Shop OBEX P1 Docs P2 Docs Learn Events
A New CNC Build - 3D Printer - El Cheapo - Page 5 — Parallax Forums

A New CNC Build - 3D Printer - El Cheapo

1235711

Comments

  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 02:10
    You can only imagine my dismay...

    There is a 0.011" difference from front of the band saw table slot to the back. I expected a little difference, but nearly not that much. I am not really certain how much that would affect the cuts, because the blade will flex a little, but the guide would defintely teeter sideways while cutting or cut the aluminum at a slight angle. ARRRGGGG again.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 03:59
    Nearly two hours later and no serious progress....

    Once again, I feel that I am making this way too complicated....

    As you may have guessed, when I cut metal, I want my ends to be as square as possible. If I can get the metal cut to within 0.030" of the final length dimension, I'd be a happy guy, because then I could just use the sander to finish the ends off to the final dimension. Even 0.060" wouldn't be to bad.

    So I have decided to just screw two blocks of UHMWPE together at a right angle and run it along the edge of the table, as I am pushing metal through the blade. The edge of the table should be fairly close to being parallel with the blade, so this should get me in the ballpark. And if I go with slightly oversized holes, I should be able to rotate the top plastic block a little if necessary, for better alignment. Worst case scenario, I will just have to add a little more length to my rough cut and then sand off a little more to true it up.

    I believe this is my next plan of attack.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 05:28
    Saw blades is a rather vast topic in and of itself. This is the first time I have ever heard of a stainless steel blade. But I suppose that if you are using a lubricant with the cutting, it eliminates corrosion as an issue.

    Methanol as I recall can be corrosive to steel.

    ++++++++++++

    Regarding my search of a set of G-code file examples and ArduinoUNO G-code interpreter code, I am still working on a choice.

    The realities are these.
    A. Many 3-D printers are now being supported by proprietary G-code interpreter code. Why so? There has been a trend by some in the 3-D to move away from open source solution as a means to increase profit.

    B. Those that are using the Arduino's and providing open-source have moved to larger Flash RAM spaces as apparently the code compiled in C demands this.

    C. Open source LinuxCNC may work well with the Propeller, but not really provide all the added requirements for 3-D printing. 3-D printing really is a 4 stepper motor scheme plus heater control versus XYZ CNC being a 3 stepper motor schem plus spindle control. (3-D printing requires about 20% more equipment be supported).

    There is an ArduinoUNO firmware code for LinuxCNC, but no heater support.

    And so, while a Propeller 1 does have the one XYZ CNC project in OBEX, the 3-D printer world has pretty much ignored using one.

    @IDBruce
    I do wonder how you decided to start this thread on the Parallax Forums without taking into account the lack of Parallax products involved in 3-D printing to date. I still have no idea how feasible a Propeller 3-D printer controller board might be. One might just run out of code space regardless of the desire to make it work.

    In other words, embarking on a project without some survey of all the resources required may lead one to rushing up blind alleys. And we still haven't determined if this is one of those.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 07:35
    I do wonder how you decided to start this thread on the Parallax Forums without taking into account the lack of Parallax products involved in 3-D printing to date. I still have no idea how feasible a Propeller 3-D printer controller board might be. One might just run out of code space regardless of the desire to make it work.

    Please refer to these two links: EDIT: In reality, the only thing the Propeller needs to do is to read inputs and write outputs according to data sent to the Propeller. Everything else can be handled by an off-board computer or on-board if someone chose to attach one. Just send commands and let the Propeller interpret the command and perfom the desired action.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 07:51
    Well, I missed that entirely. Nothing available in OBEX, so I presume the code will not be shared. I just did a Google search of "Parallax obex 3-d Printer" and it mentions your reference.

    I did do another search of Parallax OBEX to "3-D printer" and nothing appropriate came back.

    So a reverse-engineering approach of an ArduinoUNO firmware still seems required. And, it remains entirely up to you and me to resolve the lack of Propeller firmware, and an adequate front-end for 3-D CAD design.

    OpenSCAD in Linux is what I am using, but with the Slic3r doing the file conversions anything you are presently using might be just fine.

    You do realize that 3-D printer project was canceled Oct 10, 2013 without funding?

