Printed Paper WindMill v0 - Thingiverse
Printed Paper WindMill v0 by aubenc @ Thingiverse
Licensed under the Creative Commons By Attibution – Share Alike license
http://www.thingiverse.com/thing:19066
Material I used PLA (Makebot Clear PLA 4043D and also Princely Purple and Bling Bling Gold from Faberdashery) and I'm absolutely delighted on how much bot1334 loves to print with these.
Slicing/Profile Setup Wall Thickness vs Layer Height Maybe it's very logical and everybody knows already but I just find out that choosing a wall thickness that's exactly “n” times the layer height banishes weird skeinforge paths and the print end up looking awesome. That's why I was choosing a 0.9mm wall thickness sliced with 0.3mm layer height. I did a try with 1.05 wall thickness and 0.35 layer height and got the same very satisfactory results. BTW, for the rest of the profile parameters, these layer heights where too low but the wings printed very well. Also, when trying to slice this at a layer height of, say 0.33, skeinforge fails with an error. I do have try to properly adjust a bunch of other parameters to match this layer height without results.
Weird Path
Conclusion: set your layer height to “n” times your wall thickness or a close lower value! Support I turned support on for the Pin and I was setting it to 50 degrees. It's just 3 layers (or maybe 4) of support that are enough to hold the Pin and they are also very easy to remove. Fill I've been printing everything at 0.2 (20%) infill. Please take in account that I used PLA, for ABS I would increase the infill to 0.35 or 0.4. The nuts would print better and the threads, specially for the Pin, would be stronger. Speed The nuts get very easily sick at high speeds and their threads end up being a mess. My choice here for the Feedrate has been 20mm/s requiring some 4h to print everything.
Printing the Production Plates You may choose between printing each part one by one or use the “production plates” to print multiple parts in one build. Be aware that threads and tall thin walls may require different settings and printing everything in one build may end up being a waste of time and material. It's also advised to use at least 3 different colors, one color for two wings, another color for the other two wings and the third color for the rest of the “hardware”. It makes me happy to say that this is a “small printer friendly design”, the build envelope for the hole set is 90mm x 82.5mm x 100mm.
Please note that the production plates may not be centered in the build platform!
Printing One Part at a time Printing the wings Take a look to the job made by your slicer of choice. bot1334 is a Cupcake CNC and she's happy with Skeinforge Standard. The wings shape has a straight section and you want to check that this straight line is not rotated, it should deploy normal to the X axis or the print may show unwanted artifacts. I know that it sounds too weird, I did check the OpenSCAD code and it's a straight line perfectly oriented. I also tried to generate the gcode with Skeinforge 35 and 39... same result. The OpenSCAD contains a parameter intended to fix this, its name is “sra” and the provided STL have been generated with a rotation of 0.01 degrees.
1st Clear PLA wing in the left
Rotated 0.01 degrees. Purple wings.
Rotated 12.5 degrees. Clear PLA wing between the two purple ones
Printing “The Hardware” parts I did change the profile I used to print the wings, keeping everything the same but raising the layer height from 0.3 to 0.325. In fact, I was printing the wings ensuring that each layer was well smashed to the previous one. Start by the Pin, as commented before, I turned support on. If the whole thing looks like a big blob of material you may try to slow down the Feedrate and/or activate Cool and/or use a fan and/or reduce the infill (not bellow 20% for PLA or 35% for ABS). If the threads are not well defined you may need to reduce the Flowrate. In the other and, if they look super-well defined, this part may end up too weak and break!
