Here is the finished piece.

HDPE 1/2 of Spindle Mount

Spindle mount test fit, IT FITS!

 Here is a video of one of my test patterns using some 1/4″ scrap wood.

I had to make several manual modifications to the gcode to get it right. Not because CAM software did it wrong, I just didn’t have all the settings right when I created the gcode. Also I don’t have any home switches installed yet so all the absolute values are wrong or at least that is what I think you call them. The machine coordinates I know are not right until I get the home switches installed. I am still learning about all of this and I have a long way to go.

I have ordered some upgrades that should make some nice improvements in cutting speed. I have 3 new Acme leadscrews and nuts coming along with some other enhancements, such as bearing blocks and a motor mount for the Z-axis. My tie wraps are working but are a little to gehtto for me:-)

It is a very nice package. It is a 3.5 Amp stepper driver with a parallel interface built in. It is a China unit but it seems to work very well. I also ordered a new Stepper motor for my Z axis. Anyway here is a video of the CNC cutting the Roadrunner Gcode in Mach 3, into Styrofoam. Skip to about 2:08 to see a wider view of the system and the touch screen monitor.

I don’t blame you for skipping through it, kind of dry video but I was excited to see it work! The bit I had in the router was too big so some of the lines ran together but all in all it did pretty well.

More to come.

I was able to get all 3 axises put together and aligned pretty well. The only real problem is that my Z axis stepper is not going to have enough torque and it has too much run out in the stepper motor. but for testing it is fine.

First I wanted to explain how I am coupling my motors to the lead screw, AKA all thread

I took a vacuum TEE that I had and cut the barb off of one end.

Amputated Tee

This was too small to just slide onto the shaft of the stepper so I had to press it on using a C-clamp. This makes for a very tight fit. Then I score the barbs parallel to the shaft to give it teeth that can bite into the rubber tubing to keep it from slipping. I then press a short piece of tubing onto the barb and then onto the leadscrew. If needed you can add a tie wrap to secure the tubing even more. Here is a picture of how I pressed it on the stepper.

Pressing on barb

Here is a video of all 3 Axises running at the same time. I am using my arduino to run this test because I fried my buffer chips in by breakout board! That is fine, I will build something better anyway.

Here is a picture of my Z axis flange nut I built. It is hand filed to a press fit so that I can adjust it to align.

Z Axis Flange Nut

I have not put any anti backlash devices on these yet, but I have a design in mind for some.

OK, the first thing to do was tear the entire unit down. I decided that I had the uprights too high for the sides so I need to cut them down. Then it was time to work on that bearing. Let me stop here and say that I am very sorry that I don’t have pictures of all the work. My good camera was charging so I was using my backup camera. I took a lot of photos that didn’t come out due to overexposure and out of focus. Anyway, the shims I bought came in a stack of about 8 or so. They were all stuck together from where they were cut out. It was a challenge to get them apart without tearing. When I finally got them apart I started adding them under all the slides. After doing that I tried to slide the bearing back on the rail but it was obviously too tight. So now it was a game of trial and error until I got enough shims to tighten it up but hopefully, not so much that it will cause too much drag for my stepper.

Next I started working on reassembling the uprights and the X Axis rail. When that was done I had to attach them to the base. Until now I have just had the bottom rails setting on the base. The base is a piece of 1/2″ scrap aluminum plate. It wasn’t exactly cut square and I realy didn’t have a good way to cut anything that thick. So I had to square the rails without using the base as a guide. So if you notice the 1″ rails setting on the base don’t match exactly with the edge of the base, that is why. I will grind the base off square later.

The next order of business was to install the slide rails for the Y Axis. I first cut two pieces of the aluminum channel to length. Alignment here is critical. Not only do they need to be perfectly parallel, they must be exactly perpendicular to the X Axis. So in order to insure this I mounted a square to the X Axis and attached a pencil to the end and drew a line on the base. So now I had a reference on my base to the X Axis. Then I used a square to strike two lines perpendicular to the X Axis line. I used these lines then to line up my channel support. I then used C-clamps to hold them in place while I drilled the mounting holes. The positioning of these holes was also critical because my rails are pre drilled and tapped and these holes had to match up with the rails. After the holes were drilled I had to counter bore them from the bottom so my 6mm allen heads would set flush with the bottom of the base. Finally, I pushed the bolts up from the bottom and set the rails on them and tightened everything up.

