I spent all weekend and nights after work fiddling with the printer. I added the limit switches. The Z axis switch mounts to the motor which is at the positive end of the travel. The X and Y are both on the negative ends (i.e. at 0,0). This is all configurable in software but you do have to be aware of what it means for your home position, the position the machine goes to to know where it is. With the Z on positive it means my home position is (0, 0, whatever I tell it maximum Z is!).
The Z axis switch is triggered by a small bent piece of metal attached to the top of the Z base, currently with masking tape but soon to be screwed in place. The X and Y switches bolt to the frame between one of the guide rails and the lead screw. They are pressed by small discs I machined up in the lathe. They have an off centre hole so the disc can be rotated and operate the switch like a cam. That gives me very fine adjustment of the switching position by rotating the little disc then tightening the screw.
For now all the wiring it temporary and taped in place. Cable management will need to be carefully though out so the wires aren’t caught in any of the moving parts, get in the way fo prints and they also need to be run so they cables aren’t strained. I have ordered connectors so I can make longer leads to the motors and sensors as well as some drag chain to help route it all nicely.
With the cabling in place I was able to test things out. I ended up changing the firmware I was running from Marlin to Repetier as I am also running Repetier as the PC host to talk to the printer. I set everything up with guessed values I figured out from a lot of reading online. The problem is not many people make 3D printers using screws on the X and Y axes. Most seems to be belt drives so the settings are not the same. Where lead screws are used it is usually for the Z axis which tend to be run much slower than the X and Y.
I cobbled together some values that worked, set my limits and tested the switches. They need some careful configuration to get right. I also played about with the microstepping on the motors, eventually settling on 1/16th microsteps. Any bigger and the machine was noisy and any smaller (1/32) sometimes the steppers missed steps.
With all that done I was able to control the machine to move!
So the next thing I did was try it out as a plotter. This takes a little fiddling with configuration but I got that going. You also need something to plot with so I drew up and laser cut this pen holder at work.
It has a sprung attachment that bolts to the Z axis that the pen rides on. This is so I can use a constant pressure on the pen. I used a piece of MDF as a flat plotting base and managed to draw this on it. The screws that hold it to the machine are effectively captive behind the sliding arm. The holes allow an allen key through so the thing can be screwed in place.
My little table made from an old stool base comes in handy for working on the machine. It rotates and you can adjust the height! It makes a very good laptop stand too.
I was able to plot a larger version of the test file I made for my little laser cutter.
With the plotter I was able to fine tune my speed and acceleration values (down) until I could accurately draw the logo multiple times over the exact same lines.
Since I knew all that worked I looked at attaching the plastic extruder. I am using what’s called a J head extruder. First I made the heated base mount because I needed to know the height of that to know what height I would need for the extruder.
The base is another piece of scrap MDF (cut at work now I have been trained am allowed to use the table saw). That attaches to the base plate with 8mm bolts that fit into the mounting holes. The top of the bolts were machined flat and drilled and tapped for 3mm screws.
The holes allow the heated bed to be screwed down to the bolts on the base plate. The heated bed sits on screws so you can adjust the level of it. On top of it goes a flat piece of glass held with binder clips. This is to provide a perfectly flat surface to print on. The surface can be leveled by adjusting the screws. The springs ensure the level stays set. They also allow the bed to move sideways slightly which is important as it expands and contracts slightly as it heats and cools. If you screwed the corners down hard, as the bed heats it would expand and bow.
I am also using a piece of cork tile under the bed as insulation.
The original idea was the bed could be bolted to the Y axis base with 8mm nuts under the plate (remember the cut outs in the frame rails to allow for this) but so far that hasn’t been necessary as there is enough weight to hold things still and the holes are drilled to a close enough tolerance to not allow any movement.
I will make a similar, larger base for plotting/laser cutting on but with a flat, sheet steel top. The steel means you can hold things onto it with magnets.
With the bed done I knew the height I needed for the extruder. The J head uses an aluminium body with a round head that has a slot in it for attaching to the machine. I expect you can buy brackets for these (or 3D print them) but I made my own.
A small piece of L angle has a slot cut into it to match the groove in the extruder. Another small plate had a hole drilled in it the same diameter as the groove. This was then cut in two. One plate is screwed to the angle bracket. The other has over sized holes allowing it to be pushed hard up against the extruder then hold it in place.