Then print the Pin Nut. It should print fine with the same parameters that were used for the Pin. Known issues here are the threads being a mess of hanging strings. If that's the case, you need to slow down the Feedrate and adjust Flowrate accordingly. I use to print the nuts at 20mm/s to 25mm/s (it depends in many other parameters I may have been playing with). It may also happen that the speed is already very slow, say 18mm/s and there are still a little number of hanging strings in the threads (cut them out with a knife and forget about it). In this case you may try to lower down the layer height by 0.01. I'm sorry for the try & catch, the good news is that this nut is a small part that prints in minutes (even at very low speeds) and getting it right will help you with the rest of the parts. So you got the nut looking good but now it don't fits in the screw? Aaaargh! Too loose? The Pin, the Pin Nut or both where printed at too low Flowrate. Too tight? Too much Flowrate or low Layer Height. This nut has been generated with a tolerance of 0.5mm. The zip file contains this same nut together with three more with tolerances of 0.25mm, 0.75mm and 1mm. You decide if you keep on tweaking the profile or picking up another nut and hope it works. Again is well worth to get it right and keep this profile as a super fine tuned profile! The rest of the parts, Hub, Hub Nut and Fixture, should print free of issues.
Cleaning If your first layer is well smashed to the build platform you will need to do a quick clean in the hole of the Hub and the notch of the wings. Apart of that and the support for the Pin everything should be ready to assemble out of the printer.
Assembly
With all “the hardware” printed, cleaned (if required) and tested (fit for nuts and bolts, and Hub spinning freely in the Pin) it's time to put the things together.
Start with the Hub and its nut. Don't put it all the way down, just enough to be able to insert the wings in a way that they don't fall out. With the base of the Hub facing down, so the knurled nut is up, insert the chamfered edge of the wing's notch into the chamfer of the nut and swing the wing down to position. If the wing falls out you need to fasten the nut a little bit more. Once the four wings are in place fasten the nut to secure the assembly taking care that everything is well positioned: the flats in the wings' notches match the flats in the Hub's square. Add the Pin and its nut, use a couple of zip ties to attach the Fixture bracket to your support of choice. The Fixture should work fine for rounded and squared profile supports. If your support has a smooth surface consider adding a thin piece of rubber between the support and the bracket. Add the rest of the assembly and... CONGRATULATIONS you're ready to face the wind!
Changing Parameters from “paperWindMill_v0.scad” Not a “in depth” review of the parameters but “a few” notes thay may be useful. There's a huge amount of parameters, thus changing one may (most probably will) require changes in other parameters. They are organized in 3 sections of user defined parameters plus another section of computed parameters that should not be changed: Parameters (1) section contains the main arguments to be set: dimensions of the wings, tolerances and the slicer's rotation adjustment parameter discussed earlier in this document. Parameters (2) section is where you'll need to go to tweak the values for the different parts of the assembly. “Computed from above” section is where the parameters required for the shape of the wing are computed. Please note that the word “section” in that part of the parameters refers to the sections of the wing, not the different parameters sections. Sorry for the mess. Finally, all the parameters related to threads and knurls are under Parameters (3) section. Normally, the “changing parameters workflow” should be something like: 1. Use one of pps (paper size) or whg (wings height) together with wtk (wings thickness) to define the dimensions of the wings. 2. Change the wings rotation (wlf) if desired, uncomment the module “wings()” and render. 3. Select “top view” (F4) and adjust the thickness of the locking notch (dtk) and its offset (fst). Here it is important to see that the wings will not be where the notch is supposed to contact with the Hub and its nut. 4. Take a look to the Hub. Comment back the module “wings()” and use the module “section()”. The goal is to get the wings well balanced in the Pin, you want most of the weight of the wings between the two points of contact between Hub and Pin. To get this you may need to change the hbh parameter (Hub Base Height) and that may force the change of nch (Nose Cone Height) parameter. Other changes like the diameter for the Pin, the side length of the square in the Hub, ... will have more impact in the number of other parameters required to tweak. Good luck!
Sorry but somebody had to say this... This thing can spin very fast with little wind, ensure that everything that has to be locked it is locked, use glue if required. Do check from time to time that the Hub Nut is not getting loose (it should tend to get tight while spinning) and the same for the Pin Nut (this one could get loose with the vibrations). Also check that the Pin and the Hub are not wearing too much and thus there's the risk of having the Pin broken. This is intended to be a toy, take care of it so it is safe and have fun!
