New Years Resolution(s)!

Happy new year to all of you readers and random clickers! First lets get some 

awesomeness going. Y'all are here for the pictures any way...so am I really. If you don't have photographic proof, it didn't happen. I randomly posted an ad to craigslist asking if anyone could use a sign made for free. Man that was crazy. I got about 20 emails for the hour that the ad was up. Most were just people asking for random crap that didn't stand out too much. One was special. One screamed that it needed to be done.

A lady emailed asking if I could make a sign in remembrance of her daughter and if there was going to be a sign made...that one was going to be it. She didn't have a specific design in mind so I got to play around with the design. One request was to have a butterfly in the center of the sign. To the right is a shot of the initial test cut I did of the butterfly. This was just a generic Google image converted into a vector drawing. I may do a quick tutorial on how to do that later if there's interest. It turned out much better than I had anticipated. I was for sure that the intricacies of the wings would be lost but they worked out well. The only downside was that in order to get enough cut divergence to show the detail it forced the sign to be almost 2'x2'. That would be bumping up against the Y limit of the CNC.

In the end it worked out. All of the lettering was done using a 45 degree router bit. I still havent broken down and ordered a set of real CNC bits other than a couple for circuit board milling. The biggest thing that I notices about the sign that I would have liked to change is the font. I used a font found in AutoCAD and it just wasn't desined to go that big. You can see the individual vectors in many of the letter. The heart outline was done using a 1/4" endmill. As for the heart shap of the sign I found it faster and easier to draw that up in AutoCAD, add the butteryfly vectors, and then import it all into MasterCAM to create the tool paths. All in all it took me forever to get it finished with everything else that was going on but the lady was very nice and understanding since I was doing the work for free. Total cost of the project was $20 bucks for a piece of cedar. This doesn't include the time I spent joining the wood to make a sign blank or the time I spent designing the sign but all that was a learning experience that I couldn't have gotten any other way. I would say that the self test was PASSED...mostly. There's still another sign that I've agreed to make and have just let fall by the wayside due to a whole host of other things. Fortunately the dudes I promised it to have been way uber patient or just completely written it off. I hope to get around to making it... maybe I'll finish it up one day and surprise them.

The most recent CNC news is that I've moved the machine to my house! This was no small feat though. I didn't even want to take before pictures. It was impossible to walk into the office area and even if you could get into the office you couldn't have walked into the works space in the garage from the office due to all the stuff that was in there. It is still a little cramped but fortunately the CNC is on wheels so if needed it can be moved to the main section of the garage as long as my wife isn't parked in there...gah so demanding! The idea is to run the monitor in to the office so you can close the door while its running. eBay supplied a 25' long VGA cable for less than the gas that I would have spent trying to find on locally. I have a wireless keyboard and mouse but the one thing that's bothering me is zeroing the machine. So far this has proven to be my weakest skill. 

There is still more that needs to be done to get up and running again but hopefully now that the machine is in my back yard I'll get time more often to play with it. I've got a couple projects I've been kicking around. The first of which is building the tool paths to mill out the lower receiver of an AR-15. I'm not one of the fanatical fans of the AR-15 at this point but its a massive undertaking to build one from scratch. If you dont know me by now, you'll soon understand that I often do things for the wow factor. You walk into a room and say I climbed Mount Everest and people shut up an listen....baring those who are deaf or who don't know what Mount Everest is. This is no means my Everest but its a feat non the less.

The other project  I will be throwing at you guys is a cube. Much less awesome factor but it targets building that positioning skill that I mentioned I was lacking earlier. Here's a link to an instuctable that I'm going to try to follow.


http://www.instructables.com/id/Turners-Cube-A-Beginner-CNC-Milling-Project/

And for those of you to lazy to click on it, here is a stolen image from the instructable. I'm going to first try this out of wood because its cheap but if it goes well I might shell out for the aluminum because it looks pretty. 

Now that were way down here and most people have stopped reading I'm going to hide a couple of my new years resolutions. I hate announcing these things cause they always seems to be the first things to get swept under the rug and not talked about again after they are made. 

1. First and foremost I'm going to lose weight. As generic and mundane as that sounds its going to happen and I need to write it down to in an effort to commit to it. Here's what I'm going to do in order to achieve this goal.
• 5:30 a.m. outdoor runs every Tuesday and Thursday
• Indoor run on Sunday
• Nike+ fuelband (2500 fuel points a day)
• Libra weight tracking app (Android)
• This shows trend lines and tries to average out daily fluctuations
• Weighing portions when possible
2. Next is the cooler and less lame resolution. I've set an alarm on my phone to write a blog every other week. Starting today I'm going to start this and we'll see how far it goes. I hope that they are all going to be CNC related but I doubt that with school starting back up that I'll be able to generate that much CNC related news. Either way there's always something techie or nerdy enough to hit the blog.
   