This works very well.
With that in place I was able to try the machine again but I hit a snag. The Z axis wasn’t working very well. The stepper was skipping steps!
It turns out the carriage was binding slightly on the guide rails and was too still for the motor to move. That turned out to be because of an ‘improvement” I had made to the machine. I originally used button head screws to attach the plate to the carriages. I then changed to countersunk screws so I counter sunk all the holes.
There is one small problem with doing this. The countersink forces the screw into one position, even if the hole is oversized to allow for some wiggle. Since this machine is hand made my drilling wasn’t precise enough and when countersunk screws were used it forced the carriages into a position that made them still on the rails.
The solution was to go back to the original screws. I just turned the plate over so the screws had a flat surface to sit on. That allows a bit of wiggle in the positioning before the screws are done up hard. This solved the problem. Note the taped on Z axis limit switch actuator!
Steed looks on.
With that fixed I was able to start squirting plastic. The first thing I tried printing was the 3DBenchy boat. This is meant to be a good test for 3D printers. The first couple of attempts didn’t go so well. It seemed the machine was going too fast so I lowered the feed rate each time.
At 50% speed things started going better. I got my first actual print! This took about 1h 45 minutes. Apparently this print should normally take about an hour so my printer is slow but then it was never built with speed in mind. The first print was rough but recognisable.
One of the parts of the process of 3D printing is called slicing. This is where you take your 3D model and run some software on it that that works out the commands to actually drive the printer, this is the slicer. It spits out G-code commands which is what the firmware running on the printer used to know how to control the machine. The Repetier software I am using comes with two different slicers, Slic3r and Cura.
I had been using Slic3r (this is what I used to do the plotting too) but I thought I should give Cura a try too. Both pieces of software have many settings to fiddle with but I am using defaults for my setup. In Cura this is a 0.2mm print using a 0.4mm nozzle.
My Cura print came out much nicer.
I was quite pleased with that! Not bad for an untuned home-hand made printer. That print I actually left running by itself while I went to bed. It seems the way with 3D printing that everything takes forever. I have noticed this at work too. It’s quite a slow process, even on fancy printers.
Today I was working from home (I needed quiet to concentrate ) but I couldn’t use the printer as all the software is on my laptop and I was using that for work. I really must get a dedicated machine built and running. I did fire up the rescued thrown out PC from work and that works so I will build that up soon.
This evening, after work, I did some more playing with the printer and settings. I tried speeding things up again and had problems. I also played with the filament temperature. Then it occurred to me I had never calibrated my extruder. You need to tell the software how many steps it must drive the feeder to extrude a given length of plastic. I hadn’t set that up. It was set to 50 steps per mm but after doing some tests I found it should have been 88 steps. I hadn’t been extruding enough plastic!
During the tests the machine printed that by itself.
With the extruder set correctly I printed another boat! The base was immediately better. I also changed the settings to print more of a border. I think this is just to mark around the outline of the print you are doing but it also gets the plastic flowing nicely before you start printing your object.
That boat came out better than the last!
The blue tape, by the way, is there to provide a nice surface for the first layers of the model to stick to. You don’t want your print moving while printing it! Apparently 3D printing enthusiasts spent a great deal of time experimenting with anything they could print on to come up with the best surface. Blue painters tape works very well. Kapton tape also works well apparently but it is much more expensive.
That’s one of the great things about 3D printing. Hobbyists have done so much work and experimentation into making all this work. It makes it much easier for me to come along later and read what they have done and make my own.
Still lots to do. Play with all the settings. Speeds, accelerations, temperatures, nozzles, layer heights, all kinds of things. Best to change one thing at a time though! I need to sort out the wiring and a housing. As well as cooling. I am currently using the print cooling fan to cool the electronics! I need a permanent fan blowing air over the drivers to keep them cool. I also need to look at mounting a small cooling fan on the extruder head to help cool the print although I am not sure how necessary that actually is yet.
My power supply is having regulation issues. The 12 volt rail wobbles about a bit. This is a known issue with PC supplies where the 12 volt regulation is not very good without a 5 volt load. I might add some 6 volt car laps as a load and to provide lighting around the printer. Hopefully this helps. I could also run a small 5 volt cooling fan as well.
I also need to print something other than boats so I started something tonight. Have to wait till morning to see what it is!