Kiwi-M5 Mendel Battle Tank

From RepRap
Revision as of 23:46, 10 January 2011 by Sebastien Bailard (talk | contribs) (Yay, new machine!)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This page is a development stub. Please enhance this page by adding information, cad files, nice big images, and well structured data!


These are good resources for creating wiki pages.
Very Messy Example and Columbus 
--Sebastien Bailard 07:23, 10 September 2010 (UTC)
Crystal Clear action run.png
kiwi_m5_mendel_battle_tank

Release status: unknown

No image available.png
Description
Like Mendel but more tank-y
License
Author
Contributors
Based-on
Categories
CAD Models
External Link



Files - Coming Soon!

FILE ID# TYPE DESCRIPTION AVAILABLE FORMATS CREATED/RESERVED BY
Your-File-Name SOLID MODEL ASSEMBLY These are CAD files for the Solid Model Assembly .xml.zip, .stl.zip --Example User 12:00, Today's Date 20xx (UTC)
Your-File-Name CAD FILES FOR PARTS These are CAD files for each part. .xml.zip, .stl.zip --Example User 12:00, Today's Date 20xx (UTC)
Your-File-Name EVEN MORE FILES These are are even more files. .xml.zip, .stl.zip --Example User 12:00, Today's Date 20xx (UTC)|-
Your-File-Name SOLID MODEL ASSEMBLY This is the final finished machine N/A --Example User 12:00, Tomorrow's Date, 20xx (UTC)
Please edit this and click the red links to put in your own files! --Sebastien Bailard 08:34, 10 September 2010 (UTC) 


Bill of Materials/Parts List

  • Coming Soon!

This machine has been built with: - Bullet resistant material - Driven with five powerful 24 volt stepper motors - Power plant consisting of dual power supplies, 563 watt (main supply) and 150watt (heated bed) - Massive all-up weight of 25 Kg - Long wheel base (Top 2 threaded shafts are 710mm long) - Loaded and armed with 100metres of 3mm diameter filament ammunition. - Powerful direct drive filament loading stepper motor - 0.5 mm diameter howitzer extruder barrel. - Target area (Print) - X axis 200mm - Y axis 200mm - Z axis 110mm.



This machine is getting close to begin field trials for future battles. Have done some test printing using mother board 2 firmware.

Having acquired some off-cuts of 10mm clear polycarbonate sheets I thought this would be a good material to try and build a Repstrap type of machine.

Polycarbonate is what CD’s, and DVD’s are made of. Among other things it is also used for bullet resistant windows.

Polycarbonate is a very strong material, it can be easy cut with a hand saw. It can be machined drilled or cut provided machine cutting speeds are kept low to reduce temperature build up. Screw threads can easy be taped by hand.

Note: Polycarbonate is not normally cut using a laser cutter due to the high level of toxic fumes that is produced. It can be cut using a CNC water jet cutter.

Polycarbonate is not a cheap product to buy price is around NZ$460 per square meter for 10mm thick sheet pieces.

My aim was to build most of the frame parts from the 10mm Polycarbonate sheets with the least amount of machining. This means some parts will appear to be over engineered.

Due to the complexity of the design, a lot of the work has been, design and build as you go. This has meant some parts have been made and discarded until the design worked. Other designed parts were too hard to machine and a different design part was then made.

The 10mm polycarbonate worked out well in the end for mounting the standard Mendel axis support bearings. Trying to mount guide bearings at 120 degrees with only using 10mm thick material was going to be difficult, however I found that the bearings being 13mm diameter could be mounted at 90 degrees to each other, e.g. bearings could be bolted on the flat surface and then bearings could be bolted on the edge using 4 mm bolts, this then gives the bearings 1.5 mm exposure on each side of the 10mm sheet pieces. (See pictures of Y and Z axis).

I also found that the 4mm bolts that hold the 13mm bearings could be placed in a drill chuck and by using a metal file the thread could be filed away allowing the bearing to be finely adjusted.

Polycarbonate also has a degree of spring tension which can be used keep bearing against surfaces.

A lot of the parts were hand cut then machined in a drill mill. A drill mill is a bit like a bench electric drill press. It has a vice that can be moved by hand in X &Y axis direction. The chuck is replaced with a tool holder that holds slot drills that can be used to cut and shape the polycarbonate. If you use a chuck to hold a slot drill, the sideways force with cutting will cause the chuck to fall off the drill shaft. A mini lathe was used for the odd part - Z axis threaded drive shafts - this was to allow couplings to be fitted to connect the stepper motors, filament spool hub and the filament guides used in the filament spool hose connector to the extruder drive.