    Nonetheless........
    Radiant Technologies seems to be selling an $1800 3-D printer, but with no mention of open-source code at their web site. Hard to say what all that indicates.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 08:06
    A couple of more points about the Lionhead 3D printer and 3D scanner.

    1. Parallax seems to have made the main boards they are using with the Propeller included.

    2. It seems that with their system, you can have the device actually scan an object in 3-D, and then the print a replica in 3-D. Thus you don't have to do tedious 3D Cad input as your starting point.... essentially they are building 3D copiers.

    It is quite impressive and worthy of mention as doing more than making just another 3D printer.

    But a word of caution...
    When I go to their Store page, I get a 404 message. seems they are not taking orders.
  • ercoerco Posts: 20,244
    edited 2014-04-04 08:36
    You cats do know about the sub-$500 printer unveiled at CES? Rich had a nice writeup about it somewhere in here.

    http://www.tweaktown.com/news/34878/xyz-printing-shows-off-its-sub-500-da-vinci-3d-printer-at-ces-2014/index.html

    Of course it's more fun to build your own, and you'll learn a lot. I love reinventing the wheel myself. It's just up to the individual, are you more interested in making a printer, or making parts.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 09:04
    Loopy

    It seems that you are really trying to sink your teeth into this G-Code. Here is how I would tackle the problem:
    1. Choose an open source vendor that handles STL files.
    2. Slice any 3D object and get the G-Code.
    3. Study the vendors G-Code commands and the outputted G-Code.
    I would say to start off simple... First examine an Eagle DRD file.... Once you undertand that completely.... You will have a much better picture.

    To my best recollection, the DRD file contains machine and tool information, as well as the system of measure, being metric or imperial. It lists all the specific sizes of drill bits needed to drill out a circuit board. When you progress through the file, it tells you what drill bit to use and the locations to move to and drill your holes. It tells you when to change a bit for different size holes. ETC... and it tells you when it is complete. In all honesty, I believe that is a very good place to start, because it is fairly simple and it will give you a broad overview, but while reading this, you must understand stepper motors and drivers, and be able to visualize how to write the code to achieve the desired movement of the components.

    As an example, if the propeller were to parse this file, it would read the various commands and:
    • Stop the machine when a drill bit needs to be changed
    • Proceed with drilling when a button has been pressed to indicate the proper drill bit is installed
    • Tell X and Y actuators to move to an exact location according to the resolultion of the drive, stepper, lead screw, etc
    • When reaching a specified location, start drill and tell the Z actuator to move to an exact location according to the resolultion of the drive, stepper, lead screw, etc and retract and stopping drill motor
    • Meanwhile monitoring a specified input to indicate a button press for halting all operations.
    And keep in mind, that a DRD file can be altered to add proprietary commands, such as M2000 (display message on LCD, "Drilling is now complete") While parsing the file, when the Propeller comes across the text "M2000", it runs the specified code to display the message on the LCD.
  • PublisonPublison Posts: 12,366
    edited 2014-04-04 09:11
    I think the chop saw Peter was referring to is called "Cold Blade"

    http://en.wikipedia.org/wiki/Cold_saw

    I have used these in the past at a welder we used to use. They make VERY nice cuts, but at a price.

    Jim
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 09:13
    Erco

    Pretty nice looking machine, thanks for sharing
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 10:30
    I don't see any merit in using Eagle files as examples. They are for 2-D product, printed circuit boards.

    The Slic3r software specifically requests print head information and other 3d printing aspects to compile a target G-code for what others have created controller boards for. I have already verified that G-code for printed circuit boards do NOT include new additions for 3-D printer heads.

    So I have to go another route -- reverse engineering the Arduino and selecting example files from Slic3r to have a file that all the code with be recognized and work properly.

    What file formats Slic3r provides are the real reference material, but I need the ArduinoUNO C code to verify that nothing is missed and to figure out what the code is suppose to do.

    And so that is my short list of options. The list of choices gets shorter when I try to use it with reference to firmware code that fits a binary into the 32Kb flash memory space on the UNO. I can't see trying to fit code from a larger flash memory into the Propeller.

    So it is still a tedious process of searching for information that reflects reality on the internet, not vapid generalization about g-code. I may have something tomorrow.

    +++++++++++
    That portion will resolve writing a g-code parse for the Propeller, but then there are several other aspects of building a program that will work. The data needs to be off-loaded from an SDcard, all the measuments in decimal mm need conversion to integers that reflect the reality of how far the stepper motors move, the interpolation of straight movements and interpolation of curves needs to be done, output to 4 stepper motors at a steady rate of speed coordinated with the print head control must take place.

    It is a lot. I as of yet have no experience with floating point on the Propeller, and all the interpolation of X, Y, and Z are new concepts to me.

    It may be stuff you have already learned and can help resolve. As it is, I have no idea of how long my doing it all on my own might take.

    Those are just the realities.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 10:37
    I made my first cut with the band saw....

    The test piece was some 1/4" X 1" - 6061 aluminum.

    It left a nice edge, but I must admit that it was cutting way too slow for my liking. I am hoping that a more aggressive blade will speed things up a wee bit, and of course it should, hopefully a lot.

    However, I must admit that it was a lot less strenuous than cutting it with a hacksaw :)
  • MicksterMickster Posts: 2,588
    edited 2014-04-04 10:45
    erco wrote: »
    You cats do know about the sub-$500 printer unveiled at CES? Rich had a nice writeup about it somewhere in here.

    http://www.tweaktown.com/news/34878/xyz-printing-shows-off-its-sub-500-da-vinci-3d-printer-at-ces-2014/index.html

    Of course it's more fun to build your own, and you'll learn a lot. I love reinventing the wheel myself. It's just up to the individual, are you more interested in making a printer, or making parts.

    Very professionally finished product....I'm thinking that these guys are a bit better equipped for manufacturing that what is being discussed here!

    Having stated that I know nothing about 3d printing, please edify me; why all the jaggies on those "finished" components? I looked at some printer demo's on YouTube and they appeared to be very smooth.
  • Alex.StanfieldAlex.Stanfield Posts: 198
    edited 2014-04-04 10:50
    ...
    What file formats Slic3r provides are the real reference material, but I need the ArduinoUNO C code to verify that nothing is missed and to figure out what the code is suppose to do.
    ...
    So it is still a tedious process of searching for information that reflects reality on the internet, not vapid generalization about g-code. I may have something tomorrow.

    +++++++++++

    Loopy, you may find this page interesting (http://reprap.org/wiki/G-code) it has the generic AND 3D specific Gcode explanations.

    Alex
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 11:09
    Thanks Alex, that is a big help.

    Reprap has three different forks of code that may all support various Arduino devices, but they all stick with the same g-code.

    The first I looked out only supports the ArduinoDue and I can't use it, but I am trying to get to the others to see if they are ArduinoUno able.

    Any I am spending a lot of time in the RepRap Wiki pages as it seems to be an excellent research resource for this project.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 11:52
    Very professionally finished product....

    Yes, it looks nice, but also appears to be made strictly from molded plastic. I bet it is not that durable.


    With three years of wood shop and experience working in a furniture factory, I think I can design and construct an appealing and sturdy cabinet, at least well enough for my prototype. As for being equipped, I can only imagine the amount of money spent on those molds, but I bet they spew them out, as compared to building cabinets.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 18:19
    Well, the search for which RepRap firmware seems to be over as it appears there is only one canidate that fits into less than 32Kb of ram space.

    Teacup Firmware will be used as a study example to develop a Propeller solution. There are a lot of others with lots of additional goodies, but it is best to start out small and humble.

    http://reprap.org/wiki/Teacup_Firmware

    And so, it now becomes the next task, read the Teacup C code and determine the following:

    A. If the whole Teacup C code can be ported to a Propeller with minimal modification via the Ticker.zip library or whatever Martin H. has updated this to.

    http://forums.parallax.com/showthread.php/148401-Propeller-Chip-runs-a-mostly-unmodified-Arduino-program-%28video%29

    B. If the whole of Teacup C code can not be ported in C, find the ways and means to meld existing OBEX contributions with information gained from Teacup to produce a Propeller project.

    BTW, Martin H. included the ability to use have an ADC attached to the Propeller in his Adruino porting library of C code. Good stuff. And Jazz has offer to help me if I start a separate thread in the Propeller 1 forum for GCC dedicated to the porting attempt.

    So I suspect I will be over there, rather than discussing petty details here.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 18:51
    Loopy

    Nice find and I hate to see you leave this thread, but good luck with your adventures. I wish you success.

    Bruce
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-04 19:06
    I am not leaving, just working on the porting of Teacup in the sub-forum where I can get good advice from the best people. I just won't be giving blow-by-blow reports here.

    I'll be working with the Propelleruino libraries and SimpleIDE with GCC. Thank you Martin H.

    parallax.com/showthread.php/148401-Propeller-Chip-runs-a-mostly-unmodified-Arduino-program-(video)
  • idbruceidbruce Posts: 6,197
    edited 2014-04-04 22:54
    I must say that cutting aluminum with the band saw has been a learning experience for me.

    To remain consistent, a tried cutting with several different blades having different TPI, but stuck with the initial test size, which was 1/4" X 1" - 6061 aluminum. I finally settled on a 6 TPI blade, which took approximately 30 seconds to cut through the test piece.

    As mentioned, I only have a limited supply of 1/2" pulleys, and about the best speed that I could obtain without ripping a pulley off another machine was about 1775 SFPM. And I have read that the max speed should be 1500 SFPM. But it is what it is....

    Also as previously mentioned, I have been using candle wax as the cutting lubricant and it appears to work rather well, leaving no aluminum buildup on the blade.

    It is not anywhere near the speed of a table saw, miter saw, or chop saw, but I feel much safer doing it this way. As compared to making an accurate cut with a hacksaw, I would say it cut the time in half, plus the heavy labor has been eliminated.

    So I suppose I should say that I am happy with the results.

    There are two other items worth mentioning:
    1. I used a standard wood cutting band saw blade and not a bi-metal blade. I am sure it would go much better with a bi-metal blade.
    2. It cuts much better with very light pressure as opposed to moderate pressure.
    Now it is time to fix the sagging sander table and recalibrate the setup.
  • $WMc%$WMc% Posts: 1,884
    edited 2014-04-04 23:26
    I have been porting aluminum cylinder heads for years....I use wax for the lube....Proving method.
    '
    I'm not sure how the wax would be applied on a band-saw.
    '
    When I'm porting a cylinder head, I dip the cutter in the wax then cut a little and reapply the wax.
  • $WMc%$WMc% Posts: 1,884
    edited 2014-04-04 23:51
    I'm working on a RPi/Prop set-up.
    '
    512K of ram with the RPi
    '
    Just found a Basic interper for the RPi
    '
    The RPi can do floating point.math
    '
    I'll be back later with the findings
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-05 00:24
    Well all this struggle with square cuts of aluminum 6061 stock is surprising.

    +++

    I just wanted to mention that I have downloaded two .zip files from GITHUB that are necessary to work with porting Teacup firmward to the Propeller. i think this is all of it.
    But there is also the download and install of SimpleIDE and that has a new home in learn.parallax.com.

    A. The Teacup firmware files as the software to port, latest version.
    B. The lib-Propelleruino files to add to GCC in Simple IDE.

    I mention these because you are NOT going to find them in Parallax downloads and you must have a Google account to be allows to acquire them from GITHUB.

    The proceedure is a bit round-about.
    You go to the GITHUB location for the code.
    Then you have to log in with your password for GMAIL.
    Then you have Clone the code for just you by filling out some boxes.
    And finally, have to go to your new Clone account.

    Once there, you will see all the files available, and you can requires a .zip file of everything for your computer.

    ++++++++++++++
    With all that you have two zip files on your computer than can be unzipped and used with SimpleIDE for porting, or just to read for reference without being on the internet.

    You will also have to download SimpleIDE from learn.parallax.com and install.

    I just wanted to make sure that any people that want to try this independently and are new to all this have the ability to locate all the resources. Good luck.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-05 04:33
    Loopy
    Well all this struggle with square cuts of aluminum 6061 stock is surprising.

    Cutting any material "SQUARE" and "TRUE" is not an easy task, and then add to that cutting material to a specified length. To understand exactly how surprising it can be, attempt the same test I performed ealier with 2" X 2" aluminum 6061, then you will understand. When building machinery, exactness is of high importance, especially when building machinery that requires a high degree of accuracy. Different machines require different degrees of accuracy, and even different parts require different accuracy. For instance, an actuator that simply moves a part to another location, such as to a conveyor, does not need to be all that accurate, but an actuator used for high resolution positioning, must be accurate. To make it accurate, the parts must be accurate.

    The rough cut is as nearly as important as the final finish, because you want it darn close to the final dimension, to decrease material loss and avoid unnecessary labor and other costs that can be incurred, such as wear on tools, rise in electricity usage, heating, etc... The longer it takes to machine a part, the more it costs.

    As mentioned earlier by somone else, alignment issues can be overcome with larger holes, but unless your parts are true and square, your alignment with other parts will never be true and square.
  • idbruceidbruce Posts: 6,197
    edited 2014-04-05 05:51
    Since I am now satisfied with the cutting action of the band saw, I finally got around to examining the sag of the sander table. Upon close inspection, it appears that the sag is not greater than 1/32 of an inch, and I thought it was going to be more like 1/16 - 1/8. The sander support is actually an old oak veneered TV cabinet, dating back to around 1950. The cabinet is fairly well constructed, and has a some joinery which I have never seen before, but the joint looks strong. I could be wrong, but I don't imagine the cabinet will sag too much more, however we all know that gravity can create havoc.

    Instead of spending my time trying to fix the sag, I have decided to rough cut all the pieces required for my project, and I will then recalibrate the sander, and finish off all the ends, within a short period of time. I may need to recalibrate the sander for other various projects from time to time, but I figure this way, I can start my build without spending too much time on rigid support, because this project, is of the utmost importance to me.

    In addition to that, I have arranged the band saw and sander within my shop area, to keep all metal debris within one location, or should I say two, because the lathe is in another location. At this point, things are looking pretty good to get started. However before getting started, as I mentioned earlier, I have some pressing issues that require my attention. Although I will probably be making my rough cuts as time permits.
  • MicksterMickster Posts: 2,588
    edited 2014-04-05 10:15
    $WMc% wrote: »
    '
    Just found a Basic interper for the RPi
    '

    Is the identity privileged? :smile:
  • TtailspinTtailspin Posts: 1,326
    edited 2014-04-05 13:47
    idbruce wrote: »
    I must say that cutting aluminum with the band saw has been a learning experience for me.

    There are two other items worth mentioning:
    1. I used a standard wood cutting band saw blade and not a bi-metal blade. I am sure it would go much better with a bi-metal blade.
    2. It cuts much better with very light pressure as opposed to moderate pressure.
    Now it is time to fix the sagging sander table and recalibrate the setup.

    I guess i did leave some things unsaid.. Like, bi-metal blade, and lite pressure on the blade. I do tend to take things for granted when given the chance. :)

    So, I should mention this, Be sure to vacuum those metal chips up, Do yourself a favor, and keep those fine, lite weight, electrically conductive, trouble making chips
    out of your motor brushes, I know they look like they are falling to the floor, but they can be small enough to get blown around and land in some inconvenient places...

    Not that you don't know this already.. but, I didn't, and gained some valuable experience, As I learned what a small handful of metal dust can do to a motor control PCB. :)

    Keep up the good work bruce.


    -Tommy
  • idbruceidbruce Posts: 6,197
    edited 2014-04-05 15:50
    Tommy

    Now you want me to hook up a shop vac system? This project will never get started. :( Just kidding, I hear what you are saying, I do my best to keep the dust and shavings to a minimum.

    To be perfectly honest though, I am more worried about the aluminum dust and alzhiemers. I mess with a lot of aluminum, and I really should be more careful. And now I am going to be sanding aluminum, what am I thinking?
  • $WMc%$WMc% Posts: 1,884
    edited 2014-04-05 17:56
    Sorry About that Mickster...I thought I put it up there.
    '
    https://projects.drogon.net/return-to-basic/rtb-download-and-install/
    '
    I haven't had a chance to try this yet.
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2014-04-06 03:22
    FYI, I started the thread for creation of a Propeller 3D printer board in the Propeller1 forum.

    After all, this is a Parallax supported forum, so it is only appropriate that the project should have a Parallax solution. IMHO Raspberry Pi or Arduino or the both together or other 3D printer interfaces are interesting in comparison, but that is about all.
Sign In or Register to comment.