   There is is. Whew. That's rough to write that stuff down but its done. I can always go back and edit it out later! 

See you in a coupe of weeks.

Graduate school and xbee HOWTO

After many months applying for jobs and trying to figure out what I'm going to do next, Holley and I have both decided to go on to grad school!  I'll be posting about my ventures into the world of electronics design here starting with today. A part of my graduate project is going to deal with wireless communication and mesh networking.

That being said, I picked up a couple xbee modules and USB adapters to play with and get familiar with using them. Being a Linux nerd the majority of my time is spent trying to figure out how to make Windows solutions work in the Linux realm. Yesterday was no exception.

Here we have the xbee 900 pro digimesh something or other. It works on the 900 MHz band and claims a  range of up to 6 miles. Wow. That's way overkill for my needs but that is what was laying around the lab to play with. To my understanding all of the xbee products pretty much work the same with a few minor differences. I'll figure that out later down the road. First task is to get them talking.

I picked up two of the USB carrier boars along with the RF modules just so I could get them talking via USB and not have to worry about programming at first. After doing a little research I knew that getting them to work under Ubuntu wasnt going to be plug and play. The picture to the right shows the inhouse board the department made as well as an evaluation board from Digi.com. both work in Windows using X-CTU at a baud rate of 4800. X-CTU is the software utility that Digi uses to connect to and configure the xbee modules. Getting that up and running is a must if we want to do anything.

There is no Linux alternative to X-CTU so my first thought was WINE. Keep in mine I'm not particularly fond of WINE. It works sometimes and most of the time it doesn't. For this particular situation it does with a great deal of modification. I'm not going to pretend that I knew what to do to make U-CTU work but my googling skills are top notch. The following sites don't necessarily provide the working solution but go a long way to help piece together one that does. 

http://wire.less.dk/?p=162

http://aeturnalus.com/robotics/mapping-ftdi-to-files-with-udev/

The problem with the wire.less link is that /dev/ttyUSB* is owned by root. So unless you run WINE as root all the time (dont do that) then you have to change the permissions of the device every time you plug on in. It possible and only requires one command but its a pain to remember all that. This is where the second link comes in. Its possible to set rules for specific devices. When they are plugged in you can mount them with a special name or in a special place or some other more complicated things. These are called udev rules. They reside in /etc/udev/rules.d/ . The hard thing is figuring out how to write these rules so that anytime you plug in an xbee it makes it readable and writable by users other than root.

Lets start with the easy stuff. First you want to install WINE if you don't already have it. The software manager of your distro will have it so go get it.

Nect go download the X-CTU installer from digi.com. Heres the link that I used.

Open up some terminal and from your home folder

cd /.wine/dosdevices

ln -s /dev/ttyUSB0 com5

ln -s /dev/ttyUSB1 com6

ln -s /dev/ttyUSB2 com7

ln -s /dev/ttyUSB3 com8

Those steps are literally cut and pasted from the wire.less link. What they do is link the COM ports windows normally uses to the /dev/ttyUSB ports we use in Linux. So when X-CTU goes looking for COM5 its directed to /dev/ttyUSB0 which is where it needs to be looking in Linux.

Now go find that .exe installer file we downloaded. Right click on it and set the permissions to executable. Then double click on it and let it install. It should do it automatically if you installed WINE correctly.

That was the easy part. Now we need to make it to where the user can go find and use /dev/ttyUSB whenever one of these devices is plugged in. I'm big on changing permissions as little as possible. Its easy to screw something up or create a security problem you don't even know about so I only want this to change permission when an xbee is plugged in. Otherwise leave it alone.

First thing you need to find an unique id to the ftdi chip that runs the USB converter.. This allows the computer to recognize the specific device the rule is for and not just blanket the rule over every device that's plugged in. I chose to use the vendor ID but you could just as well use the product or serial ID. I actually started with the serial ID but it would require an additional line be added to the rules for each different board. This is the most specific and secure but the least convenient. 

Open up a terminal and

udevadm info -a -p $(udevadm info -q path -n ttyUSB0) | egrep -i "ATTRS{serial}|ATTRS{idVendor}|ATTRS{idProduct}" -m 3

Write the values that come up. My idVendor was 0403. If yours is different youll need to change that in the rule.