Licensed under the Creative Commons By Attibution – Share Alike license
http://www.thingiverse.com/thing:19066
Material I used PLA (Makebot Clear PLA 4043D and also Princely Purple and Bling Bling Gold from Faberdashery) and I'm absolutely delighted on how much bot1334 loves to print with these.
Slicing/Profile Setup Wall Thickness vs Layer Height Maybe it's very logical and everybody knows already but I just find out that choosing a wall thickness that's exactly “n” times the layer height banishes weird skeinforge paths and the print end up looking awesome. That's why I was choosing a 0.9mm wall thickness sliced with 0.3mm layer height. I did a try with 1.05 wall thickness and 0.35 layer height and got the same very satisfactory results. BTW, for the rest of the profile parameters, these layer heights where too low but the wings printed very well. Also, when trying to slice this at a layer height of, say 0.33, skeinforge fails with an error. I do have try to properly adjust a bunch of other parameters to match this layer height without results.
Weird Path
Conclusion: set your layer height to “n” times your wall thickness or a close lower value! Support I turned support on for the Pin and I was setting it to 50 degrees. It's just 3 layers (or maybe 4) of support that are enough to hold the Pin and they are also very easy to remove. Fill I've been printing everything at 0.2 (20%) infill. Please take in account that I used PLA, for ABS I would increase the infill to 0.35 or 0.4. The nuts would print better and the threads, specially for the Pin, would be stronger. Speed The nuts get very easily sick at high speeds and their threads end up being a mess. My choice here for the Feedrate has been 20mm/s requiring some 4h to print everything.
Printing the Production Plates You may choose between printing each part one by one or use the “production plates” to print multiple parts in one build. Be aware that threads and tall thin walls may require different settings and printing everything in one build may end up being a waste of time and material. It's also advised to use at least 3 different colors, one color for two wings, another color for the other two wings and the third color for the rest of the “hardware”. It makes me happy to say that this is a “small printer friendly design”, the build envelope for the hole set is 90mm x 82.5mm x 100mm.
Please note that the production plates may not be centered in the build platform!
Printing One Part at a time Printing the wings Take a look to the job made by your slicer of choice. bot1334 is a Cupcake CNC and she's happy with Skeinforge Standard. The wings shape has a straight section and you want to check that this straight line is not rotated, it should deploy normal to the X axis or the print may show unwanted artifacts. I know that it sounds too weird, I did check the OpenSCAD code and it's a straight line perfectly oriented. I also tried to generate the gcode with Skeinforge 35 and 39... same result. The OpenSCAD contains a parameter intended to fix this, its name is “sra” and the provided STL have been generated with a rotation of 0.01 degrees.
1st Clear PLA wing in the left
Rotated 0.01 degrees. Purple wings.
Rotated 12.5 degrees. Clear PLA wing between the two purple ones
Printing “The Hardware” parts I did change the profile I used to print the wings, keeping everything the same but raising the layer height from 0.3 to 0.325. In fact, I was printing the wings ensuring that each layer was well smashed to the previous one. Start by the Pin, as commented before, I turned support on. If the whole thing looks like a big blob of material you may try to slow down the Feedrate and/or activate Cool and/or use a fan and/or reduce the infill (not bellow 20% for PLA or 35% for ABS). If the threads are not well defined you may need to reduce the Flowrate. In the other and, if they look super-well defined, this part may end up too weak and break!