Here is what the channel looks like.

Here is one completed rail

Table setting on slide bearings.

 After the rails were done I had a decision to make. Would I use 4 bearings, 2 per side, or just 2 bearings, 1 on each side. My Y table is only about 12″ wide so I thought one bearing in the middle should be fine, and a lot less work to line up that the 4 bearings would be. So after a lot of measuring and math I had transferred the mounting patterns on to my Y table and drilled them out. Once I got the bearings mounted and the table installed, it would slide very well. However if you applied almost any force at all to either of the unsupported ends of the table, the bearings would thrust and bind. This obviously was not going to work. There was just too much play in the bearings for this type of set up. Also you could almost fill the movement of the bearings as the table slid, even when the pressure was applied directly over the bearings. I was about to pull the table and re-drill it for all four bearings, when I remembered that I had some 20mm slide bearings that another friend, also named Brian, had given me. I hadn’t thought to use them because I didn’t have a housing for them. My thought was, I wonder if they will fit in the same housing as the bearings I got from Glacern. The Glacern bearings were held captive by two snap rings at either end of the bearing housing. The slide bearings I had were a little longer so I Knew I would be able to use both snap rings. Luckily there is also an allen adjusting screw in the middle of the bearing that tightens down to adjust the tightness of the bearing. I simply installed the slide bearing and tightened the allen screw tight enough to leave a mark on the bearing. Then I removed the bearing and drilled a divot in that spot. Now I reinstalled the bearing and tightened the set screw into the divot and the bearing was now captive in the housing. After installing two of these to the table and sliding the table back on to the rails, I was amazed how much smoother the table moved and the bearings did not bind when a reasonable amount of force was applied to the outside edge of the table!

Over exposed image of the bearings

Sorry about the quality of this photo, but you can see the original bearing on the left, the removed ball bearing in the middle and the slide bearing on the right.

Next I had to attach a nut under the table for the screw to act on to move the table. I had a piece of scrap aluminum about 3/8″ thick that were cut into two rectangles 1.5″ x 2″. In the center I tapped them 3/8×16 for my drive screw. I am using just standard threaded rod for now.  I then turned one of them on edge and drilled and tapped two holes for 6mm. I drilled the table to accept the two holes I just drilled in the nutand mounted it under the table. By the way, every thing you see here was drilled with a hand drill not a drill press. Originally took some pieces to a shop to use the drill press but found the accuracy of the cheap drill press was no better than I could do by hand so I decided to just do them by hand. Don’t get me wrong, if I had a good drill press, allot of this would have been much easier. Anyway to get back on track, I took a threaded rod and screwed it through the nut until it came out the back side of the table. Now I used a line that I had already drawn on the base, that was the center of the rails, to align the nut so that I knew it was straight.

Y Axis Nut

You can actually see it better in the reflection off the base.

Then I built the mount for the stepper. This was built with some aluminum angle.

Y Axis Stepper mount

 After this was done it was time to reassemble the Z Axis. The original setup I had used for the Z Axis was not working out. The stepper motor didn’t have enough torque and it was very hard to get perfectly aligned. After a lot of thought, I rebuilt it using a different stepper, that I had from a tear down of an old dot matrix printer. I also inverted the mounting of the motor and the drive nut. The motor is now on top of the axis, rather than tucked away inside. Also for the nut I used the second piece of aluminum that I spoke of earlier. I used a grinder to get it to rough dimension then hand filed it so that I could press fit it into the aluminum mounting for the axis. I wish the pictures of this had turned out. It is hard to explain without the pictures, but I will get some in a day or so. Anyway, all I like on the Z Axis is finishing the stepper mount on the top.

Here is a picture of the unit as it stands right now.

Front view, Z Axis mount not complete

That’s all for this update, time to get some rest.

As for why I am using TurboCNC instead of Mach3, quite simply my laptop won’t run Mach3 but it will run TurboCNC just fine. Newer laptops have power conservation circuits that will interfere with the timing so don’t expect your really Rad gaming laptop to work for CNC purposes. Especially with Mach3, again some of the older software is a little more forgiving if you can sacrifice the features.