Frame

All of the Mendel frame brackets have been constructed using 2 pieces of machined 10mm Polycarbonate. This gives a 20mm thickness for each bracket. The 8mm threaded shafts nuts then hold these two frame parts together.

The basic Mendel frame design was increased at one end to allow room for two power supplies, (main power supply and a heated bed transformer). The additional frame length also allowed room to mount a spool of filament material as well as being able to mount the electronics and power on-off/reset switches away from the print area.

Increasing the Mendel frame length has increased the overall rigidity of the frame and added to the all up weight to counter vibrations from the movement of the different moving axis parts.

There is around 150 x 8 mm diameter Nylock nuts plus washers. The Nylock nuts were used more for looks, these were a pain to move along the threaded shafts, often I used an electric hand drill turning the threaded shaft and held the nuts with a spanner. Around 200 x 4mm diameter bolts of 3 different lengths have also been used. Some of the 4mm bolts required additional thread, this required using a 4mm die which was easy as the bolts already had some thread to start with. Currently I haven’t had time to document of all the parts/plans and quantities that have been used. Something I plan to work on in due course.

Power supplies

The mains power (230vac) is connected to the machine via cable lead plug-able power socket. Power is then fed to a power switch which is fitted to the front of the machine and then onto the two power supplies. An earth connection from the main power supply is connection to the Mendel frame and all threaded shafts are bonded together to reduce EMF interference. To bond threaded shafts a wire has been placed under the 8 mm nuts at frame junction areas to connect the threaded shafts together.

A 563 watt power unit that outputs +5volts, +12volts and +24volts has been used as the main power supply. All three voltage outputs go to a housing which contains 3 fuses to provide additional protection in the event of a electrical short circuit. The main power supply which has an internal cooling fan has been mounted to direct the air flow away from the print bed area. A exhaust duct has been installed to enable piping of fumes from the machine environment.

A 150 watt interior lighting transformer (240 volt ac to 12 volt ac) has been used for the heated bed. A 20amp solid state relay is used to switch the output of the transformer (12Vac) to the heated bed resistors.

Heated Bed

The Heated Bed is a 200mm x 200mm 4 mm thick aluminum plate fitted with 16 x 15 Ohm 10 watt resistors. The 16 resistors and sensor have been glued with thermal glue (Fisher Elektronik WLK5 Adhesive rated at 150C ). A Dallas one-wire digital sensor is used to monitor the temperature.

I had previously built this heated bed for a Darwin which had a separate controller. As I have run out of thermal glue and this glue is not cheap, I have used a PIC Microchip processor, programmed to convert the digital temperature from the Dallas one wire digital sensor reading to a DC signal that matches a thermocouple output. This is for the extruder processor to monitor and control the temperature of the heated bed. Planning to reprogram the extruder processor to accept one-wire input from the heated bed sensor in the future and remove this PIC processor work around.

The aluminum headed bed plate is positioned 6 mm off the Y axis polycarbonate plate with four screws and four PTFE spacers located at the plate corners . The Y axis polycarbonate plate that supports the heated bed, has been cut out in the middle to give clearance for the bed heat resistors and to provide air flow to stop the surrounding polycarbonate from heating up and wrapping. Three of the heated bed plate mounting holes on the Y axis polycarbonate plate have been slotted to allow thermal expansion of the heated bed.

Stepper Motors

Five 24 volt stepper motors have been used. (1- X axis, 1 - Y axis, 2 - Z axis, 1 - Extruder). Obtained these from a scrapped automatic cash machine (ATM). The large extruder one is the one that feeds your money out to you.

Two direct drive steppers have been used for the Z axis. Each of these stepper motors have their own Led home sensors as well as separated stepper driver boards. An additional PCB is used to allow theses two Z stepper motors to auto synchronize when being sent to their home position. Basically the stepper motors driver boards don’t receive step pulses if the LED sensor is blocked when the motor is being driven in a backwards direction. The Z axis home signal is only sent to the processor when both Led home sensors are blocked. This also stops the processor ramming the extruder heater barrel into the bed should the processor ignore the Led sensor signal for any reason.

All stepper motors cables (2 sets of twisted pairs) have an un-insulated wire that has been wrapped with tin foil and covered with heat shrink. This wire is also connected to the stepper motor and frame ground to shield electronics from EMF interference.

I found that mounting the Z stepper motors facing upwards (e.g. output shaft upwards) that the output shaft bearings were spring loaded. This could change the position of the Z axis, subject to the supported weight. (This only occurs with some types of stepper motors). The stepper motor bearing springs were replaced with fixed washers to keep an accurate Z axis position.