Then print the Pin Nut. It should print fine with the same parameters that were used for the Pin. Known issues here are the threads being a mess of hanging strings. If that's the case, you need to slow down the Feedrate and adjust Flowrate accordingly. I use to print the nuts at 20mm/s to 25mm/s (it depends in many other parameters I may have been playing with). It may also happen that the speed is already very slow, say 18mm/s and there are still a little number of hanging strings in the threads (cut them out with a knife and forget about it). In this case you may try to lower down the layer height by 0.01. I'm sorry for the try & catch, the good news is that this nut is a small part that prints in minutes (even at very low speeds) and getting it right will help you with the rest of the parts. So you got the nut looking good but now it don't fits in the screw? Aaaargh! Too loose? The Pin, the Pin Nut or both where printed at too low Flowrate. Too tight? Too much Flowrate or low Layer Height. This nut has been generated with a tolerance of 0.5mm. The zip file contains this same nut together with three more with tolerances of 0.25mm, 0.75mm and 1mm. You decide if you keep on tweaking the profile or picking up another nut and hope it works. Again is well worth to get it right and keep this profile as a super fine tuned profile! The rest of the parts, Hub, Hub Nut and Fixture, should print free of issues.
Cleaning If your first layer is well smashed to the build platform you will need to do a quick clean in the hole of the Hub and the notch of the wings. Apart of that and the support for the Pin everything should be ready to assemble out of the printer.
Assembly
With all “the hardware” printed, cleaned (if required) and tested (fit for nuts and bolts, and Hub spinning freely in the Pin) it's time to put the things together.
Start with the Hub and its nut. Don't put it all the way down, just enough to be able to insert the wings in a way that they don't fall out. With the base of the Hub facing down, so the knurled nut is up, insert the chamfered edge of the wing's notch into the chamfer of the nut and swing the wing down to position. If the wing falls out you need to fasten the nut a little bit more. Once the four wings are in place fasten the nut to secure the assembly taking care that everything is well positioned: the flats in the wings' notches match the flats in the Hub's square. Add the Pin and its nut, use a couple of zip ties to attach the Fixture bracket to your support of choice. The Fixture should work fine for rounded and squared profile supports. If your support has a smooth surface consider adding a thin piece of rubber between the support and the bracket. Add the rest of the assembly and... CONGRATULATIONS you're ready to face the wind!
Changing Parameters from “paperWindMill_v0.scad” Not a “in depth” review of the parameters but “a few” notes thay may be useful. There's a huge amount of parameters, thus changing one may (most probably will) require changes in other parameters. They are organized in 3 sections of user defined parameters plus another section of computed parameters that should not be changed: Parameters (1) section contains the main arguments to be set: dimensions of the wings, tolerances and the slicer's rotation adjustment parameter discussed earlier in this document. Parameters (2) section is where you'll need to go to tweak the values for the different parts of the assembly. “Computed from above” section is where the parameters required for the shape of the wing are computed. Please note that the word “section” in that part of the parameters refers to the sections of the wing, not the different parameters sections. Sorry for the mess. Finally, all the parameters related to threads and knurls are under Parameters (3) section. Normally, the “changing parameters workflow” should be something like: 1. Use one of pps (paper size) or whg (wings height) together with wtk (wings thickness) to define the dimensions of the wings. 2. Change the wings rotation (wlf) if desired, uncomment the module “wings()” and render. 3. Select “top view” (F4) and adjust the thickness of the locking notch (dtk) and its offset (fst). Here it is important to see that the wings will not be where the notch is supposed to contact with the Hub and its nut. 4. Take a look to the Hub. Comment back the module “wings()” and use the module “section()”. The goal is to get the wings well balanced in the Pin, you want most of the weight of the wings between the two points of contact between Hub and Pin. To get this you may need to change the hbh parameter (Hub Base Height) and that may force the change of nch (Nose Cone Height) parameter. Other changes like the diameter for the Pin, the side length of the square in the Hub, ... will have more impact in the number of other parameters required to tweak. Good luck!
Sorry but somebody had to say this... This thing can spin very fast with little wind, ensure that everything that has to be locked it is locked, use glue if required. Do check from time to time that the Hub Nut is not getting loose (it should tend to get tight while spinning) and the same for the Pin Nut (this one could get loose with the vibrations). Also check that the Pin and the Hub are not wearing too much and thus there's the risk of having the Pin broken. This is intended to be a toy, take care of it so it is safe and have fun!