So next we want to create the rule file. In the aeturnalus link they names their rule file 10-ftdi.rules. After doing some reading I saw it was recommended to name the rule file with a letter first so it is read last and overrides any previous rule. This makes sense due to the fact that the computer would read files starting with numbers first. So the next line is slightly different that what is seen in the aeturnalus link. 

Open up a terminal and

sudo touch /etc/udev/rules.d/ftdi.rules && sudo chmod 644 /etc/udev/rules.d/ftdi.rules

This is two commands in one line. it creates the file as well as modifies its permission. Next we want to add out rule to the file. Ubuntu uses vim and kubuntu uses kate to do text editing so pick your flavor. Im using kubuntu at the moment so

Open up some terminal and

sudo kate /etc/udev/rules.d/ftdi.rules 

This should open up a graphical text editor with a blank sheet in front of you. Don't panic. We just created this file. There shouldn't be anything in it.

cut and past this line in the file

KERNEL=="ttyUSB?", ATTRS{idVendor}=="0403", SYMLINK+="XBEE", MODE="0666" 

If you idVendor was different you'll need to change that here. 

Finished. Thats it. Now every time you plug into the usb port with a xbee through an ftdi chip it should set the permissions so that X-CTU can read from it. Here's a final shot of my Linux laptop communicating to my windows machine through a pair of xbees.

Found a couple spare moments!

Man, I uploaded these pictures from my phone to an unpublished blog post back in March.... Its now May if you haven't noticed. They need to hit the web though. This is a 3" interlocking cube that I designed with an interference fit. It worked and it also taught me that the Geko drive needs a cooling source or it will begin to skip steps. When I say cooling source I really mean a box fan that I put under it.






Ive forgotten what all Ive talked about on here but this is the newest purchase and build that Ive done to the CNC. Its a Porter router and DIY mount that works very well. I still have the mount for the smaller Rotozip router. I'm hoping to be able to use the smaller setup to make PCBs but I haven't been successful yet but that's mainly due to a couple of variables I will go into more detail in a later post.

In the upcoming weeks Ive got a new project I would like to start tossing around in CAD. I'm thinking about designing a boomerang. I used to make them by hand when I was young and this sounds fairly simple. I also have some pictures to throw up of a bio-reactor that I designed and cut out for work.. That design may have alternate uses in future projects. All in all, now that graduation is over, school it out, and my internship at ORNL just ended, I should have a nice amount of free time on my hands.

Ive also registered the CNC on www.HomeCut.org. Thats a cool site that allows you to register your CNC so others can contact you with project ideas. Pretty cool. That's it for now. Go build something.

Stop the CLOCK!!!!!!!!

34 : 23 : 33 : 33 Day Hr Min Sec

That's it! I have met my goal. I am almost a full 34 days past due but I'm finally there. A lot has gone on in the last two weeks so let me hit the highlights of that real quick.
I got the new hardware in not last Friday but the Friday before. As you can see I took a shot of the new motor vs. the old one. Its almost twice as large. I got all the hardware hooked up and installed the latest version of Linux CNC ( formerly Linux EMC) and did what anyone would do in their right mind. I found some double sided tape and a Sharpie and hit play. It came out twice as large as it should and reversed in the X-axis. Both easy fixes in EMC. I kept the rev per inch on the rod at the default of 20 in order to keep things nice and big since I had a whole poster board to work with







I played with a couple other test files including a couple spiral patterns and a snow flake. It was the greatest feeling on earth to see this thing run.


I spent several hours just going through the software and getting acclimated to the controls and which direction was what on the screen as opposed to on the table. My dad hung the posters from the wall in the garage like I was in kindergarten. It awesome. My parents weren't my parents during that period of my life so its fair game now. That was all the excitement I could handle last weekend.


I spent the entirety of last week thinking about the CNC. It was so bad I had dreams about it. I kept looking at parts from my small engines class and thinking if I could draw them up in AutoCAD. I picked up a copy of MasterCAM and installed it. Sadly there is almost no information on how to use it. There are two or three books on Amazon, a couple online classes you can pay for, and a couple DVDs you can buy but nothing for the person that's not sure what sources of information are really good and which ones are garbage. I finally came across this site;

http://www.cudacountry.net/html/mastercamx_toc.html

Those are free PDFs and are extremely well documented. They have window screen shots, button screen shots. Its simple to follow. I cant imagine how long it took to create them but man am I glad that I found them. I ran through a couple of the first CAM guides and waited for the weekend to roll around.

