switching the part & hotend fan connections

This allows you to use the fan speed in your slicer to just give enough air accross the heatsink to keep it cool for retractions but lower the amount of leaking airflow over the printing part. This was the chosen path because we aren't interested in changing firmware settings at this point. More on that later! 

5. Removal of all fans from the hotend

This enabled much greater chances of allowing heat creep to heat in a longer zone. And give abs more time to heat up for even faster print speeds. YES! This is against all normal printing but a plan for never needing any retractions is part of this build. If you're looking to use this as a guide just leave your heatsink fan as normal. (But upgrade because it is not a good fan shroud design.)

By this point the abs was warping less but still only allowed for slow print speeds .

6. Simulating a heated chamber

This can be done in many ways but the choice to use a hot air gun with adjustable temp was too redneck to pass up! The ace branded heat gun rests perfectly on the y axis motor mount and the extrusion blowing directly in the center where the test piece is placed. Start the temp around the "glass transition" temp. FYI abs has no official temp for this as the material is amorphous and has no exact temp, but 100-110degrees celsius is about the range that is recommended by plastic manufacturers. Due to room temp and the reducing temp as you get farther away from the heat nozzle turning the temp up was needed to reduce warping. Set the temp  20-100 degrees celsius above the glass transition temp depending on your room conditions.

Success! The heated area test worked at preventing abs warpage. On a side note if you are wanting to investigate the use of a hot air gun on open air systems post in the comment section!

Now that proving a heated enclosure is needed and that abs can be a great material for high speeds the system build can start. 

z motor mount1.stl

2020 mounted Z-axis motor Nema 17 standard bolt pattern and planetary gear with standard 4x M3X28mm spacing.

Standard Tesselated Geometry - 295.00 kB - 08/16/2019 at 21:53


  • Z Axis motor mount

    simplisticlines5 days ago 0 comments

    Z axis mount is both machinable and printable. Because the goal is to have a heated chamber print in a high temp plastic like ABS or better PC, PEEK, Ultem etc. Depending on mounting it is possible this mount will not have any direct heat within the chamber. More on that later. 

    Potential mounting solution. Still searching for a 6mm ID GT2 belt. 

  • Increasing dehydrator temp

    simplisticlines6 days ago 0 comments

    Food dehydrators, toasters and many other coil driven heating systems have very simple circuits. The upper limit of the system is controlled by a thermal fuse and have a potentiometer for adjusting temperature. For this loaner unit original fuse was rated for about 80c. Perfect for pla and petg which have lower then 80c glass transitional temps. Because this build is focusing on the higher temp ABS the requirements are for at least 110c. Opting for a much higher rates fuse a 216c fuse was selected so further testing at higher temps can be tested. DISCLAIMER AND WARNING! This is so far untested and might burn out if full power is used. Please use caution when altering heating devices!

    Original on top. New on the bottom.

    Original fuse in place.

    New fuse in place.

  • XY gantry assembly

    simplisticlines7 days ago 0 comments

    Start disassembling your ender5. You want to remove all electronic wires and set the lower portion aside leaving what you see here. 

    Use the side rails as a drilling guide. As shown below. Notice for this build 2040 extrusion was selected but is not required. The ender5 has 3x M5 threaded holes on one end and the other is a single M5 threaded hole. The drill bit should be properly sized for threading. Because in the US it is expensive to source metric drilling bits the imperial equivalent is 0.161 inch or a #20 if you have a full drill set. 

    If you have clamps great grab them and lock that firmly to the end. If you don't and have been printing with ABS you can use ABS juice to hold the extrusions down making temporary glue for drilling. 

    Your drilling should be very close to factory. Don't worry about the remaining glue it rubs off without any challenge. 

    Now let's do the same to the other side. 

    Next tap all 4 holes you just made with an M5 thread tap. At the same time glue your next extrusion down so it has time to set while you tap threads. 

    After you are done drilling and taping you can start reassembling using all the original hardware. If you did a good job of aligning your new screw holes the frame should fit square as it did out of factory. 

    Fit all the belt system on to your newly assembled frame. 

    Run the GT2X6MM belt in the center hollow channel. 