OK, enough rambling here is the video of the test, nothing exciting just a simple demonstration.

Here is a video on how I configured TurboCNC to do the quick test.

At the end of the day a few things are starting to take shape.

The X Axis with the Z Axis Front View

Look closely at the shinny Z axis plate, see that red swish? That is a little blood sacrifice I made to the project gods. My wife (the hand model because she wouldn’t let me get her in the picture) is holding the X Axis up I don’t have it secured to the table yet. Everything is aluminum except the fasteners and rails, they are stainless. The table is a whopping 1/2″ thick plate. the smaller plate for the Z and Y axis is 1/4″ plate. Yes, the Y axis table is too narrow I will address that later.

Rear view

Here is the view from behind. If this stepper works out it will be a slick install, notice you can’t even see it in this picture. The next one will show it tucked away.

I can bearly get a peek at it.

One of the biggest pains today was making the screws for mounting this. In case anyone wants to mount some of these steppers from SparkFun, the mounting holes on the front are M 3.5. I couldn’t find any screws long enough for what I needed so I used some threaded rod. Cutting them and grinding the ends so they would accept the nuts was a real joy!

Not too bad in the end though.

 I made all of my own fasteners and I will show some close ups later on how they worked. Overall I am happy with them but they are a lot of work compared to just buying some from 80/20. The drilling must be precise and there are some tricking things to do. Such as counter bore the rail face so the socket head screws can set in flush.

So, back to the topic. I had decided to use a linear bearing from 80/20 for my X axis. I could have bought one ready to go for about $70., but I decided to buy the bare pieces and build it myself for about $20. I bought the bare bearing in an unfiled and deburred state for about $5. Then instead of buying the bearing pads individually, I bought an 18″ length of the bearing material UHMW-PE (Ultra High Molecular Weight Polyethylene) for $9.54. I then bought a bag of the screws to attach it to the bearing for $3.70.  I bought 1 extra piece that I new I wouldn’t use but I wanted to use it as a reference, that was a completed UHMW-PE bearing pad.

Here is a picture of the components:

Rough bearing lower right, UHMW-PE left, reference pad upper, Screws upper right

I originally was going to cut the  bearing material into pads just like 80/20 does, that’s why I bought the finished pad. However I decided that it would be better to make longer pads that covered the entire length of the bearing in one piece. This not only will make the bearing stronger and have less slop, it turns out it is much easier to make them this way. Now someone may jump in and say that you will create too much friction if you have that much contact area. I was worried about that myself but the finished product slides smooth as silk.

The first step was to deburr the bearing housing. I did this with a hand drill and a bit about twice the size of the holes. Since the housing is aluminum, you need to be careful. I ran the drill in reverse using the bit to deburr the holes with very little force. Then I finished deburring the edges with a file. Finally I used emery cloth to smooth all the surfaces. There is no need to spend time sanding or polishing further because the housing doesn’t come in contact with the rails.

Next I cut four lengths of the UHMW-PE bearing material, exactly the length of the bearing. I used a coping saw for this and it worked really well. One tip, I laid the bearing material in the slot of a piece of 80/20 rail then laid the bearing on top. I then scored it with the saw. Then I moved the bearing and slid the bearing material down to the end of the rail, leaving it in the groove but hanging over the end so I could saw it but it was supported and wouldn’t move around. I then used a razor knife to clean up any rough edges from the saw cut.

Cut piece of bearing material in slot of 80/20 extruded aluminum

Then I put the 4 pieces of bearing material into the slots of a 10 series 2″ wide extruded aluminum. Next I slid the bearing housing onto the bearing pads. After lining everything up I took a marker and marked all of the holes that the screws were going in with a fine tip permanent marker. This left a perfect center showing for lining up the bit on the drill. Then I took a small bit, I believe about 11/64″ and drilled out the holes where each screw was to be placed.

Here is what it looks like after drilling.

On the first couple I used a bit stop so as not to drill completely through the bearing, then I realized it didn’t matter if I went all the way through because that surface of the bearing doesn’t touch anything, so I just drilled the rest all the way through. Then used the screws to attach the bearings, making sure not to over tighten them and strip out the hole. Here is a picture after I installed the first pad then pulled it off to inspect it.

First bearing pad installed.

I continued on with the other three pads in the same manner. That was pretty much it! It was really simple and I saved myself $50.

Sorry I don’t have a picture of the competed bearing but this is where my camera battery died. After this I actually went on to drill out my Z axis mechanism to accept a 3/8 ” rod for movement then attached the it to the bearing. It is starting to come together. I will get some pictures up hopefully this week if I am not too swamped at work, it is month end so work will be a bear!

More next time…

Oh yeah forgot to mention, my optically isolated parallel breakout board came in from CNC4PC. If I get time this week I may do a quick write up on it.

I have decided to put aside the plans to run this from the Arduino for now and instead I will use a parallel port on a spare PC. I am still using the Arduino for testing for now. The reason for this decision was simply that the use of the Arduino was going to severely limit my integration of some very good off the shelf software. I suspect I will be revisiting the integration of the Arduino later.

Today’s goal was to construct a proto board with the 3 EasyDriver’s on them and the breakout terminals to simplify hookup. I purchased a proto board from Radio Shack and intended to purchase the terminals as well, but much to my disappointment they didn’t have them. So I had to sacrifice a breakout board I built for my OOPIC several years ago. I had  to amputate the terminals I needed! Oh, the carnage! Well at least it had signed the organ donor card so this project my live.

Here are some pictures of the construction:

This is the layout of the stepper drivers on the Proto Board

I had to first solder pins to the other 2 driver boards that I hadn’t used yet. Then I started with the wiring on the under side of the board. Here is a picture shortly after starting the wiring.

Let the finger burning begin!

Man, I will be glad when I have my own CNC router that I can just put a copper clad board in and zap out my custom PCB.

A few minutes later I have this:

Aint it Purdy?!

OK, well maybe it took more than a few minutes and several puncture wounds and a few burns as well. You will notice every like trace is color coded.

Here is the Key:

Light Red-> Coil A

Dark Blue-> Coil B

Yellow-> MS1

Dark Green-> MS2

Purple-> Dir

Brown-> Step

Light Green-> Sleep

Light Blue-> Enable

Black-> Ground (There are two gnd sources)

Red-> V+ (12v)

If you are asking yourself, why is he giving so much information here about the color codes and such, remember this blog is as much for me as it is for anyone else. I can use this as a reference should I need it.

Here is the top view again with the wiring completed and the standoffs installed.

Board with Z axis wired up to Arduino

 Here it is labeled:

 That is where I finished today. I ordered some 80/20 extruded aluminum Friday. I hope to get to work assembling the frame and axises next week.

Videos at the bottom.

Ok, a lot has happen this week. I started out with a general ideal of how I wanted to accomplish the X and Y axis movements. I was going to go with the sliding table design rather than the Gantry. No real good reason why I have this method in mind it just seems simpler to me. Anyway I go scrounging around through some surplus junk I have picked up from an industrial auction and I came across a couple of pneumatic cylinders with a linear motion mount. This caught my eye right away. The design is quite nice but it only gives me about 10.5″ of travel. This should be fine for my first build and it should be plenty for routing printed circuit boards. Here is a picture of what one of these mounts looks like.

Linear Motion assembly

Today the Stepper motors and EasyDriver boards came in from SparkFun. This is the new version 4.2 driver boards. There isn’t much out there in the ways of tutorials for using this board yet so I hope I can help someone. 

The first order of business was to get some mounting pins soldered onto the driver board so I could plug it into a breadboard for testing. Here is a picture of the module with the pins soldered. 

EasyDriver v4.2 with Pins soldered on

Hint: it’s easier to solder the pins in if you lay the circuit board on the breadboard and stick the pins through the holes into the breadboard then when you have all the pins inserted solder them while its still on the breadboard to hold them straight)

Next I started laying out the jumpers and terminals on the breadboard. I know I could just stick the wires in the breadboard, but I find using these screw terminals work much better. The terminals I have are something like 5 mm pitch so they only pickup every other rail on the breadboard. So it takes a little thought on how to lay the jumpers out.

Jumpers

Terminals

Here is what it looks like when I finished placing all jumpers and terminals.

Completed test board

Wired up

Wired test board

In case you are trying to follow my wiring here, notice the color coding on the stepper motors. On the circuit board there are 4 terminals to hook to the stepper motor. Coil A and Coil B. Coil A terminals are next to each other and Coil B terminals are as well. However due to my use of the 5mm blocks I had to stagger the connections so what you see here  is ABAB, so Blue and Yellow are A coil and Red and Green are B coil.

The wiring of these units is pretty much straight forward. You have a direction Pin, that is held high for one direction and low for the other. A step pin that is pulsed high then low for 1 step. You have 2 other pins that enable micro stepping, ms1 and ms2. Tie them both low if you don’t want to micro step. ( I recommend tying them HIGH until you have tested for successful operation). You have two sources of power, one for the motor and one for the logic. If you are using an external power supply for your motor, which is almost always required, then the board will supply power back to the Arduino through the 5v logic connection. (However I had some issues getting my sketch to upload properly without my normal external supply hooked to the arduino). You have an enable pin that must be held low for the board to operate. The last connection that I used is the Sleep pin. If you tie it low then your motor will not be powered even though the arduino and the easydriver boards are still supplied. This also allows the motor to free wheel. I haven’t used the RST and PFD pins so I can’t speak to them just yet.

Here is the code to get the test running

NOTICE SEVERAL ERRORS WERE CORRECTED SINCE MY ORIGINAL POST!

// Some of this code was scavenged from different places on the Internet but I have changed a few things to get it working.

int dirpin = 8;
int steppin = 9;
int ms2pin = 10;
int enablepin = 11;
int ms1pin = 12;
int sleeppin = 13;
void setup() {
Serial.begin(9600);

pinMode(dirpin, OUTPUT);
pinMode(steppin, OUTPUT);
pinMode(ms2pin, OUTPUT);
pinMode(enablepin, OUTPUT);
pinMode(ms1pin, OUTPUT);
pinMode(sleeppin, OUTPUT);

digitalWrite(enablepin, LOW);    // enable easy driver
digitalWrite(sleeppin, HIGH);    // easy driver operating
digitalWrite(ms1pin, HIGH);    // microstepping disabled
digitalWrite(ms2pin, HIGH);    // microstepping disabled
}
void loop()
{

int i;

digitalWrite(dirpin, LOW);    // Set the direction pin to move forward

delay(1000);                  //Give it some time

Serial.println(“Forward”);    // Print Forward to the terminal window
for (i = 0; i<31000; i++)    // Step Forward 31000 steps
{
digitalWrite(steppin, LOW);   // Start out with step pin low
delayMicroseconds(500);    // Delay controls speed and Torque I originally had these at 100 that was too short
digitalWrite(steppin, HIGH);    // Now switch it high
delayMicroseconds(500);    // Delay controls speed and Torque you need at least a 500 delay
}             //

digitalWrite(dirpin, HIGH);    // Change direction to reverse
delay(100);                     // Give it some time

Serial.println(“Reverse”);     // Print Reverse to terminal window
for (i = 0; i<31000; i++)    // Step in Reverse 31000 steps
{
digitalWrite(steppin, LOW);   //  Start out with step pin low
delayMicroseconds(500);    // Delay controls speed and Torque you need at least a 500 delay  
digitalWrite(steppin, HIGH);    // Now switch it high
delayMicroseconds(500);    // Delay controls speed and Torque you need at least a 500 delay
}            

}

OK, now time for a couple of videos. Note the wire ties holding the stepper motor on, I couldn’t find any m3.5 screws today so I just used the wire ties for now.

CNC 040 6.5 Meg

CNC 044 5.5 Meg

I am not impressed with the torque of these small steppers I will most likely need to size up but more testing is required. This was because I had my pulse width too narrow. After increasing the delays to 500, I have plenty of torque!

I was the happy recipient of some really cool parts today (Thanks Brian). I now have some 20mm rails and 2 20mm ball screws. These will be the DADDY project after I flush all of the software and driver issues out with this junior unit. (insert smile).

I guess that’s enough for today. More to come…