Reprap Forum - Look what I made!

remade old Prusa i3 to a Mini Me printer and control box (no replies)

yar60 — Wed, 27 May 2026 12:33:38 -0400

It's Mini Me... took all of the parts of my old Prusa i3 got 20x40 and 20x20 and made it look like my CR 10, does a good job of printing it ended up 235x235x200mm.

UMMD prints another lamp (no replies)

the_digital_dentist — Mon, 09 Dec 2024 10:11:13 -0500

PETG, 580mm tall, vase mode, 1 mm wall, 0.25 mm layers, 13+ hours (mostly while I slept)

[attachment 121257 midsection.jpg]

This is what it will be when finished. 47" tall. I have to print a new clear top piece and the pieces that go into the outer "tubes", then wire the two G63 bulbs. I'll probably run wires to the top bulb through the tubes.

[attachment 121258 stacked.jpg]

EZ to print magnetic polarity stick (no replies)

LostinNashville — Sat, 25 May 2024 13:17:43 -0400

This a super simple to print, magnetic polarity press stick. Fits 6mm x 3mm round magnet. Just add the magnet make sure to reverse the direction between the two. Drop in the magnet and add a drop or two of super glue. I made it fit loose on purpose, no need to print this twice, or tweek it. Your printing this because you're making something else. Make the thing you wanted :)
Link [www.printables.com]

Toothy Sock Puppet Print (no replies)

LostinNashville — Sat, 18 Nov 2023 14:58:29 -0500

[attachment 120637 PXL_20231118_160746828.png]
[attachment 120640 puppet.jpg]
Here's a sock puppet print I designed myself over many iterations. I hope you enjoy it! My son and I had a ton of fun with this. If you do please share your stories. Thank you!

3d print link: [www.printables.com]
build video here: [youtu.be]

Build of materials---

printed parts

knee high sock

felt

e6000 glue

20mm glass eye

4 inch zip ties

gaffer tape or tape

Tool list----

pliers

side cutters – optional

paper towels

sharpie

scissors

Tool for sanding disc brake pads (2 replies)

the_digital_dentist — Sun, 24 Mar 2024 08:33:03 -0400

My bike has belt drive and a continuously variable transmission. It's been extremely reliable, except that it also has hydraulic disc brakes. Those have been not so reliable. In fact, those brakes alone have required more maintenance than any whole bike I've ever owned.

The biggest problem with disc brakes is that they squeal a lot. When they squeal, they don't stop the bike very well. Fixing them usually requires sanding the pads, but they are small and thin and hard to hold onto so sanding them usually ends up sanding the skin off my fingers, too. I decided to solve that problem. I made a tool to hold the brake pads. It took all of 10 minutes to design it and about an hour to print it.

PXL_20230512_170358483.jpg

More here: [drmrehorst.blogspot.com]

My first self made 3D printer presentation and ask for help (3 replies)

kryst3q — Sun, 04 Jun 2023 18:05:15 -0400

Hello everyone!

I'm happy to join to this forum and announce that I finally end my first 3D printer (Prusa i3 Rework) built according to instructions from reprap wiki. My specification is:
- T2 belts (X, Z)
- 20-tooth pulleys (X, Z)
- NEMA 17 motors
- RAMPS 1.4
- Arduino MEGA 2560
- Stepper motor drivers: Y (StepStick, set to 1.22V, not so loud), X (DRV8825 based driver, set to 0.85V, louder than Y), Z (DRV8825 based driver, set to 0.85V, very loud), E (DRV8825 based driver, 1.5V, not so loud)
- Marlin firmware (version downloaded from the printer page I put link above)

[attachment 120350 Prusa-i3-rework.jpg]

[attachment 120352 electronics.jpg]

I was able to perform 3 prints of calibration cube. Each one of them was better than previous one but still something needs to be calibrated. I also probably need new filament because the current one (SWIFT PET-G 1.75mm, black) was laying down few last years without moisture protection (according to what I read about it in the internet it can be a problem, right?).

[attachment 120351 calibration-cube-3rd-print.jpg]

As a beginner I have few questions and would be very grateful for any help.
1. In my Marlin firmware I have such configuration set currently: #define DEFAULT_AXIS_STEPS_PER_UNIT {5,5,250,30}. But those values (which led me to print the cube above) are not matching the calculations done according to this page. My question is: what is going on? Where is the problem? Why those values does not match?
2. When I click homing axes are moving very slowly (they're fast during a print). Should it be that way or I can change something so they will be as fast as during the print?
3. Connection place between extruder and it's radiator is leaking (but it's not disturbing printing process - at least on the first glance). Should I be bothered by that? What can be a cause of this?

[attachment 120353 leak.jpg]

Thank you in advance!

3d printed New York City Midtown (1 reply)

Maptogram — Mon, 13 Mar 2023 15:32:38 -0400

Hi,

I used my own files to produce a map of Midtown New York City. The buildings in these files are quite detailed and are not simply plain cubes. Actually, almost all of the nuances are visible. On Etsy, I sell them for a low price. Perhaps you'd like to explore it yourself. Below, I've included a link to Etsy.

Files for NYC

il_1588xN.4490298303_4kd0.jpg

gcode thumbnailer for Gnome (linux) (no replies)

Dust — Sat, 04 Jun 2022 10:45:35 -0400

A simple thumbnail generator for Gnome. This extracts the encoded thumbnail data from within a gcode file and turns it into OS useable thumbnail, such as in nautilus.

examples%20(2xZoom).png

More details and code can be fond [github.com]

My 1st home made 3D printer (6 replies)

Laneroger69 — Fri, 19 May 2023 09:21:54 -0400

My 1st home made printer.
235mm x 270mm x 250mm build area. Heated bed, 2 in 1 out hot end, Mega2560 and RAMPS 1.4, lead screw drive all axis, TMC2209 stepper drivers.

Still working out a couple kinks and calibrating.

Sci-Fi table lamp (6 replies)

the_digital_dentist — Mon, 09 Dec 2024 10:58:31 -0500

I recently designed and printed a table lamp that looks a bit like a prop out of a 1930s sci-fi movie:

The print is about 270 mm dia x 300 mm tall and the lamp is about 400 mm tall overall. It is designed to fit a Feit Electric G63 light bulb that is about 200 mm in diameter. I printed in PETG with a 1 mm nozzle, 1.2 mm line width, 0.5 mm layers, 30 mm/sec and it finished in a little over 3 hours.

More here, including a link to the STEP file.

New no-stab wire twister design (1 reply)

the_digital_dentist — Sat, 26 Mar 2022 15:11:59 -0400

If you're building a printer or other machine that has long cable runs, twisting the wires helps keep everything neat. I designed and printed a wire twister a few years ago and used it for many projects but never liked having to use a screwdriver to clamp the wires in the tool. I recently designed a new tool to use wago lever nuts to clamp the wires and it is a huge improvement.

All you do is put the spinning clamp in your electric drill/screwdriver, clamp the wires into the Wagos, and spin.

More here, including a link to the CAD file: [drmrehorst.blogspot.com]

Insane PLA melt rate. Another waste paper bin, this one in 73 minutes. (1 reply)

deckingman — Tue, 21 Dec 2021 13:11:31 -0500

Following on from my fairly crazy melt rate tests using my 6 input multi-material hot end with PET-G, I've printed another waste bin but using PLA this time. With the same 1.5mm diameter nozzle, 1.65mm layer width, 1.0mm layer height, I achieved a reliable print speed of 80mm/sec (the extruders started skipping at 90mm.sec). It's only 7 minutes - watch to the end to see what the insane melt rate was. [youtu.be]

A functional PET-G waste paper bin printed in under 2 hrs (3 replies)

deckingman — Tue, 23 Nov 2021 12:03:19 -0500

Following on from my other post about exploring volumetric flow rates with my 6 input multi-material hot end (https://reprap.org/forum/read.php?2,887063), I decided to use up some more of this 6 year old PET-G and make something mildly useful. It's a functional waste paper bin printed in under 2 hours which serves to illustrate that 3D prints can grow faster than mushrooms. :)

Link here - [youtu.be]

(Mostly) 3D printed Xmas Pyramid revisted (1 reply)

deckingman — Mon, 25 Oct 2021 10:02:51 -0400

I have revisited my (mostly) 3D printed Xmas Pyramid. I have replaced the previous fixed speed, fixed voltage, geared (noisy) motor with a stepper motor, driven by a TMC2208 driver module which has made it silent (at least to my ears). I've also replaced the LEDs with new, WS2812 addressable LEDs and use a Wemos D1 mini running ESP home to simulate a flickering candle effect. Finally, I have made a new drive shaft to replace the previous threaded rod. This is in sections for easier storage and along with redesigned rotating discs, allows tool free assemble.

[youtu.be]

Fountain pens (3 replies)

MJLew — Sat, 01 Jan 2022 01:52:37 -0500

I've spent lockdown combining my hobbies of 3d printing and fountain pens and I'm pretty pleased with the results.

The pen is made up of seven 3d-printed parts, each made from a plastic selected for its properties matching the functional requirements of each part, with a small amount of lead ballast in the grip and in the body. The ballast is eccentrically located so that the pen is less likely to roll off a flat surface. (It’s not a perfect roll-stop, but better than nothing.) The surfaces of the pen are ‘straight off the printer’ apart from the ends of the cap and body, which have been sanded and polished.

The body, grip and cap outers are all made of Polyalchemy Elixir PLA chosen entirely for appearance. The body liner is Xtron3D Marble, selected for its high hardness. The section inner and cap liner are made of a slightly softer PLA, Aurarum PPLA, so that the threads will deform to a perfect fit rather than wear when they are used.

Only the section inner is made in the conventional 3d printer manner (i.e. 3d model [OpenSCAD] then slicer [Simplify 3D]). The other parts are all made using G-code files that are made by software that I wrote in LiveCode and are printed with a single wall, “vase mode”. The single wall components are far too weak to stand alone—particularly given the relatively poor inter-layer adhesion property of Polyalchemy Elixir—but they are reinforced by being made into an epoxy-filled sandwich with their respective liner parts.

That unusual approach to 3d-printed construction means that I gain freedom of design and am easily able to achieve the intricate surface structure that gives the pens its sparkly glistening shine.

Golf Tee Area Markers (no replies)

RepRot — Thu, 26 Aug 2021 22:24:57 -0400

I made some large concrete golf balls for use as tee markers at a local golf club.
To do this, I made a mould with my 3D printer using ABS plastic filament.

Each concrete golf ball marker is 120mm in diameter (4 ¾ inches).

The 3D printed mould I made consists of eight 3D printed parts which I hold together in a polycarbonate box when filling.
The top 4 molded parts have a small hole which I use a funnel to push down a mix of cement and sand.
Once the cement mix has harden but not solid, I remove the ball from the mould and smooth off the area where it was filled. Once fully harden they are painted.

As golf balls have dimples I had to make the mould with 8 pieces rather than two to be able to remove the concrete ball from the mould. I made the dimples larger to make them stand out more.

One issue I had was the sticking of the cement mix to the mould which can easy destroy the mould pieces.
I tried a number of different products but found using Vaseline cream to be good.

Each tee area required 6 balls to be made; Ladies, Men’s & Pro’s (Play off is between and behind two balls) and with being an 18 hole course, 108 balls had to be made.
Each ball took an hour for mixing and filling the mould and removal and cleaning up the mould for reuse.

The course green keepers like them as they can move them easily without getting off their ride on lawnmowers. Also they have a small resting footprint which is good for the grass they sit on.
Being concrete they don’t blow away with the wind and people don’t try and hit them with their clubs.

Flight Simulator Aircraft Gauges – Instruments (Arduino controlled) (7 replies)

RepRot — Fri, 01 Jul 2022 10:10:46 -0400

In the pursuit of a more realistic flight simulator controls and display and also having already designed and built a control console and foot controls with force feedback; I thought I would try and design and build a set of aircraft gauges. I used my 3D printer that I had built to make a lot of the gauge parts.

With each aircraft generally having a set of gauges that are suited for its design and having already built controls based on the Cessna 172 I therefore based my gauge designs on the Cessna 172 aircraft.

I am using the following PC programs:
Prepar3D version 5- Flight Simulator.
A2 Simulator Cessna 172 aircraft addon.
FSCUIPC Registered version for Prepar3D v5 . (Used for extraction of data from Prepar3D to Arduino processors).

One of my requirements was not to have a real sized cockpit layout, rather a more compact setup that wouldn’t take over my computer room space.

One of the differences with aircraft gauges that are displayed on a monitor flat screen and a physical gauge is that physical gauges have a lesser viewing angle due to there depth. Therefore I had to consider the size I was going to make my gauges and to where I was going to position them.

The other thing I had to consider is that I wanted to have a total of 12 gauges that would be arranged in two rows of 6 gauges that would fit in a box on top of my existing control console being 425mm wide.
With some gauges being a dual type and others requiring different drive motors I worked out that I could reduce the size of the gauges to 70mm x 70mm (height x width) with a panel hole of 65 mm diameter with a viewing size of 60mm diameter. This is slightly smaller than normal aircraft gauges.
The other thing I needed to do was to mount the gauges in a box with a sloping front to provide a better viewing angle.
--------------------
Overall it’s been an interesting project, somewhat of a learning curve and not a quick project....
I am pleased with the finished product and it’s enjoyable to use…
And it all started after I started building 3d printers some years ago….

----------------------------------------------------------------------------------------------------------------------------------------
Warning: Only read the following if you wish to now about the technical stuff…. .

Gauge Motors:
I have used Switec X27 automotive geared stepper motors for the single axis pointers gauges
They can be purchase from the likes of Aliexpress and Ebay for a couple of dollars each.

I also used Switec X40 automotive geared stepper motors for the dual pointer single axis gauges.
The X40 is a dual version of the X27. The X40 motors are end to end with the output shafts inside the other.
They can be purchased from Aliexpress for about five times the price of an X27.

The X27 & X40 geared stepper motors are low current motors and can be driven from 5v to 9v. They contain two coils for each motor unit (4 wire connections for each motor).
Wire leads can be soldered onto the motor connections.
X27 & X40 have 315 degrees of rotation with an internal mechanical stop.
The X27 & X40 geared stepper motors are used in many vehicles for speedometer and tachometers.

You can connect and drive these gauge motors directly from an Arduino micro processor input/output pin (I/O).
Information about this can be found on the internet.
While the X27/X40 can be connected directly to the Arduino pins, protection diodes should really be used to stop the possibility of electrical spikes damaging the Arduino.

I tried this direct connect method but found that it didn’t provide a very smooth movement. My next thing was to try using micro stepping.
I then tried using a spare Polulu stepper driver which uses the A4988 driver chip. I had to supply a driver voltage of at least 8 volts for the A4988 to allow an output to the stepper motor. I found using the A4988 with micro stepping produced a lot smoother movement. However the stepping count wasn’t correct and movement was noisy.
Looking into this issue I found that the X27 & X40 style stepper motors use a 6 stage stepping sequence where as the A4988 stepper driver commonly used for Nema type motors use an 8 stage stepper sequence.

I then tried a AX120728SG driver chip which is equivalent of the VID6606 (Uses 6 stage stepping sequence). This produced a lower noise and smoother movement . Note: Refer to the VID6606 datasheet for the AX120728SG specs.
The AX120728SG driver chip is a surface mount device. However they can be hand soldered onto a SOP28 adapter boards. These adapter boards can be purchase for about $0.40 each. It can be a bit tricky to get all the pins connected first time but not impossible. With soldering 140 pins I failed with 3, even testing them with a meter won’t always indicate that you haven’t soldered them properly.

Each AX120728SG can drive 4 stepper motors. It requires a step and direction input signal for each stepper motor and has a reset pin which needs to be held high and requires a +5 volt supply.

As the X27 & X40 motors have no output to determine the output shaft position. An initial homing function needs to be performed to determine the shaft position.

With a mechanical 315 degree of movement, the trick is to send it enough step pulses as to move it to an end stop limit position. As you properly don’t know where the motor is, it means sending it the amount of pulses that would move the motor though it’s full range.
I did find that once the motor reaches its full movement and you keep sending pulses, it does produce a small amount of noise. This doesn’t cause damage to the motor.

I found on startup of my complete gauge display unit, which has multiple gauge stepper motors all homing at once, it can produce quite an amount of noise. This is subject mainly to the issue of the motors hitting against the end stops.
I found I could reduce this homing noise by knowing where the motor were when last used.
I built into my Arduino program for controlling the gauges that if no data was received from the PC for a set time, then certain gauges would be moved to their zero position. I used the flight simulator “TIME” extraction data as it is always being updated every second.
While this reduces the issue with noise with the motors hitting the end stops on powering up, the down side is if you pause the flight simulator, some gauges will move to the zero position. On un-pausing they then move back and indicate as they should. I don’t find this a major issue.

Gauge pointers:
I 3D printed gauge pointers (needles) for the X27 & X40 stepper motors.
The X27 motor has a 1 mm diameter shaft.
The X40 motor has a 1 mm diameter inner shaft and a 2.5 mm outer shaft.
It can be difficult to glue the pointers correctly in position. Fortunately I found I didn’t need to use the full 315 degrees of movement.
What I did in the programming was to have an offset variable for allowing fine tuning of the zero gauge position.
The gauge label zero position is usually further away from the start movement end stop.
I also use a variable for the gauge full movement step count as some gauge motors have been modified for reduced movements.

For homing (knowing the last position being zero), I move the gauge forward (full movement less the offset), then move it backwards (full movement) and then move forward (offset) and set the gauge position counter at zero. This reduces the noise of the motor being driven against the end stops and ensures gauges are in there correct position.

Normal mounting of the X27 & X40 motors is by soldering the motor coil connections to a PCB, however I found I could use two M2x12mm counter sunk screws that will self tap into the existing holes in the X27 & X40 housing. Care needs to be taken as not to place any warping pressure onto the X27 & X40 plastic casing, as this can damage the motors gear meshing which can result in noisy or non operation. Best to self tap screws before motor is fitted to motor mounting plate

Gauge Construction:
I designed the gauges to be a number of parts that are laminated together which allows many of the same parts to be used for different gauges
- Front bezel – 3D printed
- Perspex window – (2 mm thick)
- Gauge label – Bubble jet printed. (Only some gauges have this label, two motors type)
- Spacer – 3D printed - (5 mm thick) (Can have two 3mm LED’s for night flying).
- Pointer (Needle) – 3D printed
- Gauge label - Bubble jet printed
- Motor mounting plate – 3D printed

Four M3x25mm counter sunk screws with flanged nuts are used to hold the gauge parts together and four black M4 x 25mm flanged head screws with flanged nuts are used for fixing the gauge to the gauge box front panel.

I created the gauge labels using the free download Inkscape program. SVG format was used to create clean edged images that can also be scaled without loss of quality unlike Jpeg/bmp. The images were bubble jet printed onto white photo quality paper and then laminated with clear plastic. Holes in the labels were then made using hole punches for the required screw and gauge holes. The label edges were then coloured with a black felt tip pen.

Gauge Motor Control Software:
All gauge stepper motors use the Arduino AccelStepper library.
You can have multiple numbers of stepper motors using this library with one Arduino.
For each stepper motor, you need to set a Max speed and Acceleration rate. Acceleration rate is both acceleration and de- acceleration. These Max and Acceleration values can be changed within the running program if you need to.
You can set the current position value of the motor at any time.
You also can set the motor to move a set distance from it current position or move to a set count position. However setting the position doesn’t mean the motor moves.
After setting a move, you need to have a run command. This run command needs to be repeated multiple times as it only moves the motor a single step subject to Max speed and Acceleration rates and whether the motor has reached its move position that was set.
You also have a command function to known how much further the motor has to go before its going to get to the move position it was told to go.
You can use commands like this after setting the move position-
while(Stepper1.distanceToGo() != 0 {Stepper1.run();}
This provides multiple runs until the motor gets to the move position it was told to go to.

This is great for homing type movements but for normal flight simulator operation you don’t wish to move to a position that may have been updated with a new lesser movement.
I used Arduino communications interrupts to update the move position and then only in the loop part of the program I have the command for the stepper motor run.
Note: If you have too many commands taking too long in the main program it can slow down the movement of the stepper motor.

To reduce the time it takes to home gauges, I would home some gauges that are controlled by the same nano together. After setting each gauge movement I would use a command like the following..
while(Stepper3.distanceToGo() != 0 ||Stepper4.distanceToGo() != 0){Stepper3.run(); Stepper4.run();}

When designing and building each gauge I wrote two pieces of Arduino software for each gauge…
One allowed me to input data via serial monitor and the second one that allowed me to extract data from flight simulator via the FSUIPC program and compare the physical gauge movement with the gauge on the monitor.

RPM Gauge:
This gauge was the first gauge I designed and built as it is the easiest. Hardware side is fairly straight forward.
I used a fishing line glue for attaching the pointer to the stepper motor shaft
For the software side, the extracted data from flight simulator needs to be scaled to provide the correct movement range.
What I did find later with multiple gauge extractions and that not all gauges have a straight linear scale.
Thanks to Martin Nawrath -- [interface.khm.de]
I was able to create a re-mapping section in the program which allows for linear and non linear scaling for both positive and negative values.

Airspeed Gauge:
This gauge is much the same as the RPM gauge with a different label.
The airspeed gauge uses 3 different scales, e.g. 0 to 40 (half scale) 40 to 160 (normal scale) 160 to 200 (half scale). You can use a few “if” commands to get the correct gauge scale movements. However this is where the use of the re-mapping program section comes in useful.

Vertical Speed Gauge:
This gauge is much the same as the RPM gauge with a different label.
Homing of the gauge requires the pointer to be positioned in the middle of the movement range.

Oil Temperature/Pressure, EGT/CHT, VAC/AMP and Fuel Left/Right Gauges:
Each of these gauges uses two X27 motors side by side:
Note: The positioning of the motors side by side is one of the restrictions to how small all the gauges can be made.
The X27 motors for these dual type gauges require less movement that the standard 315 degree of movement.
The X27 motors can be carefully unclipped and by using a soldering iron, plastic can be melted into the limit channel to reduce the range of movement.
While it may not be easy to get an exact reduced movement with the melted plastic, the use of variables in the program for the range and offset can adjusted for this issue.

Reducing movement of the pointers is required as you don’t want pointers hitting parts of the gauge which may result in the pointers coming un-bonded from the X27 motor.
An additional printed label (center is cut out) is used with these gauges between the gauge window and the spacer.

Turn Coordinator Gauge:
An X40 motor (dual axis motor) is used for the Turn Coordinator gauge as both roll and side slip is required to be displayed.
Note: The X40 motor coil connections are usually facing the output shaft direction. These connections leads can be changed by re-soldering to face the opposite direction.
A white plastic disk which was 3d printed is used for the displaying side slip ball movement. A black circle dot was painted onto the white disk. The white disk has been glued onto the outer shaft (2.5 mm diameter) of the X27 motor.
A 3D printed turn indicator pointer has been fitted to the inner shaft (1mm diameter) of the X27 motor.
Both motors of the X40 motor are homed as like the other X27 motors e.g. RPM, airspeed etc.
An additional spacer is used with this gauge due to the white plastic disk for displaying the slip ball.
A curved slot has been cut out of the printed label to allow viewing of the slide ball.

Compass Gauge:
This gauge requires a stepper motor that has 360 degree of rotation and that it can drive a 3d printed disk (2 mm thick - 55mm diameter) that has a compass label attached to it.
While an X27 motor can have the limit stop removed, it doesn’t have enough strength and speed to rotate a disk at any great speed. The other issue is attaching a compass disk to a 1mm motor shaft that won’t fall off over time.
I used a 24BYJ48 5 volt geared stepper motor. These motors have 360 degree of movement and have a lot more driving force than an X27. The 24BYJ48 requires to be driven by an 8 stage stepper sequence driver.. e.g. DRV8834 , A4988. They also have a 6 mm diameter output shaft with flatten sides, which makes a good bonding area to attach a 3D printed disk. They can be purchase for a few dollars each.
The correct stepper motor current levels needs to be set on the DRV8834 to avoid noisy operation and over heating.

As the 24BYJ48 motor has no output to determine the motor shaft position. A method is required to be able to detect the position of the motor shaft that has the 3d printer disk attached. I have used a hall effect magnetic sensor (A3144) that is attached (glued) to the 3D printed motor mounting plate. On the 3d printer disk I have installed a neodymium (2mm long x 1 mm diameter) magnet. This allows detection of a home position.
The A3144 hall effect magnetic sensor is a 3 wire device. One wire requires +5 volts, another wire requires 0 Volts and the third wire is the sensor output. This sensor output wire can be connected directly to an Arduino I/O pin. It requires that the Arduino internal pull up resistor to be enabled. When the A3144 detects a magnet, the sensor wire is driven low (0 volts).

Homing operation of the compass gauge:
1st – Check the magnetic sensor. If magnet is detected, then move the motor a short distance forward.
(The reason for this is the sensor can sense the magnet though a few degrees of movement.
2nd – Move the motor backwards until the magnet is sensed.
3rd – Move the motor offset distance
(Reason for the offset is it can be difficult to position the printed label on the plastic disk accurately).
4th – Set the Motor position at Zero (e.g. 0 - North).

I painted an image of a plane which shows the heading direction on the gauge window. I did this by using a laser engraver that I had built. I engrave a plane image onto a stick-on label to create a stencil. Using the stencil I painted a white plane image onto the Perspex window.

One of the issues with the compass gauge is the data from flight simulator is just a degree figure 0 -359.
When moving from 359 to 0 degrees or back for 0 to 359 degrees you don’t wish to take the long way around.
Therefore you have to use the current position and determine from the new direction figure which direction and amount of movement is required. The Arduino Modulus formula(%) comes in useful here.

CurrentHeading = Stepper1.currentPosition();
LastHeading = CurrentHeading/11.377778; // The figure 11.377778 is steps/degree
Movement = ((Data + 180 - LastHeading) % 360) - 180
if(Movement < -180){Movement = Movement + 360;}
Stepper1.moveTo(CurrentHeading + (Movement) * 11.377778);} // The figure 11.377778 is steps/degree
Note: The correct stepper motor current level needs to be set on the DRV8834 to avoid noisy operation and over heating.

Altitude Gauge:
A X40 motor (dual axis motor) has been used for the Altitude gauge.
As the altitude gauge requires 360 degree movement the X40 motor was carefully dismantled and the end stop lugs for both motors was cut off.
Note: The X40 gears are very fine and you don’t want a small piece of plastic to get jammed in there.
Motor connection wires were also swapped to have them facing away from the output shaft direction.

The altimeter pointers were 3d printed to allow a small neodymium magnet (2mm long x 1mm diameter ) to be installed on the rear side of both pointers. The larger pointer has the magnet positioned towards its outer end away from the center of the gauge as not to interfere with the smaller pointer.
Two hall effect magnetic sensors (A3144) were mounted into the motor mounting plates. These sensors were positioned 180 degrees from each other and aligned under the pointers magnets
Homing is required for each pointer and is just like the compass homing.
I did find it looks less weird to home one pointer in one direction and the other in the opposite direction.
After the initial homing of both the pointers, the speed for the stepper motor for the smaller pointer (1000 feet) is set at 1/10 the speed of the larger pointer (100 feet).
This allows the pointers to move together at what appears as a clock geared motion.

Digital Clock Gauge:
For the digital clock gauge I have used a 6-Digit LED 0.36" Display , 7-segments GREEN, TM1637, size 46x14mm. I control it with the Arduino TM1637_6D.h library.

The 6 digit display can display the Time, Outside Air Temperature (OAT) and Voltage (bus voltage).
6 digits are used with decimal points for TIME. (hh.mm.ss)
3 digits on the left are used with decimal point for OAT ( xx.x )
3 digits on the right are used with decimal point for VOLTAGE ( xx.x)

To select the different display readings, I have used two stainless hex screws located by the Select and Control which would normally be press buttons.
The select one is connected to an Arduino analog pin and functions as a capacitor touch sensor that allows stepping though the three different displays.

Attitude Gauge:
This has been the most challenging gauge to design and build. It has taken some time and effort..
I first tried using two 24BYJ48 geared stepper motors (Same type I used for the compass gauge) to rotate a ping pong ball in two directions via brass gears to display roll and tilt. While I got it to work, movements weren’t fast enough and gear backlash was too much.
I then sourced a larger motor a Nema14 hybrid stepper motor direct from Casin in China. They were very helpful.
The motor cost NZ$20 and freight was NZ$20.
The Nema14 hybrid stepper motor is two stepper motors that are end to end with there output shafts inside the other. (Inner shaft 5 mm dia - Outer shaft 8 mm dia).
Using the Nema14 hybrid stepper motor I was able to directly rotate the ping pong ball without the use of gears and not have any gear backlash issues, Tilting still required the use of gears and using a bit of software I was able to reduced backlash, basically just over drive a bit when moving in one direction.
I used two DRV8834 stepper drivers and the correct stepper motor current levels needs to be set to avoid noisy operation and over heating.
I have used two hall effect (A3144) magnetic sensors for detecting home positions for roll and tilt. These sensors are mounted on a motor support plate. While the rotate sensor will detect every time the ping pong ball rotates, the tilt sensor only will detect when the rotate is in a set position. Thus on initial homing, the rotate motor rotates until detecting the home sensors, it then rotates to a set position to allow the tilt motor to sensor its home position.
Hopefully picture explains how it works..

Gauge Controller PCB:
To control the 12 sets of gauges, I designed a two sided 370mm x 100mm circuit board.
I used Kicad PCB program (can be downloaded for free).
It was the first time I had used this program. Previously, I have used Eagle for a few small boards.
It took a little to bit to work out matching up the footprints to symbols and 3D objects.
I do like the Kicad function of being able to display a 3D picture of your PCB with all the components.

I got my PCB made by JCLPCB in China. Minimum order is for 5 sets of PCB’s. Being my first order they gave me a US$8 discount. Total cost including delivery of the minimum 5 PCB’s was NZ$40. It only took 6 days from when files were emailed to have the PCB’s delivered to me in New Zealand. No Covid 19 then…

The PCB is designed more as an Arduino shield.
The PCB has 6 x pluggable Arduino Nano’s, 5 x pluggable SOP28 adapter boards that have AX120728SG stepper drivers. and 3 x pluggable DRV8834 stepper drivers.
There are a few capacitors as per specs for the AX120728SG and DRV8834 stepper drivers as well as a few diodes (Surface mount device - SMD), LED (SMD), Resistors (SMD), fuses and power sockets (+5V/GND).
There are also a number of Dupont 2.54mm header pin and socket connectors.
The SOP28 adapter board pcbs used for the AX120728SG’s can be purchase for about 40 cents each.
-----------------
As the AX120728SG require 5 volts, I decided on a 5 volt power supply bus. This required a 5 volt stepper driver for the larger 3 stepper motors used thus I chose to use three DRV8834 stepper drivers. Also I chose to supply 5 volts to each of the six Arduino Nano’s 5 volt supply pins.
The only issue with using a 5 volts power bus is that when programming any of the Arduino Nano’s with the 5 volt power supply off you don’t wish the PC to be supplying power to the whole gauge controller PCB. A diode (2N4001) has been used on the 5 volt supply line for each Arduino Nano processors.

The PCB design has six Arduino Nano’s micro processors, one of these Nano’s is connected via USB cable to a PC that is running flight simulator program.
This Nano that is connected to the PC is the “Master” device for I2C communication bus. This Master device is connected via the I2C bus to the other 5 Arduino Nano’s which are “Slaves.
For connecting all the six Arduino Nano’s together to use the I2C bus, all the Nano’s SCL pins including the master has to be connected and a resistor is required to pull this signal to 5 volts. Same applies to the Nano’s SDA pins. The resistance value is dependant on how many I2C devices are connected.
Each of the 5 slave Arduino Nano’s are connected to one AX120728SG which allows for 4 stepper motors (X27 or X40 type motors.
One of the slave Arduino Nano’s is connected to a single DRV8834 (for driving Compass)
One other slave Arduino Nano is connected to a two DRV8834 (for driving Attitude)
The Arduino’s Nano’s that control stepper motors with hall effect magnetic sensors (Compass, Attitude & Altitude) have additional connections which include +5v, 0v and the magnetic sensor signals for the gauge home sensors.

The controller PCB can control a total of 23 stepper motors ( 20 x X27/X40 style motors and 3 x 24BYJ48 or Nema type motors.
I am currently using a total of 18 stepper motors. The remaining 5 available are for an additional engine or they could be used for some other gauges that use the X27/X40 style motors.

I have designed the controller PCB to have two 5 volt power connectors at each end of the PCB.
Also at each end there is a Dupont socket connector that has the I2C bus signals SCL & SDA & +5V/Gnd.
This design, allows the controller board to be daisy chained to another controller board or to another PCB design board for aircraft designs requiring more gauges. The I2C pull-up resistors and locations would need to be matched to the number of I2C devices.

While I could have had the AX120728SG driver chips directly soldered to the controller PCB, I chose to have them soldered onto the SOP29 PCB’s. This allows swapping them about which makes troubleshooting quick and easy.

For connecting the master nano to the PC, I have used a USB Type-A female to female converter attached to the rear of the enclosure box. This allows for a more robust external USB connection rather that having a direct USB cable to the master Nano USB Mini type B fragile connector socket.
I have also cut the 5 volt wire on the USB cable connected to the master nano.
For the master nano to be seen by the PC, the master nano requires a power source (+5V). This is supplied from the a 230vac /5 volt power supply. E.g. Turn the Gauge Controller Unit on for it to be seen by PC..
---------------------------------------------

Although not yet tried, the gauge controller PCB could properly be used for inputting of aircraft switches via FSUIPC to flight simulator.
If the stepper driver chips PCB’s were removed and wire jumpers are used on the SOP28 adapter boards, the Arduino I/O’s can be connected to the row of gauge connections that the stepper motors use.
This would allow 9 x I/O’s per slave processor and with 5 slaves; giving a total of 45 switches that could be connected. Additional there are also the another 11 x I/O’s header connectors that can be used for switches.
These 11 can be used as analog inputs and could have more than 1 switch per I/O with using resistor logic.

Software:
When I first started out with the gauge design I was using flight simulator FS2004 with LINK2FS to extract data.
I got every thing working and then decided to upgrade my flight simulator to Prepar3D v5.
With Prepar3D v5 not having a default Cessna 172, I then purchased A2A Cessna 172 add on.

To extract data from Prepar3D v5 I tried using a menu driven program called Nunalink which is an updated version of LINK2FS for use with Prepar3D v5.
While I was able to extract data, I had an issue with the A2A aircraft. The issue being that A2A manipulates the RPM data that is written to the RPM offset to give a more realistic flying effect. The RPM offset data has a delayed effect and thus it can’t be scaled to give a true reading. Same issue with the prop speed. What is required is to read the local RPM variable.
This lead me to purchase a registered version of FSUIPC for Prepar3D.
A registered version of FSUIPC allows you to write a Lua file that can send extracted data to Arduino processors.
I was able to write a Lua file to output data to Arduino’s similar to how LINK2FS does it, that being data the Arduino receives starts usually with < or ? and is followed by a letter and then a number which is of variable length. The number may be positive or negative number with or without a decimal point and no new line.

For the Arduino to read the data, is to read the first character which will be a ( < ?) .
For each of the ( < ?) you have a section that allows the next received character to be identified to what the data category is. Usually the CASE command is used for this..
Additional received number characters are added to a string variable which is the converted into an int type variable.
I am currently extracting 20 variables from Prepar3D.

With my gauge controller, I have used the I2C Anything library for the I2C communications.
I2C Anything allows for sending different types of variables rather than sending just a couple of char bytes.
Except for time data, I have used int variables, time data (hhmmss) requires a long int variable.
With I2C Anything library, you can send multiple variables. E.g. If you send 6 variables you need to plan and expect 6.

With the gauge controller having 6 Arduino Nano’s, one being a master connected to the PC and the other 5 nano’s being slaves controlling gauges. As each slave controls a number of gauges, I needed a method to distinguish the data received by the each slave nano. I used the method of sending 2 variables from the master Nano to slave nano’s. First variable being the description and second the data value. For the description variable I just use a single digit number.
As the two variables are sent to a specified nano, the same description variable value can also be sent to a different nano for another gauge.

When a slave nano’s receives the I2C two variables it occurs in an interrupt. A couple of “if” commands are used to determine what the data is for. The data is then re-mapped (scaled) and the gauge movement value is set.
After the slave finishes the interrupt and returns to the loop which contains the gauge stepper run command which allows the gauge motor to move.

Issues:
- One issue I found was the Airspeed and Turn coordinator gauges were moving a lot slower that they should.
These gauges were controlled by one Nano that also controlled the digital clock which has a touch sensor.
After some troubleshooting I found that using any analog reading (this includes libraries for touch sensor) slowed the program looping thus slowing down the stepper motor run commands.
I then changed the way I was reading the analog signal as per the below helpful information about using an interrupt and this fixed the slow gauge movement issue.
[meettechniek.info]

- Another issue I found was gauges that required electrical power to operate, the extracted gauge data value from flight simulator didn’t change when the master electrical switch was turned off.
Gauges displayed on the monitor do turn off. This issue required a small amount of programming to monitor the status of the master switch, if master switch is off then have some gauges to show a zero reading.

Controller Power Supply:
I have used a 230VAC to 5 volt 6 amp power supply. I measured the current to be around the 1.2 amps when using a bench power supply. With the future possibility of adding LED’s to the gauges for night flying , I choose a higher rated power supply as the price is not a lot more than using a 2 amp supply.

Gauge Enclosure Box:
The gauge enclosure box I made from 0.9mm thick aluminum sheet.
I cut, drilled and bent the aluminum sheet as to make a box with a sloped front.
I have also used 12 x M3 clinch nuts that I pressed into the aluminum for holding the two parts of the box together with M3 flanged headed screws.
The power socket and USB connection sockets are at the rear.
The enclosure box has a foot print of 425mm wide, 180mm depth, 150mm high and the front is sloped back 65mm at the top. The box has been painted black with being professionally powder coated.
------------------------------------------------------------------------------------
If you read this far you must have missed the warning about the technical stuff…….
Hopefully it explains how it all works…

Ender3 heatsinks on 5er diamond hotend (no replies)

heinz — Sat, 14 Aug 2021 05:17:52 -0400

The 5 lathed down E3D heatsinks are the most expensive parts of the 5er diamond hotend. The nozzle itself is only ~16$.
Ender3 heatsinks are small enough to fit side by side without any modification.
Not only cheaper, but it allows you to use different heatbreaks. The original ones are integrated into the heatsink and have 4.1mm bores to completely run the PTFE tube through.

I've been printing for a few weeks now with this setup, even challenging parts with lots of retractions and unused colors idleing and hydrolyzing,
but no clogs.
Even if you had any clogs, the grub screw mounted heatsinks can be more easily removed, without risking to unscrew the heatbreak from the nozzle.
Works with a 50x50x10 fan (2,4W) so far and no specifically designed fan shroud.

So if anybody is thinking about a 5in1out hotend, this might be a good start.

I tried with direct drives at first (see attached pic, don't get fooled by the printing time, eeprom was deleted inbetween)
Second pic shows the bowden setup, which is easier to work with and requires no adapters.

Smart, automatically watering plant pot with App-control :D (no replies)

reerb — Sat, 12 Jun 2021 10:43:11 -0400

I 3D printed this smart plant pot to save my drying plants. It comes with a capacitive soil moisture sensor, automatic watering, an integrated water tank, WiFi and data display plus push notifications in a smartphone app.

I have been working on this for almost 2 years now and have refined it many times. But I still find more things I would like to improve every day and this process is so much fun :D

A few days ago, I decided to start a youtube channel to share my learnings and create some step-by-step tutorials how to build this plant pot. Feel free to watch the first teaser :D

Youtube-Video LINK

[attachment 118353 Flaura1.jpg]

Sixi3 robot arm (1 reply)

i-make-robots — Sat, 08 May 2021 04:11:49 -0400

I was really inspired when the reprap project wanted to make self replicating things.
Ive been obsessed with making a low cost robot arm for about a decade now. Here's the latest news.

[www.youtube.com]

Where the Sixi2 was 1300 parts in 300 types, Sixi3 is 300 parts in 25 types.
By using a single actuator design I get low backlash, high power, and really affordable manufacturing. cheap, fast, strong: pick two. I chose cheap and strong.

My new goal is to make practical demos. Unlike people here, most folk need to be shown how it could be useful.

The firmware is based on Marlin, adapted with no heating or extrusion. I've got FK, IK, jacobian, inverse jacobian... I have challenges now with making it easy to drive the arm and adding new kinds of tools.

[reprap.org]

I'd love to talk to anyone that is eager to build an arm based on this design.

Printed with Wax filament Wax Casting Model (2 replies)

zurferjoe — Mon, 29 May 2023 03:28:43 -0400

I have at last got the wax filament to do what I want it to do !! Print ! This is my MV Agusta 175 cylinder head , my one was damaged beyond repair so copied the bits and here it is ! 60 hours printing later .And also the cam holder also replicated this first model was done in PLA to see about shrinkage etc . Hope it all works out and I can at last ride my bike ,

Monorail / duorail based on a 4040 V-rail (1 reply)

sjostedtio — Fri, 02 Apr 2021 01:55:33 -0400

Link to my video

I've started working on a Monorail / Duorail implementation inspired by Nicholas Seward:s Monorail.

I call it the Zipline, since it kind of looks like a monkey riding a zipline :)

However, I'm not sure I've gotten the geometry correct. I know it should be three intersecting, in my circle demo in the video it appears that all axis are moving in sync with each others...

The angles of the arms are 45 degrees on the side with two carriages and 30 degrees for the side with a single carriage.

I made a Benchy.... kind of fast X( (no replies)

Nitram — Sat, 23 Jan 2021 16:10:35 -0500

Video

Printer:
Anycubic Kossel Mini linear
Enclosed with plywood
E3d volcano extended
16 tooth pulleys and idlers
Reworked idler assembly
Gates 6mm belts
piezo probe Orion
5015 radial and 2 4010 fan as part cooler
tmc2130 over spi on all axes @ 24v on VM @ 1/8 microstepping 35v on E
PEI on glass
BMG clone extruder
E3d PT100 amplifier on aux board
Trigorilla as Main AVR (AtMega 2560)
Arduino NANO as auxboard
RPI 3B+
Klipper
Fluidd
and other tweaks...

#SpeedBoatRace

3D Printed Working Toy Bicycle (no replies)

Roberts_Clif — Wed, 18 Nov 2020 07:57:24 -0500

Before you think this was made by me this is Not my Design.
Though thought it was so amazing would post it for all to see.

You-Tube Link

Fox Theater (no replies)

jsw — Sun, 16 Aug 2020 16:38:53 -0400

Theater facade for O scale model railroad

Eat! (1 reply)

jsw — Sat, 01 Aug 2020 02:43:45 -0400

Streamline moderne diner for O scale model railroad. Layered construction, mostly Hatchbox silver ABS. Mk3S.

O scale model railroad service station (2 replies)

jsw — Fri, 31 Jul 2020 21:10:12 -0400

Based somewhat on the typical 1950s moderne-styled Sunoco stations that used to be commonly seen in the States.

The car in the photo is not of the correct scale. Car is S scale.

gassta1.jpg

Working on a new timelapse method (4 replies)

the_digital_dentist — Fri, 29 May 2020 15:34:34 -0400

First test- some bugs, but give me a day or two and they'll be dead, I think.
The extruder should be at the back of the printer in each frame.

https://vimeo.com/410890331

A new cell phone mount for my printer (1 reply)

the_digital_dentist — Wed, 15 Jul 2020 14:40:16 -0400

I made a new mount for my old cell phone so I can monitor long prints. It clamps to the t-slot frame of the printer.

Here's the original mount that sticks to the front cover of the printer.

IMG_20190308_222237.jpg

The new mount bolts to the printer's frame and works when the front cover is open- when I print TPU or PETG:

on%2Bprinter.jpg

More here...

Extra Cooling (2 replies)

leadinglights — Sun, 01 Mar 2020 18:59:50 -0500

I recently did a print that required more cooling than normal. Not wanting to use a desk fan I decided that I would have to make up something to do the job. Photo below of the fan thing cooling a problem print. and a photo of the business end

[attachment 114689 DSCF03471.jpg]

[attachment 114688 Fans2.jpg]

A small feature that I am quite pleased with is the use of a narrow but deep groove to hold the wires for the fans. The printing 'flash' allows the wired to be pushed in and then helps to hold them.

Thingiverse files, including a PDF with a diagram of the wire routing are at [www.thingiverse.com]

Mike