X Y Z Axis

The X Y Z axis are all guided with standard 13 mm diameter ball bearings at 90 degree angles. 12 bearing were used for the X axis. 10 bearing were used for the Y axis. 10 bearing were used for the Z axis.

2 larger bearings with ¼” shafts were used to on the X axis drive belt (opposite end to stepper motor). 2 larger bearings with ¼” shafts were used to on the Y axis drive belt (opposite end to stepper motor).

The Z axis threaded drive shafts lift nuts are designed to self align. Basically they are allow to move back and forth and left and right but not to rotate. This is to reduce any binding with misalignment of the Z drive shafts/stepping motors.

The X axis stepping motor has been mounted on its side to reduce the use of multiple bearings. A single continuous drive belt is used to move the extruder. Belt tension can be adjusted by moving the complete stepping motor assembly or the belt holding clamps on the extruder.

The Y axis stepping motor is mounted on the inside of the Mendel frame. Belt tension can be adjusted by moving the stepping motor.

Electronics

The electronics circuits are located on the back of the Mendel frame in an enclosed polycarbonate box. The box assembly is enclosed except for an opening along the bottom edge to allow electrical cabling and an upwards air flow.

A 12 volt PC CPU fan has been installed at the top of the electronics box to provide air movement from the electronics enclosure box. The 12 volt PC CPU fan is connected to the +5volt supply to provide a slow and quiet air flow. Used 2G electronics to test all 5 stepper motors and extruder heater. Used separate controller for heated bed.

PCB’s contained in enclosure. - Custom build Sensor and Z axis PCB. - Extruder processor PCB (A work in progress- Upgrade Mother board 2 to 3 work around – Non-surface mount components). Have installed two RS485 chips back to back to allow current firmware to work. Purchased new ATMEg168 processor. Found that you have to load a bootloader for the arduino environment. This required building a programmer to load the Arduino bootloader. Used a simple parallel programmer. Refer http:/www.arduino.cc/en/Hacking/Parallelprogrammer Found that you have to use a 16mhz crystal and 2 caps to get this to load a bootload . Still to be wired up and tested.

- Sanguino processor PCB. -Customer heater PCB, (Extruder and heated bed, heated bed driver goes to solid state relay. Extruder and heated bed temperature sensor signals pass thru this PCcool smiley. - Five stepper driver PCB’s Other design features: - Electronics components enclosed allows for greater protection from un-authorization contact with items that may short circuit components. - Modular construction – separated PCB’s allows quick and cheaper exchange parts replacement. - A barrier support has been fitted below the heater PCB to stop solder or parts dropping directly onto stepper driver PCB’s should a fault occur that results in a melt down. - A reset switch has been fitted to the front of the machines alongside the mains power switch. - Electrical cables have been wrapped in Tessa woven fabric tape for bundling cables.


Extruder Assembly

A powerful direct drive extruder has been used with a Pirated CupCakeMakerbot PTFE and 0.5mm barrel. A thermal coupler PCB interface has been mounted at the extruder assembly.

The use of a large extruder stepper motor and a large extruder area has been designed to allow future redesign and experimental work. Future plan is to try 0.3mm custom built extruder barrels and also switch-able multiple filaments (different colours) using just one extruder motor, e.g. different pinch wheels that can be engaged as required to feed different colour filament.

Spool assembly

A feed supply spool that can hold up to 100metres of filament has been fitted to the front of the machine on the Mendel frame extension. The spool being position at the front is to allow easy loading and viewing of filament feeding.

Two ball bearings have been used to proved easy feeding. The spool is tilted using a 6mm diameter bolt against the Mendel frame. The spool is a 6 spoke assembly. The filament is held with 6 x U shaped pieces. These U shaped pieces have been angled to be parallel to the base of the machine when position at the top of the spool. The theory behind this is to keep the filament better placed on the spool as it pulled off.

A piece of cable holding wrap with 5 nylon guides allows the filament to be pulled by the extruder motor to the print area. Heat shrink has been used to hold the nylon guides in place. The nylon guides which are barrel shaped have been counter sunk drilled to allow easy threading of the filament to the extruder pinch feed.

X, Y, Z Axis Movement

The X and Y axis has a full movement of 200mm x 200mm. The Z axis has 110mm movement. Installing a heated bed and the extruder reduced the normal Mendel 140mm Z axis movement.

By replacing 9 threaded rods (currently 390mm long) with longer ones, replacing the transformer mounting plate with a larger one and extending the stepper motor and sensor wiring would allow the full 140 mm Z axis movement. This is not something I am planning at the moment.

Once I sort out the upgrade to the electronics I will be looking at adding it to a wiki page.

No worries.  Give this one some love, eh?  --Sebastien Bailard 03:46, 11 January 2011 (UTC)