Once it did, we built a mount for a supercharged Dremel looking tool. Its made by Bosch but I didn't thing it was going to handle this job. I was glad to be wrong and another moment even better than what I mentioned came about. I crossed the line that I set for myself shortly after starting this blog. I cut a tangible object.  I cut an EMC logo in the double size and then recalibrated to cut a logo in a one-to-one scale. The one-to-one scale is sitting in front of me as I type. I may sleep with it under my pillow. The larger one stayed with my dad and I wouldn't be surprised to find it hanging on a wall somewhere the next time I go over to my parents place.

Here's a video just to show off. Enjoy the screaching sounds. I wanted to cut the sound and have it be a silent film but apparently that's not an option. YouTube should get one that but until then turn the volume down.

Where to now that I've met this goal? Well I'm going to take it all apart and sell it for scrap....Nope. Not even funny. Even as I look at the screen I'm not laughing. Well I've got some ideas. I'm thinking about making an enclosure to hold all my air ride equipment and sound equipment for the Blazer. Im also thinking about guitar bodies, and missing dashboard buttons, and signage, and random small engine parts.

Short Update!

 
23:14:07:13 passed the deadline

I'm now five hundred bucks poorer! Why am I sooo excited about that you ask? If you'll take a look at the nice picture on the right ----------------------------->

You will see that the money will soon be converted into a set of three high torque stepper motors, a Gecko g540 motor driver, and a matched power supply to get this party started! Hell yeah I'm excited. This should make getting up and running a lot easier. No more dealing with the Arduino and those tiny tiny stepper motors I was playing with. On a side note I will most likely be building a 3d printed or something smaller when I get the CNC up and running. So in the end those stepper motors will have a home....even if I end up selling the whole thing.

For those wondering, I got the job at ORNL and that's the main reason I have the excess financial funding needed to acquire these final pieces. There are upsides and downsides to this deal though. The upside is that I have a job and am being paid good money. The downside is that this is only a spring term internship with the possibility of extension. So the hunt is still on for a permanent job once I graduate. I've got some leads but nothing solid at this point. Fortunately my resume is getting full of all kinds of different and important acronyms that make me a strong candidate for many positions.

Tip for those looking for a job in the science/research field. Mention specific instruments you have worked with in your resume. HPLC, GC, NIR, Bioflo reactors, Fermentors,  Mills, etc, etc. It tells the reader what kinds of things you can already do and also says that someone trusted you with high dollar research equipment.

More to come when it unfolds.

23:14:20:37 passed the deadline

Late...again

Well its now 55 minutes into the new year. 55 minutes past my CNC dead line. Sadly I didn't make that dead line but I am close.
What do I have? I have a CNC table with working X and Y axis. The Z axis wouldn't run due to the less than capable stepper motors. the final thing I have moving forward us a clear path and end goal.
That path is to buy a gecko 540 drive and stepper motors. This will run over $500 but its going to be controlling a $1500 table so that price is on par. I've been hired on at ORNL so I will be able to generate the capital required to finish this project up. So here's the big question. When? That's the hard line. This next semester is going to be jam packed. Senior design. The ORNL internship. Searching for a post graduation job. The list goes on.  My initial though was to push the deadline out till the end of February but to be safe I'm going to say March 15th. This gives me time to gather the funds, wait for it to be shipped, have some unknown delay, get it together, and crank out a piece.
There it is. I'm happy. Big things are happening this year. I graduate. I get a real time job. I finish the CNC.
Thanks for reading
Steve

IT'S ALIVE!!!!

Yeah that title is cliché as all get out but deal with it. :-) Currently I'm excited that the projects not dead. That sentence needs a little back story. Last week (that time I should have been studying for finals) I got everything hooked up and ready to go with the motor drivers and Boarduino sled I constructed to hook all those motors up in a cleaner manner. I eventually got enough spare time to get it all hooked up to the CNC and with bated breath to see if it would work. Yeah I spelled that right too! I googled it. Anyway, the whole thing went to hell as soon as I gave it the command to move any of the axes. The motor were skipping steps like crazy. There just wasn't anything that was going to make it all work.

I didn't throw anything. I didn't even get super upset. This project has been going on for years. Literally. Its just another hurdle in the road I told myself. So I left the garage and started looking for the solution. The Gecko g540 is probably the most recommended motor driver  over on CNCzone.com so thats where I started. Ahren over at CNCRP has a kit that comes with everything to fit the need I was having.

http://www.cncrouterparts.com/3-axis-electronics-kit-p-74.html?cPath=22

For the small price of $520 the quick and easy solution could be mine. Right? Well no. Since Ive spent the majority of my project fund on the table itself I don't have enough left over to purchase such a fine piece of equipment. More on that later though.