    Install the original x axis and zip tie the belt to the original belt hooks. This is temporary so you can test movement and make adjustments as needed. 

    You should have a fully assembled XY assembly. From this point if you are using linear shafts you are on your own. For this build we are adding 4 gantry plates to the ends and using the 2040 extrusion as the z axis rails. 

    Fit the gantry plates you source the best you can. The parts sourced for this build nicely aligned with original Y axis bracket. The only ugly was drilling space for the 8mm belt shaft and the replacement to a flat M5 because the stock screw prevented the 2040 extrusion from sliding down the new Z bearing. 

    The idler side is a bit more tricky because it does not align with any of the gantry plate holes. However it is a quick modification to use the gantry plate as the new idle pulley bracket. Please note the motor side still has a 3mm plate between the gantry plate and the extrusion. Make sure to add a 3mm spacer to the front idle plate so the frame remains square. The original idler bolt will need to be replaced or grinded down for the extrusion to fit properly. 

    The upper XY gantry is now mostly complete with the exception of the z motor mounts. This is currently in design and will be added to the next build log. 

  • The build out in theory

    simplisticlines08/13/2019 at 18:29 0 comments

    The ender 5 is based on 2020 & 2040 extrusions which are 20x20mm and 20x40mm V slot. Important to use V slot as standard T slot extrusions are not the same dimensions. The build plate style is a cantilever with its support rails on only one end. This design is just fine for small light weight parts but not possible for the 20kg+ weight which is possible to print with this large scale. In this build the building plate will be completely stationary and be integrated into the lower frame with the XY frame being in motion along the z axis. This has several advantages like all motors and electronics can be attached to the moving frame allowing for short wiring to motors and reduced electrical noise.  


    Source the applicable length for your build in this case the Y axis is going to use 2x 2040 extrusions because of the long length the weight of the X arm traveling the center would potentially bend but more on that fix later. The X axis will retain the stock size extrusion but could be made to any length also. This allows for only 2 extrusions needed for the upper gantry.  

    Z axis frame

    You will also need 4x extrusions for the z axis which currently is using 2x 10mm linear shaft. You will be replacing those with the 4x 2040 extrusions with the same gantry style plates as the stock Y axis

    Using Gantry plates is optional and can be replaced with linear shafts/ rails depending on your requirement. The 2040 extrusions have enough accuracy for this build and the gantry plates can be found from several sources for very little cost in comparison to industrial linear rails.

    Z axis motion

    Z motion normally is done with a lead screw which has many advantages and disadvantages. To start with the major disadvantage the stepper motor is holding all the build weight axially which will reduce the service life and be ruined unless you change the internal bearings to a thrust style bearing. This is not recommended as some motors can be damaged during disassembly.  The alternative is to use a belt driven Z axis. The major issue is if power loss occurred the motor will give way and potentially break something as the gantry comes crashing down. This can be solved by using a planetary gear set which mounts to the stepper motor. This has a larger holding force because the planetary gears are setup to be torque amplifiers meaning a single turn on the input shaft will turn less times then the output shaft. With a high enough ratio the internal magnetism of the stepper can hold without power. Additionally this rotates the force transaxial and into the planetary gear set so the motor will never see axial force and should have great service life. The biggest con to this setup is belt selection. Cheap belts without fiber or steel reinforcement will stretch over time and have a slightly lower service life. Luckily belts are very inexpensive and are easy to replace which is why this style of motion system has been selected. 

    Building surface

    Many build surfaces can be used to fit the material needs. In this case the best option is granite because it is stiff, heavy weight, handles moderately high temps without any expansion or bending, and can be sourced locally in every major city at a decent price if a remnant can be used. Because of the mass and thermal conductivity of granite the preheating process will be extended but the trade-off is if power loss occurred the heat will last for longer and allows for more time to save a build if a failure occurred. 


    The enclosure will be mounted to the frame adding stiffness and thermal insulation value. The load acting on the exterior is relatively low but needs to account for the XY motion during printing. The exterior can be almost any sheet of building material even wood has an ignition point well above the environment temperature. The simplest and cost effective option is reflective XPS foam and hardboard sourced locally. ...

    Read more »

View all 4 project logs

Enjoy this project?



Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates