My 3d printer is tied up doing production runs so not yet had the opportunity to study cause and effect with different Skeinforge settings. Basically I’ve been studying live graphs from the production runs; checking for consistency between print jobs and during print runs. The graph below is a 100 second snapshot of a production run I was doing.

3D Printer Extruder Force Sensor Graph Sample

From the graph it is easy to tell when the printer is printing and when it is not. The drops are the retractions and you can see that the nozzle loses pressure after each retraction. The retractions also show a period of travel without printing, and the longer the travel – the more pressure is lost. Also, a bunch of retractions together shows a decline in nozzle pressure with each retraction.

The pressure sensor is preloaded with 200g of force so that the retractions can be absorbed by the sensor without bottoming out. This has not been accounted for in the graph and so 200g needs to be taken off the force sensor graph readings in this case.

By looking at the graph I would deduct that my retraction setting is not long enough. Even after deducing the 200g pre-load, there is  still a round 500g of force applied by the extruder.

Extruder Force Sensor – The Next Step

There is plenty of room for improvement on my 3d prints and I’ll be looking at the graphs to see if the readings can help identify what settings needs to be changed to get the best print quality. Also I’ll be testing the force sensor on printing simple shapes to better judge sensor reading consistency.

How do I get one of these sensor thingies?

I don’t want to reveal the set-up until I’ve got all the support documentation ready to publish. This is because I need the time to focus on the documentation that will allow others to replicate the sensor kit as easily as possible. If I reveal my set-up now, many people will try to build there own sensor kit and I could be bogged down with providing support.

Due to the low investment cost of the hardware and the potential it has for testing 3d printer extruders and hot ends, the force sensor gauge is likely to be very popular. So I’ll be looking to get the airtripper extruder filament force sensor properly backed up with good support documentation.

If you would like to speculate how the force sensor works and how it is fitted, use the comment section below.

Got myself a J-Head MK-IV Hot End clone, from Ebay (snipermand), to see if it will be good enough to replace my heavily modified Mendel Parts Hotend V9 clone; and since the J-Head MK-IV is a clone, this is my quick review to share my purchase experience. The review also includes an illustration of how the different components fit together, how the J-Head clone stacks up against the original J-Head and conclude whether Hot End clones are really worth considering.

The J-Head MK-IV will be the third Hot End clone I’ve purchased in the last 12 months with the hope of putting together a decent extruder system for 1.75mm PLA filament. The Hot End I’m using right now is not as good as it should be and it’s a Mendel Parts V9, the first clone I purchased, which I had to heavily modify due to a manufacturing or design error made by the supplier in India. The second clone I purchased was the MBE Extruder V9 from qu-bd.com, and like for many other users, it just would not work with the 1.75mm PLA filament. This Hot End kit is a Makerbot Stepstruder clone.

J-Head MK-IV Hot End Clone Design

The illustration above shows the clone version of the J-Head MK-IV Hot End and looks similar to the original J-Head supplied by hotends.com. The version I have is with the aluminium nozzle/heater combination which is also available in brass (J-Head MK-IV-B). The overall machining quality of this nozzle is very good and was delivered with the main components pre-assembled, and has some kind of red sealant locking the peek insulator on to the aluminium nozzle/heater. A resistor and thermistor was included with some wire, wire insulating PTFE tubing and connectors which all needed assembly.

The peek nozzle holder has perhaps gone through the most evolutionary changes during the life of the J-head which demonstrates the importance of cooling; and version four is used here, but with only four vents instead of the five as per original design. The extra machining to the peek nozzle holder has greatly increased the surface area to allow heat to dissipate more efficiently to allow better cooling for extruding PLA filament. The series of vents round the peek nozzle holder would allow a cooling fan to be effective at lower RPM speeds making the 3d printer much quieter during operation.

The aluminium nozzle/heater combination that make the J-Head, is probably the best and most reliable 3d printer Hot End design in it’s price range. You have what is commonly three separate components (nozzle, heater block, threaded tube), combined into one. With the machining required to produce the nozzle and peek nozzle holder together, it still manages to be the best value for money Hot Ends out there; with the important good usage track record. The aluminium version I have is based on the blueprints for nozzle/heater combination version one.

PTFE liner tube is used in the peek nozzle holder which is held between the nozzle and the set screw under pressure. The design uses the PTFE tube to create a good seal between parts and also provide an almost instant transition from cold end to hot end for the filament path. The PTFE liner also provides some heat shielding from the peak nozzle holder as well as the nozzle itself, however, some active cooling would be required for PLA filament to prevent the PTFE liner from heating up too high. An extra PTFE tube was supplied as a method to convert the J-Head to a 1.75mm extruder from it’s native 3mm design; a method not supported in the J-Head Wiki because the melt chamber is machined for 3mm filament and not 1.75mm.

J-Head MK-IV Clone v J-Head MK-IV

If I’d done my research properly I might have decided to get the J-Head MK-IV from hotends.com instead of getting it from RepRap.me through Ebay. With the various clones available you just can’t be sure that they’ve been manufactured using techniques that achieve the kind of precision and quality you would find in the original item. Basically, these Hot Ends are sent out to customers untested, and when ordering a clone Hot End, you are putting a lot of trust in the supplier, their understanding and knowledge of the product, and the quality of manufacturing.

Had I done some research on the J-Head MK-IV Hot End I would have realised that the 3mm to 1.75mm conversion method, applied to the J-Head clone, is not recommend. The conversion involved inserting another PTFE tube inside the existing PTFE liner as shown in the above illustration. The PTFE tube is used to reduce the melt chamber diameter size to 2mm. It’s not clear how well this conversion works or whether there are maintenance issues.

A 1.75 mm J-Head  Hot End ordered from hotends.com would have a melt chamber machined for 1.75mm filament rather than being reduced with PTFE tube from 3mm. This is likely to improve reliability over the Hot End clone version and have better control over to temperaature in the melt chamber.

Hot End Purchase Experience

As I noted earlier, I’ve purchased three Hot End clones in the past 12 months with the third purchase being the J-Head MK-IV clone from RepRap.me via Ebay (snipermand). I ordered the J-Head MK-IV Hot End with a 0.4mm orifice and everything looked in order on delivery, the build quality looked good as far as I could tell and it seemed like I made a good purchase. However, while working on the J-Head, to get it ready for extruding, I kept noticing the Hot End orifice looking a bit big for 0.4mm. So I took some close up pictures with the nozzle tip up against a ruler and added some reference points to the Hot End nozzle images with an image editor. The image above right shows the grey reference points added.

The reference points on the image was used to accurately gauge the size of the Hot End orifice against the ruler, the orifice size appears to be 0.5mm in diameter. To confirm the size of the Hot End orifice I found an electronic component with 0.5mm leads, confirmed with calipers, and I was able to insert them in to the orifice with a snug fit.

I’ve since sent a message to RepRap.me about the issue and still waiting for a reply. As it stands now I’m unlikely to use the nozzle since it does not meet the specification I wanted, and also unlikely to order another clone as this is the third clone failure to meet the specification claimed. Most people that order Hot Ends may not consider checking the nozzle size because the orifice is so small and take it on faith that they have been sent what they’ve ordered

Hot End Clone Conclusion

The J-Head MK-IV has few parts but a lot can still go wrong through poor quality manufacturing and poor assembly. Things like poorly tapped screw threads, PTFE liner not retained properly, PTFE liner ends not cut squarely or cleanly, nozzle orifice drilled larger than specification, and assembly errors caused by lack of product knowledge. It only takes a single fault to cause the nozzle to fail and in view of this, it would be better to put your trust in the original designer and supplier.

All in all it was a bad decision to purchase this Hot End clone due to the lack of backing from the supplier. The sale campaign was mostly backed up with a copy and paste from the J-Head wiki with an added thermistor table for only the Sprinter firmware. The thermistor supplied had no brand or type to identify it and would be difficult for the less than average 3D printer user to set up correctly in firmware. Also, the supplier did not declare any working experience or manufacturing process to back-up the reliability or build quality of the their Hot End clone design.

A 3d printer Hot End is what makes a printer a 3d printer and so is a critical component that needs to be right. So the advice would be to buy your Hot Ends from the original designer and manufacturer that is backed up with good documentation and support. Basically, if you are out to buy a J-head Hot End, get the J-Head from hotends.com. Supporting the original designer/manufacturer/supplier will help with further Hot End research and development.

I’ve just updated the Marlin firmware on my Sumpod 3d printer since I’m always keen to have the latest features and bug fixes. To be honest, I don’t mess with the firmware that much, and if it wasn’t for the configuration file from my last version, I would struggle to remember what sort of configuration I would need to set in the latest firmware version.

Anyway, while my latest experience with 3d printer firmware is still fresh in the mined, I’ll share some notes about what settings you need to know to get a basic Marlin firmware configured enough to get a 3d printer working. The notes will focus on the Marlin firmware v1 and will include setting up a click encoder and LCD panel. But before going straight into getting the Marlin firmware configured, I’ll first quickly introduce you to a handy tool called WinMerge.

WinMerge

For anybody that’s in the business of editing and configuring 3d printer firmware files such as the Marlin firmware, I would suggest downloading a copy of WinMerge. It’s free, open source software, and is cross platform, so the same tool will run on Windows and Linux.

You can use WinMerge to compare a clean version of your Marlin firmware against your edited version that you are using on your 3d printer. This will help to keep track and note all the changes made to the files that you might want to transfer to a newer firmware version.

You can open just two files to compare or you can open two folders to compare. Comparing 3d printer firmware folders will allow you to quickly spot which files that have been edited.

Marlin Firmware Basic Configuration

Some motherboards listed in the Marlin firmware configuration file may not have support for some of the options or features available to configure. The notes will be biased towards the Ramps 1.3 board, but the note will still be valid for any Marlin firmware compatible board.

This Marlin firmware will be useful to those who wish to configure their own 3d printer firmware and want information that expands on the comments already made in the Marlin firmware. The notes are a brief guide on what the settings are and how to use them, leaving it up to the person who is configuring the 3d printer to decide what the final settings should be. I don’t guarantee that the information in this Marlin firmware guide is accurate, however, if you spot any mistakes please leave a comment at the end of the post. The Marlin firmware guide is likely to be updated to improve the information where necessary after publishing.

For Marlin Firmware V1 you will need Arduino 0023 IDE to save, compile and upload to the 3d printer motherboard. In the following notes, to edit the firmware, I’ll be using line numbers to reference the location of the code in the Marlin firmware configuration file. Although there is no line numbering in the IDE editor window,  you will see the line number at the bottom left of the IDE showing the current cursor position. Just move the cursor to any line with the mouse to update the line number.

To get the 3d printer up and running you only need to configure one file in the Marlin firmware and that file is the  Configuration.h

Marlin firmware Configuration

So, to get started, open up the Marlin firmware Configuration.h in the Arduino IDE and work down the notes below. Use WinMerge to compare the changes to a clean non-edited version of Configuration.h for final review before uploading to the 3d printer motherboard.

Baud Rate – line 20

#define BAUDRATE 250000
//#define BAUDRATE 115200

Line 20 marks the start of the Marlin firmware configuration journey and this is where the baud rate is set to determine the speed of the communication between the printer interface software and the Marlin firmware. Two common baud rate options are defined with one, preceded with two forward slashes (//), commented out to disable. The default enabled option, 250000 baud rate, works well when used with Printrun/Pronterface and RepSnapper software.

Motherboard – line 49

#ifndef MOTHERBOARD
#define MOTHERBOARD 33
#endif

You will see a list of motherboards to choose from preceding the code snippet shown above in the Marlin firmware configuration file. The code snippet above defines the Ramps 1.3 as the motherboard, you can select a board just by changing the number to any board that’s on the list. The Marlin firmware will manage circuit board pin assignments to match the motherboard you have selected. Pin assignment details for each motherboard type can be found in the pins.h file of the Marlin firmware.

Marlin Firmware Thermal Settings

Now we come to the thermal settings section of the Marlin firmware where things get a bit more complicated, however, we don’t need to touch the complicated stuff to get the printer up and running. Basically all we need to do is select a temperature sensor type for each of the sensors installed on the 3d printer. If you did not install the temperature sensors yourself, you may have to investigate what sensors you do have so that you can select the best match from the list. More about thermistors on the RepRap wiki.

Thermistor – line 78

#define TEMP_SENSOR_0 1
#define TEMP_SENSOR_1 0
#define TEMP_SENSOR_2 0
#define TEMP_SENSOR_BED 1

If you are looking at the Marlin firmware configuration file you will see a list of temperature sensor types preceding the code snippet shown above. The code snippet above is setting up the 3d printer that features one hot end and a heated build platform. The temperature sensor selected for both features is 100k thermistor.

If you are using the Ramps 1.3 motherboard with the default pin.h file in the Marlin firmware, the motherboard connectors T0 and T1 will be enabled for the hot end sensor and the heated bed sensor. Changing the 1 to a 0 will disable that sensor. Change the number to select the best match for your sensor from the list.

Maximum Temperatures – line 99

#define HEATER_0_MAXTEMP 275
#define HEATER_1_MAXTEMP 275
#define HEATER_2_MAXTEMP 275
#define BED_MAXTEMP 120

Some hot ends and heated build platforms might have a maximum temperature rating much less than the default settings in the Marlin firmware, reducing the default maximum temperatures will help avoid accidental damage to the 3d printer if set too high in the interface software.

Marlin firmware Mechanical Settings

The Marlin firmware Mechanical Settings section will be about configuring End Stops, Stepper Motors, Build Platform Printable area and Steps Per Unit.

End Stops & Pull Ups

This section will be about configuring end stops in the Marlin firmware that are the limit switches for each axis on the 3d printer. Issuing a homing command from the interface software will cause the 3d printer to mechanically move each axis towards the end stop until the limit switch is triggered.  ENDSTOPPULLUPS will need to be defined where you have limit switches that don’t supply a voltage to the signal pin to generate a digital 1. Enabling pull up resistors will ensure that the end stop signal line will read a digital 1, and when the signal line is shorted to ground by the limit switch, you get a digital 0.

For more information about end stops please refer to the RepRap wiki for Mechanical Endstop, OptoEndstop 2.1 and Gen7 Endstop 1.3.1.

Coarse End Stop Pull Up Resistor – line 194

// corse Endstop Settings
#define ENDSTOPPULLUPS // Comment this out (using // at the start of the line) to disable the endstop pullup resistors

In the Marlin firmware ENDSTOPPULLUPS is defined by default, and ENDSTOPPULLUPS for each end stop connector on the motherboard are enabled individually from line 207. However, commenting out line 194 will only disable ENDSTOPPULLUPS that are also commented out optionally for each connector from line 197. Having this kind of fine tuning makes it easier to configure different types of end stops connected to the 3d printer. You may have mechanical end stop switches for axis X and Y that need pull up resistors enabled and optical end stops that don’t need pull up resistors enabled.

Fine End Stop Pull Up Resistor – line 196

#ifndef ENDSTOPPULLUPS
// fine Enstop settings: Individual Pullups. will be ignord if ENDSTOPPULLUPS is defined
#define ENDSTOPPULLUP_XMAX
#define ENDSTOPPULLUP_YMAX
#define ENDSTOPPULLUP_ZMAX
#define ENDSTOPPULLUP_XMIN
#define ENDSTOPPULLUP_YMIN
//#define ENDSTOPPULLUP_ZMIN
#endif

If ENDSTOPPULLUPS at line 194 in the Configuration.h file of the Marlin firmware is commented out, then the code snippet above, starting at line 196, will execute. This code snippet will allow you to configure individual pull ups for each end stop connector on the motherboard. You would comment out a define statement for an end stop where you don’t need a pull up by preceding the line with two forward slashes. You would normally need to enable pull ups when using mechanical end stop type switches that just simply sink the signal pin to ground on the motherboard end stop connector.

Inverting End Stops – line 216

// The pullups are needed if you directly connect a mechanical endswitch between the signal and ground pins.
// set to true to invert the logic of the endstops.
const bool X_ENDSTOPS_INVERTING = false;
const bool Y_ENDSTOPS_INVERTING = false;
const bool Z_ENDSTOPS_INVERTING = false;

For a mechanical end stop that is wired as normally connected (NC), a digital signal of 1 will be read when an axis is in contact with the limit switch. In this case End stop inverting needs to be false. End stop inverting needs to be true if the mechanical end stop is wired as normally open (NO).

For optical end stops, you will need to refer to the suppliers specification or a previous configuration.h file like the one in the Marlin firmware.

You should test the homing command while the axis is positioned at the centre of the travel distance. If you find that the axis won’t move when the home command is sent, then you may have the logic incorrectly configured. Be ready to reset the motherboard or turn off the power to avoid axis crash. Before making any test, complete the Marlin firmware configuration as much as possible. If you can reach the end stops easily, you can trigger them early before the axis has completed it’s travel for a safe test.

Disable Maximum End Stops – line 219

//#define DISABLE_MAX_ENDSTOPS

The above code snippet found on line 219 of the Marlin firmware Configuration.h file is commented out by default, which allows homing axis to end stops where end stop switches are connected to the X+, Y+ and Z+ connectors on the motherboard.

It is common to have the Z axis end stop switch connected to the Z+ connector in order to home the Z axis away from the hot end. In order to do this, line 219 needs to be commented out. Line 219 may behave differently for boards that don’t have X+, Y+ & Z+ end stop connectors.

Disable Axis – line 228

// Disables axis when it’s not being used.
#define DISABLE_X false
#define DISABLE_Y false
#define DISABLE_Z true
#define DISABLE_E false // For all extruders

Normally the above code snippet would not be changed in the Marlin firmware and all the settings would be set to false by default. However, if your 3d printer has a Z axis handle fitted like my 3d printer you might want to disable the Z axis so that the stepper motor can be turned by the Z axis handle while the 3d printer is printing. I’ve often made a Z height adjustment to fine tune the gap between the nozzle and the build bed as the first layer begins to print.

Stepper Motors

We have reached the part in the Marlin firmware configuration file where you configure stepper motor rotation direction, end stop direction, travel limits and steps per unit. As long as the end stops are configured correctly, the following settings should be easy enough to sort out during testing.

Stepper Motor Rotation direction – line 233

#define INVERT_X_DIR true
#define INVERT_Y_DIR true
#define INVERT_Z_DIR true
#define INVERT_E0_DIR true
#define INVERT_E1_DIR false
#define INVERT_E2_DIR false

This is where we decide which direction each axis will go when we control the 3d printer through the interface software. When we send a command to move +10mm on an axis, we expect the axis to move 10mm in the direction expected. The initial stepper motor direction can be difficult to predict without switching on the printer and performing a test. So I would suggest leaving these settings till last and complete the rest of the Marlin firmware configuration before proceeding with the test.

Once the Marlin firmware Configuration file is configured enough to operate the 3d printer, you can perform a test to check that each axis move in the correct direction. Set each axis midpoint of their full travel distance and then switch on the printer. After connecting to the 3d printer through the interface software such as Printrun/pronterface, test each axis by jogging them 10mm in the positive direction. The stepper motor rotation direction for each axis can be corrected from line 233 in the Marlin firmware configuration by changing the logic.

You would need to test the extruder stepper motor direction as well. This can be done without filament loaded and bringing the hot end temperature up to 175 degrees so the Marlin firmware will allow extrusion. Send an extrude command through the 3d printer interface software and observe the direction the filament drive gear pulley rotation. The rotation direction for the extruder can be corrected from line 236 by changing the logic.

When performing axis homing and axis jogging for the first time, it is important to be ready to press the reset button or be ready to turn off the power to avoid axis crashing. Home each axis separately to confirm that the axis is homing towards the end stop. If you can reach the end stops easily, you can trigger them early before the axis has completed it’s travel for a safe test.

End Stop Home Direction – line 242

// Sets direction of endstops when homing; 1=MAX, -1=MIN
#define X_HOME_DIR -1
#define Y_HOME_DIR -1
#define Z_HOME_DIR 1

Basically you tell the Marlin firmware which end of the axis the end stop switch is located. It is common for X and Y axis to home the hot end to the Zero location and Z axis end stop home to the maximum positive location. The code snippet above and the image to the right shows that configuration.

Some motherboards like the Ramps 1.3 have a connector for each end of each axis, totalling six connectors. You will need to be sure that the settings above match the end stop connections to the motherboard.

Travel Limits – line 249

#define X_MAX_POS 130
#define X_MIN_POS 0
#define Y_MAX_POS 130
#define Y_MIN_POS 0
#define Z_MAX_POS 107
#define Z_MIN_POS 0

The above defines the printable area of the 3d printer after homing. For the X and Y axis you just measure the travel length of the nozzle from the home position. The maximum travel length will either be restricted by the size of the build platform or by the maximum travel distance of the axis.

When setting up the Z axis for the first time, it is best to set the Z axis travel length a bit shorter than what is measured until a software and hardware test of the 3d printer is completed. This will help to avoid accidentally crashing the build bed into the hot end during initial tests. The Z axis travel length can be fine tuned later after the tests have been satisfied.

The measurement units are in millimeters and are defined from line 249 for the maximum positions in the Marlin firmware configuration file. The minimum positions can be left at the default 0 for this configuration.

Steps Per Unit – line 275

#define DEFAULT_AXIS_STEPS_PER_UNIT   {106.76, 106.76, 800, 48.14}

Configuring steps per unit will be one of the last bits of fine tuning you do before you start printing for the first time. Calculating steps per unit accurately will give the 3d printed parts the best start possible. However, if you are just upgrading the Marlin firmware or upgrading from any other firmware, you can get the steps per unit figures from the old configuration file if you still have it.

Steps per unit means the number of steps a stepper motor has to turn to equal 1mm of axis travel. The units at line 275 of the Marlin firmware configuration.h file are in the order of X, Y, Z and E. Getting the steps per unit from another printer of the same design would be very close to what you need, and will help you run some stepper motor tests before you get down to calculating accurate steps per unit for the Marlin firmware configuration file.

The code snippet above shows the steps per unit for my Sumpod 3d printer and is not the default setting in the Marlin firmware configuration.

Steps Per Unit Calculation

To calculate steps per unit (mm) to put in the Marlin firmware configuration file, you need to find a good way to measure axis travel distance accurately. You may have to temporary remove the hot end so that the filament can be extruded in order to make measurements. Using a dial indicator in place of the hot end and a ruler taped to the bed can provide a good accurate way to measured travelled distance. To get the best accuracy you should sample at least 100mm of axis travel. You command the printer to move the chosen axis 100mm using an interface software such as Pronterface. You then measure the actual distance the the axis travelled. Using the formula below you can calculate the new steps per unit.

Steps Per Unit Formula

NewStepsPerUnit = SampleTravelDist / ActualTravelDist x OldStepsPerUnit

You then repeat the above formula as many times as necessary until the commanded travel distance matches the actual distance travelled, using the NewStepsPerUnit as the OldStepsPerUnit each time.

Marlin Firmware Basic Configuration – The End

At this point, you have done enough configuration in the Marlin firmware and can now start 3d printing. If you are interested in LCD display and click encoders, read on.

Marlin Firmware additional Features

The last section in the Marlin firmware configuration file is for additional features, this section allows you to configure some of the optional extras you might have attached to your 3d printer. For the purpose of this guide I’m just going to include notes for the LCD 16×2 and the LCD 20×4 with click encoder control panel. The RAMPS 1.3 Arduino shield and my Marlin firmware configuration will be used for this guide.

Enabling an attached 16×2 LCD or click encoder control panel is straight forward in the Marlin firmware. However, the pin assignments for the attachment connectors need to match those in the Marlin firmware pins.h file. You can check if the LCD and click encoder panel features are supported for your motherboard by looking through the the pins.h file of the Marlin firmware. If you are just updating the Marlin firmware you can check for pin assignment changes by comparing your old pins.h configuration file with the new version of that file. Any changes found can be used to update the latest version of the Marlin firmware.

Adding LCD Support – line 303

#define ULTRA_LCD

To enable any type of LCD support in the Marlin firmware, ULTRA_LCD needs to be defined. To add support for 16×2 LCD display, line 303 needs to be uncommented like the code snippet above, by removing the preceding forward slashes. By default, the Marlin firmware does not have LCD or click encoder control panel enabled. Enabling line 303 and nothing else will give you support for 16×2 LCD screen that is connected to the motherboard pins as assigned in the pins.h file. Pin assignments are found in the pins.h file of the Marlin firmware under each supported motherboard type.

Adding Click Encoder Support

There are two popular types of click encoder control panels that can be enabled for the RAMPS 1.3 board. The first type is the Ultipanel, which can be found on Thingiverse, and the other type is the RepRapDiscount Smart Controller. For this guide I’m just going add notes for the Ultipanel since the other type is supported by RepRapDiscount RepRap wiki for the Marlin firmware.

Enabling ULTIPANEL – line 307

#define ULTIPANEL

Enabling NEWPANEL – line 331

#define NEWPANEL

Both lines 307 and 331 are not enabled by default. To enable 20×4 LCD display and click encoder, uncomment both lines by removing the forward slashes. This would also enable SD Card support as well, a feature that will be covered in another guide. It will not be necessary to uncomment line 303 if line 307 is enabled by uncommenting. ULTRA_LCD will automatically be defined when ULTIPANEL is defined.

Marlin Firmware Guide – The End

Well that concludes the Marlin firmware guide for now, and I would expect to be making updates going forward to correct errors if any are found or just to improve certain aspects of the guide.

If you have any questions or comments about the Marlin firmware guide, please leave them below. However, if you need Marlin firmware support, this is perhaps not the best place to get it.

The Airtripper’s Direct Drive Bowden Extruder is now at version 3 with the design files ready to download from Thingiverse. A lot of work went into the design to improve the usabillity and the look of the extruder. The design is stronger with a much cleaner 3d printed finish, and filament changing is now much easier than before.

The bowden extruder was originally designed to fit the Sumpod 3d printer to replace the bulky MDF extruder housing that was awkward to use. However, the bowden extruder can be used for other 3d printers making use of it’s simple bracket, and the extruder has been popular with the Rostock delta 3d printer. A tube bracket is now availble for attaching to the bowden extruder to help guide the filament from the filament spool. more about that here at Sumpod 3D Printer Filament Handling for Bowden Extruder.

You will find more about this bowden extruder on the following page: Bowden Extruder Upgrade Part 3.

The popularity of The Airtripper’s bowden extruder was boosted when the extruder was included in the development of the awesome Rostock 3D Printer (delta robot 3D printer). To see the bowden extruder in action on the Rostock, watch the youtube clip below.

Click here to view the video on YouTube.

Bowden extruder V3 Update Details

A number of updates were made to the bowden extruder design with some minor updates on the strut and idler housing.

To get a better fill between perimeters around the screw holes, the rounded ends on the strut have been increased to 10mm in diameter to improve print quality on wider 3d printer settings.  A small taper was added to the edge of the idler bearing housing to make it easier to slide the rubber tube over the preloader hook.

1. The filament drive gear housing has been filled-in to improve overall print quality by minimizing stringing.
2. Filament guide funnel size increased and angled for (a) a more gentle filament bend around the drive gear and for (b) easier filament threading.
3. Holes opened up and angled to give the idler preload screws better clearance from the idler housing hooks.
4. M4 nut for bowden cable (PTFE tube) now drops in to position much easier than before, also, reduced filament snagging when threading into the bowden cable.
5. Screw column housing lowered and widened at the top to minimize shrinkage and deformation. The M3 screw now drop in without drilling out, although a 25mm screw is now required instead of a 30mm. However, a 30mm screw will fit with a washer so the screw cap does not drop into the recess.
6. Holes removed from the fixing bracket to improve overall 3d print quality.
7. The filament in-feed bracket is re-designed for a much cleaner look and is now attached to a screw column for added strength.
8. A spacer is added so that extra washers are no longer needed.

Bowden Extruder V3 Parts List & Files

Most of the items below can be acquired cheaply from Ebay. The MR105ZZ Ball Bearing is optional but recommended, and the Rubber Diesel Hose can be replaced for coil springs. The M6 nuts, bolts & washers are for attaching the extruder to the Sumpod 3d printer or any other printer with similar fixing arrangement, just decide what length of bolts you need.

Extruder 3D Design Files

All the files for this extruder project are on Thingiverse for download. I’ve Supplied STL files that combine selected 3d objects for printing in one session. This is good for the smaller 3d objects because the printed layers will be given more time to cool. You can download the files from here:

http://www.thingiverse.com/thing:35404

Special Parts

• Nema 17 Stepper Motor. Since this extruder is a direct drive type, a powerful stepper motor will be needed. Point your browser to http://reprap.org/wiki/Stepper_motor for a good source of stepper motor information.
  
• Filament Drive Gear. For direct drive extruders, I can only recommend the MK7 drive gear at this point. If this does not work, you probably have Hot End issues. Starting with a tried and tested drive gear will help with your extruder system trouble shooting.
  

Screws, Nuts & Washers

• 1 X   M3 x 25mm S/S Cap Screw Allen Bolt.
• 2 X   M3 x 30mm S/S Cap Screw Allen Bolt.
• 2 X   M3 x 45mm S/S Cap Screw Allen Bolt.
• 1 X   M3 x 6mm S/S Button Head Allen Bolts.
• 3 X   M3 Stainless Hex Full Nuts.
• 1 X   M4 Stainless Hex Full Nuts.
• 2 X   M3 washers.
• 2 X   M6 S/S Hex Head Bolts.
• 2 X   M6 S/S Flat Form B Washers.
• 2 X   M6 S/S Hex Full Nuts.

Ball Bearings

• 1 X   608 ZZ [8 x 22 x 7] Roller Skate Ball Bearings.
• 1 X   MR105 ZZ Model Miniature Ball Bearing 5 x 10 x 4mm.

Tube

• 1 X   PTFE Tube 4mm x 2mm.
• 1 X   1/4″ 6mm id Rubber Diesel Hose Tubing Line.

Printing the Bowden Extruder

As a guide for printer set-up, I’ll list some of the settings used to compile the G-code. The toolchain I normally use includes OpenSCAD, Skeinforge, Printrun/Pronterface and Marlin. The settings listed below will be those used in Skeinforge, just the notable settings are included that works for this 3d print. Printing thicker layers and adding extra shells could create gaps in some surface layers, especially around screw holes that are close to a surface edge.

• Carve: Layer Height = 0.25
• Dimension: Filament Diameter = 1.75
• Fill: Extra Shells on Alternating Solid Layers = 2, Extra Shells on Base Layers = 1, Extra Shells on Sparse layer = 1, Infill Solidity (ratio) = 0.25

The STL files should be all you need to print off the extruder successfully, and I’ve provided extra STL files that will allow you to print a pair of selected items or all the items in one go.

3d model images have been uploaded to Thingiverse to help identify which STL files have multiple objects in them. The STL files rendered to images by Thingiverse don’t clearly show the files with multiple 3d objects.

Assembling the Bowden Extruder

The original assembly instructions are still valid for this bowden extruder update and you can find it here: Extruder Upgrade Part 3.

To add to the original instructions

The bowden extruder V3 bill of materials is slightly different from the last version because of a couple of small changes made to the extruder main body. However, all the parts used to assemble previous versions of the extruder will still fit the new version without buying new parts.

As mentioned above (bowden extruder update details), an M3 x 25mm screw is now required for one of the screw posts for attaching the extruder to the stepper motor. However, an M3 x 30mm screw can still be used if a washer is added to the screw before inserting in to the screw column. This will shorten the screw enough to fit the stepper motor body.

The bowden extruder features an axle bearing support for the optional MR105 ZZ ball bearing to spread the load on the stepper motor shaft. Because of variations in 3d printer outputs, it may be necessary to add a bit of extra diameter to the ball bearing. When assembling the bowden extruder, take note of the amount of contact between the ball bearing and the bearing support. If you don’t think there is enough contact, try a piece of electrical insulation tape to add some diameter to the ball bearing. Add as many tape layers as needed to get good contact.

Some types of filament drive gears, after lining them up with the filament on the stepper motor shaft, will leave a gap between the ball bearing and the drive gear. For proper stepper motor shaft support, the ball bearing should be positioned at the end of the shaft. This position can be maintained by adding M5 size washers to fill the gap between the bearing and the filament drive gear.

Related Topics

Sumpod 3D Printer Filament Handling for Extruder
Bowden Extruder Upgrade Part 3
Bowden Extruder Upgrade Part 2
Bowden Extruder Upgrade Part 1

Bowden Extruder Gallery

The more improvements and new features added to the Sumpod 3d printer the more I want to use the 3d printer, and adding better filament handling has gone a long way to improve the 3d printer’s ease of use.

The Sumpod’s sturdy construction has allowed me to set up a filament spool rack on the top of the printer, and adding filament feed brackets to guide the filament round to the extruder keeps friction to a minimum during the printer’s operation. This set-up will go a long way to reduce the printer’s set-up and shutdown time because the filament spool can now be left at the printer.

I’ve made the design files available for download from thingiverse should anybody want to use them. The design files might not suit all Sumpod 3d printer configurations, but the designs should provide inspirations to those looking to improve their own filament material handling.

Sumpod 3D Printer outside – are you mad!

Well, to get the best clear pictures, I make the effort to get the Sumpod outside. I have to make sure it’s a dry day though because getting the MDF case damp might upset the printer’s build platform levelling . I’ve got more features and improvements lined up for this printer so it looks like I’ll be taking it outside a few more times yet.

And the Problems Before - Basically, I had to put the filament spool where I could get it, and that was mostly on the floor in front of the printer. The spool got kicked over a few times due to lack of space and people walking past, and when done printing, I had to remove the filament from the printer and put away the spool until needed next time. I also had to turn the printer side ways facing so that the extruder was a bit more in line with the filament spool. Which made it difficult to check the LCD screen on the front of the printer.

The solution to these problems will allow me to keep the filament set-up on the printer and have a permanent place for the Airtripper’s pocket reel rollers. Having a spool rack on the Sumpod will allow the printer to stay loaded with filament reducing set up and shutdown times.

Sumpod 3D Printer Spool Rack & Guides

Spool Rack - The spool rack legs stand on the ridge just inside the top edge of the Sumpod’s outer casing, and anchored down with small screws. There is an option to have four screws to anchor but I’ve just used one in each leg here. Once the anchor screws are in place there is no need to remove them to take down the spool rack, just slacken off the screws a bit and move each leg inward to remove or to place.

The spool rack shelf is just a scrap piece of 8.6mm ply measuring around 255mm by 100mm, basically the size to fit snugly inside the recess in the top of the Sumpod (after the legs are fitted), and the size to fit the Airtripper’s Pocket Reel Rollers. Having the spool rack stand inside the recess prevent sideways swagger, improving stability. M4 Wing nuts are used to attach the 100mm tall legs to the ply shelf which allows for easy and quick packing for transportation. Due to vibrations from the Sumpod during operation, it was necessary to fix the reel rollers to the shelf to avoid spools or reels toppling over the edge.

Filament Guide - Now that the 3d printer has a spool rack, I needed to set up a filament guide for the extruder driver on the back of the printer. The plan was to use existing fixtures to avoid drilling new holes or making new screw holes in back of the case, spoiling the paintwork.

I suppose any tube that has very low friction properties will do for the filament guide, I used PTFE tube since I have plenty to spare. Tube brackets are in place to hold the PTFE tube in position to guide the filament round to the extruder from the spool. Without a guide, the filament is at risk of folding or breaking when pulled round sharp bends.

I designed two tube brackets, one to fit on to the Airtripper’s Bowden Extruder and one to fit a case fixing bolt on the top corner of the Sumpod 3d printer. The 4mm o.d. tube is in two pieces where one piece fits between the brackets while the other is used to guide the filament in to the first bracket I call the in-feed. The out-feed bracket is the one attached to the extruder driver.

Conclusion

The spool rack is working very well and it is wide enough to hold more than one filament spool. However, narrow spools are at risk of toppling if the printer is used in an area where it could get disturbed, like people bumping in to the table that the printer is on. For a more secure set up, a spool rack could be made using screw rods that attach to the spool’s hub, similar to whats already out there but made to sit on top of the Sumpod 3d printer.

I can’t guarantee that the spool rack is fit for purpose and regular checks may be necessary. If I had to print these again I would make the rack legs a bit thicker and more robust, and I would also add another 20mm to the height to give the Hot End bowden cable more headroom.

The Files

Get the files from here: thingiverse

Sumpod 3D Printer Gallery

I’ve designed and printed a bracket for attaching a Dial Indicator (purchased on Ebay) to the 3d printer for accurate build platform levelling. I can now test to reduce inclines on the heated build bed to a very small fraction of the 0.25mm layer height I normally print at.

I have made the design files for the bracket available for download and the link to the files can be found towards the end of this post. The OpenSCAD file can be edited easily to make the bracket fit different 3d printers. 

Levelling Without the Dial Indicator

I had a problem with getting large 3d printing projects to work due to the 3d printer build bed not being accurately level. I say accurately level because there is very little tolerances when printing layers as thin as 0.25mm and less, and the larger the print project footprint on the build bed area, the smaller the tolerances acceptable. While I did not have a Dial Indicator, I was using a business card of some sort for bed levelling, and this worked well enough for most printing projects, especially those with the smaller footprints on the build bed area. So to improve 3d printer build bed levelling accuracy, I decided to go with the Dial Indicator.

The Symptoms Of Poor Levelling - One of the more serious symptoms is caused when the Hot End nozzle gets too close to the build platform while extruding plastic up an incline. The extruded plastic spreads sideways from the nozzle tip causing ridges to form in the first printed layer. The ridges usually form towards one edge or one corner where the incline of the build platform is at its highest.

If ridges are formed, the first printed layer will look like a ploughed farmers field but messier, with ridges getting deeper towards one side. You usually get large clumps of plastic build up along the edge of the problem area. If the ridges are formed high enough, the Hot End nozzle could collide with the ridges which can cause axis motion to be disrupted and cause subsequent layers to be printed out of alignment. Using a Dial Indicator to level the 3d printer build platform would help to avoid those ridges and clumps of plastic build up.

The Dial Indicator

With the Dial Indicator (sometimes called Dial Test Indicator) fitted, I can now level my heated build platform with some accuracy. It does mean removing the Hot End from the 3d printer to fit the Dial Indicator but I don’t expect to have to do this too often, and I will only need to level the platform again if symptoms occur or I’ve crashed the Z axis.

Specification - The Dial Indicator has two dials on the face, the large dial measures at a 0.01mm resolution while the small inset dial measures at 1mm resolution with a range of 0-10mm. The numbers on the large dial can be rotated so that zero can be set at any location around the face. Some dials come with a pair of markers that attach on to the outer edge of the Dial Indicator face so that a more visible tolerance range can be set for test measurements. Outer diameter of the dial is 58mm, while the probe length is 43mm.

Levelling The Platform - The Dial Indicator bracket default dimensions will allow it to fit the Sumpod (MDF original) easily after removing the Hot End. The dial bracket dimensions can easily be altered, in OpenSCAD software, and printed to fit other 3d printer makes. After fitting the the Dial Indicator, adjust the Z axis to bring the 3d printer build platform into contact with the Dial Indicator probe. Adjust the z axis so that the dial reads at least 2mm to allow for negative readings when moving the probe across the platform.

The next step is I choose a corner of the platform to set the probe and adjust the level height I want on the platform level adjuster to set as a template for all the other corners. I set the Dial Indicator to zero by rotating the outer dial and proceed to level one edge of the platform. The Dial Indicator probe is sensitive to imperfections of the platform, the bearings and the rails as the probe moves across from corner to corner. So I try to get the probe as close to the adjustment screws as possible before making adjustments, and I move the probe back and forth until the measurements match as closely as possible. I then proceed to level the next edge. I find that it is better to get the platform roughly level first before levelling for accuracy.

Dial Indicator Bracket

The Dial Indicator bracket is my first attempt at scripting a parametric 3d model in OpenSCAD, so there is a bunch of variables at the top of the script that can be changed to alter the dimensions and characteristics of the bracket. The default values configures the bracket to fit the original Sumpod MDF 3d printer but the values can be change to make the bracket fit other printers.

Parametric OpenSCAD File - To make any changes to the Dial Indicator Bracket you will need the OpenSCAD 3d modelling application to open the file. Once you open the file and view the script, you will see a list of adjustment variables, one set of variables for the bracket base and one set for the Dial Indicator. There is a short description that will provide a clue of what each variable is for, pressing F5 on the keyboard after changing a variable will redraw the 3d model. All the values are in mm, some values will either add or subtract while other values are actual measurements. Experimenting with the values and pressing F5 while viewing the 3d model will give an idea of what has changed.

When the Dial Indicator bracket is ready to be saved to STL, press F6 to compile and render the 3d model and then select Export as STL under the Design menu. Save the STL file to a directory and use your favourite program to convert the file to G-Code.

Find the files here for the Dial Indicator bracket here: thingiverse

Dial Indicator Gallery

Hope you enjoyed the Dial Indicator solution, leave a comment below if you have issues with printing the bracket or working with the OpenSCAD file.

So, this is an introduction to my latest 3d printer extruder system with a detailed view of the Hot End, Cold End and Nozzle. There are plenty of pictures and a detailed illustration that shows details about the 3d printer extruder system I’m currently using. I explain some of the pros and cons, and explain why the latest extruder system I’m using works.

I’m still using a Twin Drive Extruder System I developed to push the 1.75mm Polylactic acid (PLA) filament into, what used to be, a very stubborn nozzle. However, forcing the filament into the nozzle was not the answer and some investigation work needed to be done to make the system work better. A new Hot End is purchased and after much tweaking, the extruder set-up is now working as well as it can be and I should be able to revert back to the single filament drive extruder upgrade, freeing up a stepper motor. I’ve got a new extruder stepper motor drive gear coming from the US which should provide improved grip on the filament giving more pushing power with a single stepper motor.

What was

The Hot End has caused the most frustrations and headaches during the 3d printer ownership, and at first, I was not sure if the Hot End was at fault or the fault was with some dodgy PLA filament. It seemed that some types of PLA filament extruded better than others, but I still had performance issues with them all. Rather than build a collection of PLA filament that failed to extrude, I decided to develop a set-up that was less fussy about extruding different PLA filament types.

After a number of different extruder mash-ups with some endless tweaking and putting new bits together, I finally have a 3d printer extruder system that works. Through tweaking, the Hot End part of the extruder has increased in sophistication due to having better nozzle heat control and active cold end cooling. This has allowed for better filament management during it’s journey through the 3D printer extruder system that is fitted to the Sumpod 3D Printer.

About The Hot End

Where it’s from

My latest Hot End is a derivative of the Mendel Parts V9. The parts kit I got, shipped from Make Mendel in India, was supposed to be a Mendel Parts V9 copy, but there was an error in the main Peek housing that allowed the tubes to connect together without a thermal barrier between them, this meant the kit could not be used without a fix or part swap. Instead of returning the Hot End kit, I decided to use the parts to build my own derivative version.

How it works

The basic operation of the extruder system is to feed the filament, using a stepper motor drive gear, into the Hot End melt chamber to extrude melted plastic out of the nozzle tip. In order to achieve good extrusion performance for best 3d print quality, Some conditions in the 3d printer extruder system need to be controlled.

The Hot End has two chambers ( M6 threaded tubes), one melt chamber and one cold end chamber. The chambers are separated by a thermal barrier so that each chamber can be controlled to maintain separate temperature targets. The melt chamber is heated to the point where it melts the filament to a level that can be extruded with minimum pressure without the plastic burning. The cold end chamber, to avoid jamming, prevents the softening and swelling of the filament. A fan and heat sink is attached to the cold end chamber to keep the heat off the filament until the filament reaches the melt chamber. If the filament softens in the cold end chamber the filament will swell and become jammed under pressure from the extruder stepper motor drive gear.

Due to PLA’s relatively low glass transition temperature, the heat sink cooling fan needs to be switched on during 3d printing. Without the fan, the cold end becomes very hot which could lead to filament jamming. The Hot End is capable of extruding 1.75mm PLA at temperatures up to 230 degrees C without changing the glass transition of the plastic in the cold end.

Hot End Pros and Cons

Pros

• 1.75mm PLA filament can be extruded at temperatures as high as 230 degrees C.
• The Hot End reaches the target temperature easily because heat transfer to the cold end is kept low by the PTFE thermal barrier.
• Filament swelling, causing extruder jamming, is prevented by using cold end heat sink and fan.
• M6 threaded cold end chamber allows for easy attachment to heat sink.

Cons

• The Hot End is difficult to assemble and has a lot of parts.
• The PTFE thermal barrier is difficult to get right because it deforms very easily, under pressure, when the M6 threaded tubes are screwed against it.
• The PTFE thermal barrier needs to be drilled on each assembly to align with the M6 threaded tubes. This causes extra wear on the inside of the tubes.
• The cooling fan adds extra noise to the 3d printer.

Conclusion

If I’d have got this Hot End from Mendel Parts instead of Make Mendel, I’m sure I would have had a few less problems. However, Make Mendel was the only company that had the parts and could deliver quickly.

1.75mm PLA is probably the most challenging Filament to extrude due to it’s relatively low glass transition temperature and of course being really thin as well. The Mendel Parts V9 Hot End derivative I created works well with this filament, and I’m sure Mendel Parts V9 original does work just as well if set up correctly. Anyway, working with a faulty Hot End has been very educational and has made me a bit wiser for my next purchase.

As it happens, I have a new Hot End on backorder, a Makerbot MK7/8 and Makergear Plastruder derivetive, so looking forword to getting that in the near future.

Hot End and 3D Printer Extruder Image Gallery

The notes will be about building a 3d printer heated build platform out of 3mm window glass and nichrome wire (resistance wire). This setup is ideal for printing PLA because it sticks to window glass without the need for any kind of tape, and PLA pops off the glass easily as the platform cools at the end of printing. Window glass is good for temperatures up to 80 degrees C which may not be hot enough for ABS printing. If ABS is how you roll then oven glass or ceramic glass may be a better option than window glass. The temperature I usually set, for printing with PLA filament, is between 55 degrees C and 60 degrees C.

Use the information at your own risk and do not leave the 3d printer unattended during operation because of risk of fire, safety first.

Resistance Wire Heated Build Platform Planning

Ohms and Watts

To build the heated build platform we need to figure out a few things first so that we can get the right bits together. We know that the RAMPS 1.3 board (RepRap Arduino Mega Pololu Shield) recommends a power supply be rated for at least 11 Amps to supply the heated build platform, that will give us a starting point to work out the lowest total desired resistance for the heating elements combined. The calculations will be based on 12v power supply that is rated to supply more than 11 Amps of current. So from that we can work out the total minimum resistance we need for our heating element by dividing the power supply voltage by the maximum amount of current we want to draw. I’m going to use 11 Amps as a guide to calculate the minimum resistance we want for the heated build platform, anymore than that will overheat the RAMPS 1.3 board. So we can calculate that 12 Volts divided by 11 Amps will give us a resistance value of 1.09 Ohms. Calculating power = 12 Volts x 11 Amps will give us 132 Watts of power which should provide a fast heat bed warm up, the lower the power rating, the longer it will take for the resistance wire heated build platform to reach the target temperature.

Now that we know what the minimum desired resistance we want for our heated build platform, we now want to work out the maximum desired resistance. This is important because we need to know what margin of resistance is acceptable when working out the design of the heating elements. The wider the resistance margin the easier it is to design the heating elements. We want the heated build platform to reach the target temperature quickly so we will work out the maximum resistance that will give us at least a 100 Watts heated build platform. So, Amps = 100 Watts divided by 12 Volts will give us 8.33 Amps, and Ohms = 12 Volts divided by 8.33 Amps will give us 1.44 Ohms resistance. Now we have a resistance margin to work with, which will be between 1.09 and 1.44 Ohms.

Our target resistance is set within the resistance margin we have defined above, which is the resistance of all the heating elements combined built into the heated build platform. If the heated build platform requires two heating elements then each element needs to have a resistance two times that of the target resistance, so if the target is 1.5 Ohms then each element should have a resistance of 3 Ohms. Each heating element should have a resistance of the target resistance times the number heating elements used in the heated build platform.

Target Resistance = Element Resistance / Number of Elements

Element Design

One way to design the resistance wire heating elements for the heated build platform is to use a vector graphics editor like a free to download program like Inkscape from inkscape.org, or you could use paper, pencil and ruler. The idea is to sketch the heat bed outline and the resistance wire heating element design to scale so that the glass can be placed on top of it for tracing the resistance wire around the element design.

Depending on the size of the heated build platform you are working on, you may need more than one resistance wire heating element and ideally, you want all the heating elements resistance wire ends to finish at the same end off the platform. Also you want to decide which end of the platform you want the electrical wires to lead off and route to the electronics and enough wire slack will be needed for axis motion.

Designing a heating element with one continuous resistance wire to cover the whole heated build platform may be possible, but the resistance wire required, will be far too thick to work with and keep it stuck down. I would recommend using a resistance wire of 22 gauge and up for easier usability, a table of resistance wire gauges is shown below with corresponding resistance values per meter.

Nichrome Wire / Resistance wire table

A single heating element is one continuous resistance wire that is spread across the whole of the heated build platform or spread across a portion of it, in which case, more than one heating element is used to heat the bed, with all the resistance wire elements being the same length and pattern. Glass is a good thermal insulator so it is important to make sure the resistance wire is spread across it as evenly as possible to avoid cold spots during heating. Uneven heating may result in the glass cracking or breaking and could cause issues with 3d print’s bond to the heated build platform during printing because of wide temperature differences and changes.

When designing a resistance wire heating element for 3mm window glass, I would recommend having the resistance wire no more than 20mm apart down the length of the glass and also have the resistance wire placed with a gap from the edge of the glass measured less than half the gap of that measured between resistance wires. I’ve included images as an example of what a heated build platform plan could look like which includes some calculations to work out the ratings.

Choosing the resistance wire

It is a lot easier to match a resistance wire to the desired plan we want for our heated build platform because it gives us more flexibility in creating the desired heating element pattern. If you have the resistance wire already, you’re plan options are going to be limited.

Unless your heated build platform is a lot smaller than mine, I would recommend starting with at least two heating elements. It is just a case of sketching out the first heating element to cover half the platform to get a rough idea of the length of resistance wire we need for one heating element. Assuming we are going to use two heating elements in our plan, each element needs to be twice the target resistance to fit within the resistance margin calculated earlier which is between 1.09 and 1.44 Ohms. So if our heating element resistance wire is 880mm in length we can select a resistance wire from the list and divide it’s resistance by 1000 and times it by 880 and then divide by 2 to see if the result meets our target resistance range. If you tested the 22 SWG resistance wire from the list in the calculation, it will be found to be the best match and this would be the resistance wire or nichrome wire to buy for the heated build platform. If you needed 3 elements at 880mm in length then the 24 SWG resistance wire will be the best match having calculated resistance = resistance per meter / 1000 x 880 / 3 elements.

Bits you may need

Nichrome wire – or resistance wire, a wide range of gauges can be found on ebay. Should be easy enough to find any of those in the list above.

Glass – Window glass has been used successfully by me and other people, just keep the temperature below 80 C and you should be fine. If you can get heat resistant glass then even better, but can’t be sure if PLA sticks to this type directly, in which case, tape may need to be used. 3mm window glass is used for this tutorial project.

Power supply – for example if using the Ramps 1.3 board to power the printer and the heated build platform, you will need a power supply capable of supplying at least 16 Amps. A modded PC power supply can be used and may be the cheapest option.

Exhaust wrap – This can be used to insulate the underside of the glass and help keep the kapton tape stuck down on the glass that’s holding the resistance wire in place. This can be found on ebay but shop around because prices can be a large difference between suppliers.

Kapton tape – If you’re already into 3d printing, then you may have this already but again, available from ebay. A roll 10mm in width is the one I used for my heated build platform.

Blu-Tack – This is good for holding down the resistance wire onto the glass while forming the heating element.

Electrical wire – Two sorts of wire may be needed, suitable wire to connect each end of the resistance wire to a terminal block and suitable wire to connect the terminal block to the Ramps board, each type of wire needs to have a power rating better than what was calculated for the heating element or heated build platform.

Terminal block – May be needed to connect all the wires from the heating elements, use one with at least a 15 Amp rating.

Thermistor – Refer to the RepRap Wiki for more information which covers the types recommended for use with your 3d printer firmware. A suitable connector will be required to fit the board like the RAMPS 1.3. The thermistor I used is one from rapidonline.com part no. 61-0452, Gt Thermistor 100k 3%.

Crimp connectors – or bootlace ferrules to join the electrical wire to resistance wire, solder may not stick to resistance wire but soldering the crimped join should make a more robust connection which is needed for a moving heated build platform. Solder may not be useful on resistance wire connections for higher bed temperatures due to the low solder melt point.

Heat shrink tubing – Some of this will be recommended for use to install the thermistor, it would help to reduce the possibility of the thermistor wires shorting.

Acetone – or nail polish remover, the type without added oil, is used to clean the heated glass bed if printing PLA plastic directly on to it. Dirty or greasy heated glass bed will prevent PLA sticking to it.

Heated Build Platform Assembling

So, you have all the the bits and ready to start building your heated build platform for your 3d printer. To build the heated build platform, follow the steps below:

Getting ready. If you are using a transparent glass platform and you have made a resistance wire heated bed plan like the above, then you just start with placing the glass platform onto the plan. Next, cut the resistance wire to length to match the length decided on the plan, and if using more than one heating element, cut all the resistance wires to the same length so that they all have the same resistance.

Shaping the resistance wire. Before you attempt to make the resistance wire heating element, roll up a few small balls of Blu-Tack ready to tack down the resistance wire onto the glass. Place the resistance wire onto the glass and first tack down the ends of the wire where the heating element begins and ends at the edge of the glass. Now form the rest of the heating element by working towards the middle starting from the ends of the resistance wire. Keeping the wire u-turns rounded during the heating element forming will allow you to make adjustments to the pattern easily to best fit the heating element plan placed under the glass.

Sticking it down. Once you have all the heating elements in place and applied the finishing touches, the heating element is ready to be stuck down with Kapton tape. Kapton tape the resistance wire to the glass, first between the Blu-Tack, then fill the gaps with Kapton tape after removing the Blu-Tack. Leave a bit of the resistance wire ends bare so that they can be connected to electrical wire.

Connecting the wires. How the electrical wires are going to fit will depend on the 3d printer design. Place the heated bed in the 3d printer, to help decide what wire lengths are needed, and plan the electrical wire routes from there. Remove the heated bed from the printer to attach the electrical wires to the resistance wires. Some of the images included will show an example of how the electrical wires might be connected up. I used 0.1 inch crimp terminals to connect electrical wires to resistance wires but bootlace ferrules can be used as well. Make sure the connections are robust enough to cope with heated build platform axis motion.

Adding the thermistor. Prepare the thermistor to be stuck on the glass by connecting electrical wire to it’s legs using bootlace ferrules or some other clamping type of connector. Insulate the bare wires up to the thermistor bead with Kapton tape to avoid shorting, also use shrink tubing where necessary. For best thermistor placement checkout the post Heated Build Platform Rework, stick the thermistor to the glass with Kapton tape.

Fitting to 3d printer. Fit the glass bed on to the printer and connect all the wires. For best heat bed performance and to protect the 3d printer, add a thermal barrier to the glass underside. Route the thermistor wire and the heated build platform wire to the control board. Fit a connector to the thermistor wire if required and connect to the control board. If using the RAMPS 1.3 board, the thermistor would be connected to header pins labelled T1, while the heated build platform would be connected to D8 connector. Attach the power supply, to feed both the 5 Amp and the 11 Amp connectors in the case of the RAMPS 1.3 board.

Final checks. Check that all the wires are securely connected and check that axis motion does not trap any of the wires. Make sure that one end of each heating element is connected to ground while the other end of each heating element is connect to +12 Volts. Disconnect the heated build platform from the control board and test the resistance of the platform with a multimeter to see if it meets within our target margin resistance. If no resistance is detected then there could be a short circuit or no electrical circuit and will need to check the wires again.

Resistance Wire Heated Build Platform Testing

Once you are happy with the resistance wire heated build platform installation and all the electrical wires are connected correctly and securely, then the next step is testing.

Before the heated build platform and thermistor can work they need to be enabled in the control board firmware, but before you do that, test that the printer is functioning normally with the new power supply if you changed it. Now that you know the power supply is working, update the firmware on the control board to enable the heated build platform and thermistor. With the 3d printer powered up and connected to interface software, confirm a proper reading is being made from the thermistor, about ambient temperature, and if that looks OK, test the resistance wire heated build platform starting at a low temperature first and increase from there until you reach your desired target temperature. An infrared thermometer would be useful to compare the bed temperature with the thermistor reading.

Adding a heartbeat. An LED can be added to the resistance wire heated build platform, as shown in some of the images, to show that the heated bed is live. Just add a 1k Ohms resistor to the longest leg (anode) of the LED and add extra wire to the LED legs to the desired length. The wires and the resistor can be connected to the LED with bootlace ferrules without soldering. Use heatshrink tube or insulation tape to insulate the wires. Connect the LED leg with the 1k resistor to 12v and connect the other leg to ground, the LED is connected to the electrical wire that is connected to the resistance wire from the control board.

Closing. Thanks for your interest in this tutorial and hope it was useful. If you managed to build a resistance wire heated build platform from the notes, please leave a comment.

Resistance Wire Heated Build Platform Gallery

The hand nibbler tool has provided an opportunity to recycle old household electrical enclosures in to something new when added to 3d printer printed designs. The sheet metal nibbler has worked very well for me, operation is easy, quiet and clean and importantly, convenient and accurate.

Nibbler Tool Features Review

The nibbler tool I have is the Draper Expert Hand Nibbler sheet metal cutter 35748, and I got this from Ebay for less than eleven pounds delivered. It can cut sheet metal up to 1.2mm thickness and laminates and plastics up to 2mm, but I’ve only used these for metal sheets so far, and I’ve been able to cut metal sheets down a scribed line with good accuracy. However, the nibbler tool is not great for cutting out round holes but curvy shapes are possible if they are not too tight. Usually, I can rough cut curves before rounding them off neatly with a Dremel or file.

It is claimed that a wide variety of materials can be cut cleanly without distortion, a claim  I can confirm as being accurate, at least in the case of the salvaged DVD drive enclosures I’ve been using. Other features are the interchangeable cutters and spring loaded handle with slip guards that allow you to comfortably push the hand nibbler forward while cutting sheet metal, and it’s quiet and clean operation will make it ideal for late night hacking.

Nibbler Tool Prerequisite and Usage

When working with sheet metal, other tools and safety gear may be required besides the nibbler tool to complete a project, and a steel rule and scriber would be ideal to have at hand for best mark out ability. Pliers are also good to have for straightening and at least two pliers would be needed for bending and folding the work piece.

While the nibbler tool doesn’t normally make sharp edges from cuts, as a precaution, it’s best to have gloves to put on which would make handling sheet metal a lot more comfortable. Eye protection kit should always be at hand making it a good policy is to wear them as a routine when using tools or when filing and sanding.

I always try to get the best lighting conditions possible when marking out and cutting and sometimes prefer to go outside of the house for best results. Sometimes scribed lines can be difficult to see in shadows so good lighting is essential for cutting accurately. Also when marking out, allow at least 3mm for the cutter to take out as waste from cuts and for the best cutting accuracy, I like to use the hand nibbler tool upside down to make it easier to see the scribed lines.

Putting a little light oil on the moving parts including the cutter will keep the nibbler tool functioning smoothly and make easier work at cutting sheet metal.

Nibbler tool used in Projects

I originally bought the nibbler tool to put some vent holes in the side of my computer case for extra cooling and now, since owning a 3d printer, it is being put to further use in much more interesting ways, I may even need to buy some spare cutters.

My SUMPOD 3d printer extruder and hot end have gone through a lot of modifications in order to get better performance, and the nibbler tool has been one of the tools essential for making those modifications, such as cutting out hot end brackets, bowden cable brackets, nozzle cold end heat sinks and peek insulator heat sinks.

The materials used for making these brackets and heat sinks have come from old PC DVD drive enclosures and copper pipe, and without the nibbler tool, it would have been difficult to make the cuts and form the shapes needed.

With the nibbler tool there is no need to make everything entirely out of 3d printed plastic, things such as enclosures can be made to look much more interesting by sticking metal panels to plastic frames, hot glue works very well for this.

Sheet metal is all around us waiting to be recycled in to our next project, so before you throw out that old set top box or VCR, skin it first and turn the old enclosures in to something new with the help of the nibbler tool.

I’ve been using this heated build platform for a while now and so far been happy with it’s performance. However, for the larger prints and longer print runs, the base of some prints start to warp and become unstuck from the glass on the heated build platform leading to some prints not making it to the end of printing. To keep warping to a minimum, the first layer of the print needs to be stuck down well to the heated build platform. To achieve this, some investigation is needed so that a solution can be considered.

When testing the temperature of the heated build platform I found that when the heating element is powered, the temperature difference went as high as fourteen degrees C on the glass between the thermistor and heating element. I believe this temperature difference, which was happening about every 230 seconds, caused enough contraction and expansion to cause prints to become unstuck from the glass. Clicking and cracking sounds can be heard during printing as the print starts to become unstuck from the heated build platform.

Heated Build Platform Analysis & Rework

Analysis

Armed with a laser guided heat sensor gun, a stop watch and a spread sheet, I set about finding out what was going on. I preheated the glass bed and let it go through the heating cycle a few times before taking the measurements.

I pointed the heat sensor gun at the target and took measurements for the duration of a heating cycle. The targets for this exercise is the glass area above the nichrome wire and the glass area above the thermistor. See the chart below.

From the chart, when the heater is powered, the temperature on the glass above the nichrome wire climbs to 68 degrees C, 13 degrees C above the target temperature. It takes about 30 seconds for the thermistor to detect a temperature rise, and when the target temperature is reached, the nichrome wire heating element is powered off.

While the nichrome heating element is powered off, the temperature in the glass above the thermistor continues to conduct heat from the hotter areas of the glass which lasts for around 50 seconds. During the next 140 seconds the temperature starts to drop gradually and becoming more evenly spread across the glass before the heating cycle starts over again.

Conclusion
The low thermal conductivity of the glass and the position of the thermistor being far from the heat source is causing uneven heating of the glass and poor heat control. Areas of the glass get far too hot, much higher than the target temperature. The target temperature is not maintained and only reliably sets the lowest temperature of the platform.

Remedy
Ideally, the glass platform could do with more nichrome wire which would improve the heat spread across the glass and reduce temperature differences between the heat element areas and non heat element areas. Unfortunately, adding more nichrome wire will mean buying a different rating wire to replace the wire already fitted and involve rewiring the circuit.

Before I start buying and replacing nichrome wire, I’m going to try changing the position of the thermistor to see if that will get me close enough to a better temperature controlled heated build platform.

Heated Build Platform Rework

The first attempt at modifying the heated build platform involved in moving the thermistor closer to the nichrome wire heating element. I secured the thermistor to the glass leaving a gap of around 2mm between the thermistor and the nichrome wire.

After returning the heated build platform to the 3d printer for testing, the results came as expected, the thermistor heated up too quickly causing the heater to turn off before enough heat could be transferred to the glass. The glass took too long to heat up from cold and could not reach the target temperature. The thermistor did not give a correct temperature reading from the glass because the thermistor was too close to the nichrome wire. The thermistor could not supply a consistent, stable reading around the target temperature. However, the heat is more evenly spread across the glass than before, which is what I’m looking for, but can only maintain a temperature at around 10 degrees C below target.

The second attempt was more successful, which involved in using a copper heat spreader attached between the thermistor and nichrome wire heating element. The heat was absorbed and transferred very quickly to the thermistor with a much improved short regular heating cycle. The target temperature was maintained with a stable reading from the thermistor and the reading matched that of the heat sensor gun when pointed at the glass area over the thermistor. The temperature difference across the glass build platform is a lot better than it was before which has been measured at around 6 degrees C just after the nichrome wire is powered off.

Conclusion

Changing the position of the thermistor worked really well, I now get an accurate stable reading from the thermistor and the temperature across the heated build platform is kept within the target temperature. This should be enough to stop the prints becoming unstuck through expansion and contraction of the heated build platform.

Heated Build Platform Gallery