Back to the drawing board I went. (another cliché phrase, is this guy kidding?)  The RepRap stepper driver V2.3 that I've built and am using is built around the Allegro A3982. Not much information is out there about the max specs of the v2.3 driver boards but Allegro provides specs for their chips so I just assumed that the circuit was built to utilize its capabilities.

"The A3982 is a complete stepper motor driver with built in translator for easy operation. It is designed to operate bipolar stepper motors in full- and half-step modes, with an output drive capacity of up to 35 V and ±2 A. The A3982 includes a fixed off-time current regulator which has the ability to operate in slow or mixed decay modes."

Hey I've got four, count them four, of these little drivers and I planned on frying one in the process anyway. After frying a HobbyCNC driver back in the first days of this project Ive come to expect to break a few things from time to time. So I scrounged up the biggest laptop supply I could find, cut the end off of it, found another old ATX power supply, cut the end off that too. Then I wired it all together and gave it a go.

After all that you're wanting me to say it ran perfect and I went on to cut something out or anything spectacular. Well not exactly. It bound up but it did have moments where it ran like it should. Even after cleaning the screw, the rails, and a light WD-40ing, it still bound in places going at the G0 rapid linear motion speed. At the slower G1 speed it did much better. This signifies that the motors are at the edge of their performance curve. Fortunately, Im still only running them at a fraction of what their capable of. Currently Ive got a 20v 6A power supply. Meaning that if I can get a hold of one closer to 35V it should make enough difference to get back on track to hit that dead line.

 

For now, enjoy this video of my latest Rube Godlberg project that knocks over a can of WD-40

Making the stable connection

 Its like one of those movies that starts with the ending scene. I put together a "Shield" for the Boarduaino to connect the motor drivers. Now its one step further away from looking so ghetto. Unfortunately one of the shafts of the steppers is slightly bent. Its not something I've noticed before so I'm pretty bummed. Never the less I'm hoping that its not enough to matter with helical cut motor couplers. 

I'm weird about symmetry so three motor connectors just wouldn't work. Which is cool because I have four stepper drivers to. Maybe one day Ill incorporate a fourth axis. For now its just a spare connector on the proto board since there's no designated pin for a fourth axis in the GRBL config file.

 There are 4 pins required to run each of the stepper drivers. 2 ground pins, a STEP pin, and a DIR pin. Its complicated to route the wiring on the bottom of the proto board without making it look like a rats nest. A finger nail clipper makes a great wire cutter. Its easy to handle and if you have a hang nail its good for that too. :-)









Here's a picture of the final wiring on the underside of the proto board. It's all pretty close but there's no shorts. The wiring will be hidden in the end when the whole thing is mounted in the computer case.

Hopefully Ill get a chance to connect this all back to the table later this week. Ive got some finals and a proposal to give to the Biosystems Engineering Department. Here's a couple pictures for you interwebians.

Trying to revive the dead.

As an engineer and resident nerd, when a piece of electronics breaks the first thing that comes to mind is can I fix it? I've managed to bring a sound system back to life with a single capacitor. Actually I met a guy while working at Radio Shack that was buying LCDs on Craigslist and fixing them by replacing capacitors on the main board.

So when my router stopped working I decided to crack it open. Netgears routers are simple to take apart if you have a set of torrix bits. There are four screws under the feet. Typically these feet are glued in place but it seems that Netgear may reuse these shells or repair them often because the feet are hinged rather than stuck in place with an adhesive.

 Looking at the internals, its pretty easy to look around. There's no metal covers hiding anything. Everything is open for the pokin'....Sorry. Ill never say that again. Anyway upon inspection I saw that it indeed had a blown capacitor. The green one on the left had a bulge in the top of the canister.





This gave me an opportunity to play with the macro setting on my Nikon D3000.  Enjoy the eye candy. It hard to tell its blown without comparing it to the brown cap beside it so I took a picture with both of them in the frame.

 You can see how flat the one on the right is. Its almost convex compared the blown capacitor.
Radio Shack is way overpriced on parts but the convenience of being able to go down the road and pick it up the same day makes paying 1.49 for a single capacitor ok. Its slightly larger than the original cap due to the fact that it has a higher voltage. When dealing with capacitors you can substitute one with a higher voltage but its important to keep the farads the equivalent. They balance the impedance of the circuit as well as other fun stuff.

So after all this did it work? Sadly it did not this time but it did give me a good reason to upgrade to a N router. It also may provide some useful spare parts in the future. There's several SMD LEDs, resistors, and other junk that could be salvaged with the aid of my high tech solder reflow system (an electric skillet). For now I'm going to keep it intact just in case I come up with another idea to fix it. Drop me a line if you have any clues.

Placing Random QRcode on walls!

HEY! Thanks for scanning one of my secret QRcodes strategically placed around UT campus. You are now one of the elite few that will be counted in this social experiment. Years have been added to your life. I swear.

If your interested in electronics and CNC related projects feel free to check out the rest of my blog!

Hope you enjoyed the link! If you're interested in making your own QRcode you can visit some one else's website.


http://qrcode.kaywa.com/

Oh, and feel free to comment on where you found the QRcode placed. I gave several to friends and told them to spread the love.

Steve-O

Mechanical Portion....CHECK

The CNC has a method of travel now! I finished up adding the power-train components I picked up a couple of days back. I'm pretty excited about that. The X and Y axis are pretty smooth. The Z axis is slightly rough to turn but this may be to the weight of the carriages and router plate. I'm afraid that the motors that I have wont be able to work that axis. We'll see when we get there. For now, we celebrate the victory at hand.

Now I'm going to make you quit being lazy and read from the other side of the screen.

A small list of items that will need to be acquired is growing as we move forward in the project. First is a table surface. A simple half sheet of MDF should do the trick there. Next is a router mount. I haven't decided on what I'm going to use for a cutter but its going to start out small. I have available to me something about twice as large as a Dremel but half as big as a typical router.  Its my dad's. This will most likely be the starting implement due to it's weight.

The next chapter of the project begins today. The electronics section of the project will be executed mostly in my lab (AKA a small room of my house I clutter up with my junk) . I've dug out the hardware that I was preparing for the previous iterations of the CNC. It was like digging for fossils. I had to blow the saw dust and metal shavings off of it with compressed air. As I've mentioned previously the brains of the setup is a Boarduino with an Atmega328. That's now running the latest version of GRBL. Version 0.6. Fortunately flashing GRBL to the Arduino was much easier this time. Used to be you had to compile it from scratch and then run a string of commands to get it to flash onto the processor. Now you just load a hex file using the Hex Uploader found here. It seems that if I was booted into UBUNTU at the time I was doing this that it might still be as hard as it used to with the AVRDude commands and what not. If your forced to do that and get hung up I have instructions on my old blog at www.stevetotheo.weebly.com.  Things may have changed a bit over the past couple of years but the idea should be the same.

Where am I now? So far I have repaired one of the broken Reprap drivers that ended up with a damaged connector early in the process during my last attempt at a CNC. I built four because I fully expected one to not work since it was the first time I had ever ventured into the world of surface mount soldering. It turns out not to be as hard as people make it out to be. All four turned out great and working. Back on topic. I have a working boarduino loaded with the latest version of GRBL, four working v2.3 RepRap stepper drivers, and a modified computer power supply unit to run it all. Things I need. I need to run back and grab one of my stepper motors to test each of the driver boards and get the wiring to a more permanent stage. Then I need to mount it all up, find a test Gcode file,  and hope that the motors are beefy enough to do the job. So that's the plan. Ive got 47 days to finish this up and at this point I am on schedule to meet the goal! Tomorrow its back to the daily grind of school and end of the semester Senior projects. 

A home of its own.

Well, I finally got the spare cash flow to order the rest of the power train components for the CNC. I picked up the couplers, shafts, and collars from Ahren at CNCrouterParts. That dude is fast. I emailed him a question and within the hour he had replied. I ordered the gear from him and it was at my house 2 days later. He is legit and a great person to do business with. I also got some astounding service from McMaster. I ordered the ACME rods from them and they were on my door step the next day. Crazy stuff compared to some of the other troubles I've had in the past from other companies.







I went over to my Dads shop to get to work on the power train and ended up building a rolling cart from random scrap pieces of building materials he had laying around the shop! As you can see I've found out you can post pictures beside text. So enjoy this new feature! Any way. There it is. Its held together with what I would call sewing needles (air-gun "nails") and glue. ;-) Its on wheels so its easily movable and once I cut some slots out in the faces of the cart it will be able to hold the electronics pretty well. Unfortunately, building this also means that I didn't even touch the CNC other than to pick it up and put it on the cart.  With classes coming to an end I'm going to have to hit it hard to make some free time to finish this up.  As I sit here and type this post Ive got 52 days 10 hours 40 minutes and 5 seconds...4 seconds...3 seconds... Its coming up on the CNC deadline.

That brings me to my next point. What is finished? Where is the line? I'd like to define that. The end goal is to have a working CNC machine....and something cut from said machine. I want to start the 2012 with a working tool that can be used to take a drawing and translate it into a physical piece you can hold in your hand. With the current hardware I think I'm going to be limited to circuit boards and 2 dimensional objects  due to some of the limitations of my driver setup and the gcode interpreter. We'll have to start there.

What's my hardware setup? Fair question. I think I may have alluded to it in the past. Its another DIY project I did in the past. Its troublesome...its unreliable, Its even ugly but I built it. Its a trio of reprap driver boards that I learned to make using solder paste, surface mount components, and an electric skillet. Yeah it worked great. Really. Anyway the thing is that there's a bunch of wires coming and going from each of the drivers to a boarduino (also a DIY project. First thing I ever soldered). It needs to be cleaned up and wire managed. Currently its all mounted in an old PC case. It ran one of the previous CNC machines I attempted. When I say "ran" I mean that very...very...very loosely. I was able to feed the boarduino a single gcode command through a terminal and it would execute it. Above that things would go wrong instantly. There were timing issues, wiring issues, and above all the screws were just that. They were threaded rods from lowes. The would bind and get all kinds of messed up. Ill try to pull a picture of the one thing I cut with it. Its just a couple of lines in a piece of MDF. I use it as the picture for my Dad when he calls my phone. He takes time out of his life to help me with these projects so I figure he deserves that.

Moving on past that learning experience. This time I have real ACME rods and legit hardware. Things such as helical couplers and thrust bearings.  Things that I couldn't afford back in the first two iterations of this project. Ill end the post with this.

Precision isn't cheap and no amount of engineering and creativity can be substituted for a well designed part. Sometimes you have to bite the bullet and buy the $20 dollar couplers or the $30 dollar carriages rather than trying to engineer your own.

Steve-O

Ardunio GPS on the CHEAP

I'm still waiting on money to get straight before I go off and purchase anything else for the CNC. I've got about 300-400 more to go on the table and then Im going to have to figure out the electronics I've already got. I'm sure they are going to need repair and Ill have to navigate my way through the software. That's all for another day.

I get antsy when I've got nothing going on in the project realm so I went ahead and spent 20 bucks on a Pharos 500 GPS. Its one of those stamp sized units that came with Microsoft's streets and maps software...I think. Anyway I've got a couple Arduino Mega 2560 boards and since the 500 natively outputs in a serial fashion it was easy to communicate with..

I say easy but it took me some tinkering. The mega has several serial communication pins and if you hook the GPS up to the main one, it trips up the Arduino on start up. So you have to (well I did, I dont guess you have to) unplug the GPS and then plug it back in after you get booted up. The solution to this is to move to one of the other serial pins and just edit the code. Ill post a follow up with the modified code and a picture of the modified receiver.

Modified receiver? Yeah! That's how its done. Everything Kuztoms! No for real I had to break it open and solder pins to the board to make it bread boardable<-- Spell check says Im making that word up].

All in all the project has been a success. It was relatively cheap and  filled that whole in my boredom. So check it off the list. Where can I go from here? I'm thinking a reverse geo-caching project but that will have to be another day.

Competition motivates me!

I wasn't planning on working on the machine today but I was inspired by one of the other members on CNCzone. KriegKuts is in the process of building a FLA-100 as well. Dude's Build Thread can be found at the link provided! The rail system I set up to drill the holes for the X rail needed to be modified for the 6 inch Y rail. Basically instead of the small bolts and wing nuts I had to use a C clamp. Not as elegant but it worked. Today I didnt have the help of my dad so having a plethra (<---Nice word huh?) of C clamps of all shapes and sizes to hold the rail as I drilled it. I skipped the mid size bit this time and went straight from the center bit to the 3/8 bit.

 

I was going to stop here for the day but then I realized that the Z axis only needed 6 holes to be mounted. I didnt trim the rail or the supporting 8020 extrusion so pretend you don't see that. As for going straight to the 3/8 bit.

 

 

You can see it turned out fine. Im sure it shortened the life of the bit but after today Im done with it and its paid for itself. I bought 2 and only needed 1. That means Lowes gets to have one of their nice $11 kick ass bits back Im am so amazed that Im able to work with raw steel and get precise enough results to even slide the rail into the extrusion. 

Without adjustment of anything I checked the height of the bottom of the Z rail on both sides. Less than 1/16th of an inch different. Thats nuts. Now Im off to check my bank account and see what I need to do to pick up the power components.

$310 plus shipping. Ouch. But that's all that stands between me and moving on to the electronics....which Ill be salvaging from my last build. Arduino + DIY Reprap drivers + some creative thinking. Thats the teaser of the day.

X Rails and Ganty assembly

Its days like this that make me proud to be a DIYer. Drilling the holes for the rails was much simpler than I had anticipated! I also noticed that CNCRouterParts just put out a jig to help people do the drilling. My dad owns a lower end drill press from Harbor freight. It works pretty well but you can tell it has a noticeable amount of play in the chuck so every hole has a small amount of variation in it.

In order to avoid compounding the variability I took extra precautions to ensure each hole was as close as possible. It worked out really well this time and I saved a good amount of cash by drilling the holes myself rather than buying the rails pre-drilled from FLA (not to mention the time I would have spent waiting for them to be shipped and get here )

For those of you willing to take the venture into making your own rails heres what I did.

Set up a guide to make sure your at least the same distance from the edge each time. I used a small piece of aluminum angle I picked up from Lowes  Use a center bit or a counter bore bit to start a small pilot  hole. Use some cutting fluid with every hole you drill. It makes things go so much better...like butter. lol not really but it did make a big difference. After you have each of the holes started, move up to a drill bit about half the size of the final 3/8 bit your going to use. Any thing will do, just make sure you use the cutting oil. With this bit you can either go all the way through or only partially. Both are good but we found that not going all the way through helped to hold the cutting fluid when you move up to the final 3/8 bit.

Here's a couple pictures of the setup I had going on the drill press. It worked really well.

Once all the holes are cut it all goes together like seen in the drawings. An interesting side note is that the assembly forum post that I mentioned

http://www.overclock.net/case-mod-work-logs/755828-complete-2-x-3-custom-cnc.html

Yeah that one. It shows him placing the carriage bolts on the 8020 and then tring to get the rail to go on it. That seems really difficult so I just put a bolt and nut in each of the holes and slid the rail into place. It only took a couple minutes. One of the legs was a wee-bit (sp?) close but in the end it all worked out.

That's all I had planned on doing today but I finished pretty early in the day so I got to throw the gantry together. That was much more complicated. The 3030 extrusion is extremely hard to drill through in a straight line. Even when using a jig made from another piece of 8020 I was typically not perpendicular to the piece.

Getting to this point never loses its appeal. Being able to push the gantry along that x-rail is the first pay off you get in a CNC build like this. This time I plan on cutting something. The original machine I tried to construct out of MDF never made it past the gantry push stage.

This machine is without a doubt the most sturdy machine I've attempted to construct so far. I feel confident that it will withstand the abuse were going to put it through.

Ok so whats next? Next I need to drill the holes for the Y rail. Mount that and the associated bearings. Drill the Z axis rail and mount that plate.. Then comes the hard part.... Finishing. Ive got some power train parts I need to order but that going to wait until I get the table assembled.

One step closer - Base Assembled

Here is a couple pictures of one of the more beautiful things I have made. Alright alright. I didn't really make that. I assembled it. A lot of work went into it to get it from the raw t-slot to its current state. Its incredible how many times you have to take that thing apart in order to fit everything and then you realize that something else needed to go on there so you have to take it all apart again

Extruded Aluminum CNC frame

Extruded Aluminum CNC frame

Next task is to drill the hole in the x-axis rails. That's going to be a long hard task to do in a steel rail.

Turning $350 in $670

Before I go into the aw inspiring way I cut $320 from the bottom line of this project, lets first take a moment to look at a fine piece of hardware purchased from FLA and explain why I have this big countdown running now. The bearing blocks are crazy cool in the way that they are adjustable. Much better than anything that I have attempted in the past.

Now on to this count down thing. As I mentioned in a previous post, I need a dead line. so here it is.



If that's not enough time then I deserve to fail.




Here sits $350 worth of aluminum.


And here is $670 worth of aluminum.






The difference is a five dollar saw blade from Lowe's and the ability to accurately cut the different lengths. The lengths needed to make the bed have been cut. Holes still need to be located in all of the pieces in order to assemble the bed.

I have ordered the flat steel bar stock for the X and Y rail. I'm not looking forward to drilling the multitude of holes in each of them but I will save over a hundred dollars by drilling them myself...If I can manage to get them accurately located.

Other news is that I found out I'm going to have to order some 8020 gussets that aren't included in the fastener bundle I picked up. Not a big deal but another expense to add to the list.