Getting it right

Having put the Prusa kit together and calibrated it, I printed a small logo and it seemed OK. On looking closer it didn't look quite as neat as I expected. I decided I needed to check a bigger print. First I changed the spool of filament to orange and recalibrated. 

I think I'll describe the process of designing an object in another post, so I'll just say that I created a very simple square drink coaster. It is 100mm x 100mm x 3.5mm. When I tried to print it the print failed. 

 

The photo is not very good, but you can see that there are diagonal gaps and clearly the corner has come unstuck and curled up. The print head snagged on this raised corner and the whole object came unstuck. I watched this all happen and it was clear that the extruded filament was not properly adhering to the print bed. A couple of stray lines wafted about and where this overlaid the print this made a raised bump. As the nozzle passed this it snagged and started to move the object. 

It seems that there are two main reasons for failing to stick: the print bed might need cleaning or the print nozzle is not aligned on the bed. I cleaned the bed before the print, so that left the calibration issue. I re-read the instructions and some comments online. It seems that the the nozzle needs to be very close to the bed, but not touching. As the line of plastic is extruded it needs to be slightly squashed by the moving head. I had set the head higher than that. The First Line Calibration allows a fine control of the height of the nozzle to be made using the knob on the controller. I wound the head down by another 100 microns. I ran the print of the coaster again and this time it printer well and looked much neater. 

Keeping the print bed clean seems to be important, though the manual does say that it doesn't need to be cleaned before every print. I'm trying not to touch the centre area where the prints are made so I don't leave finger marks. The kit came with two cleaning wipes soaked in isopropanol. I need some more. I have some lint-free cloths so I thought I'd buy some liquid isopropanol. The only suggestion as to where to buy any is pharmacies. It seems that the national chains of Boots have decided not to stock it, not sure why. When I looked around the local villages almost all of the pharmacies have been sold to Boots, so no joy there. Eventually I did find an independent pharmacy that had a few wipes which will do for now. I ordered a bottle of the solvent online which should arrive shortly.

Testing and Calibration

The assembly stage of the Prusa I3 MK3 went pretty well. Before the power is switched on the PINDA probe height needs to be set. The manual describes it well enough. The next stage is to test everything works. I plugged the power in and switched it on. No burning smell, no smoke, no grinding noise so the first test was passed. The display lit up and offered a self test, so I pushed the button to accept.

A motor whirred and the head moved left and right and promptly reported an error. It seemed that part of my careful wire routing meant the filament sensor cable was snagging on part of the frame which stopped the full movement. I realigned the cable and tried again. This time the head moved left and right, the bed moved in and out and then the head went up and down followed by some more complex moves. The screen reported that all was well and that the X/Y axes are perpendicular which is the best result.

Calibration followed and I didn't do that very well. Part of the process is configuring a fine adjustment to the height of the nozzle. I was cautious and didn't adjust it low enough, so as it laid down filament it didn't stick to the print bed. I tried again with a better result, so I tried to print a simple design supplied on the SD card in the kit. It started to print and quickly failed as the block came unstuck.

I returned to calibration and watched very closely as I lowered the print nozzle to be sure it wasn't going to touch the print bed. The calibration print was immediately cleaner and straighter. The test print then worked without a problem. So I have my first printed plastic object.

I did put the failed first print to use by trying to paint it with acrylic paint. It took the paint well enough with good coverage but when it had dried I rubbed it fairly hard and some of the paint came off. I'll find some enamel paint and try again, just out of interest.

The final push

The printer now looks substantial, but also clearly incomplete. The basic structure is all there, but there's no control system and there are wires everywhere. So now it's time to ... add more wires.

The first step is to assemble the LCD display. This is a circuit board with a four line LCD display and a couple of controls that sit in a printed case. This is fixed to the frame with a couple of brackets and a pair of ribbon cables snake out to add to the nest.

Next we fix the metal plate that carries the print bed. This includes a heater and a temperature sensor, which are both preinstalled. The plate also has a number of powerful magnets that hold the steel print bed in place. The heater and sensor add to the wire count and still no destination for them all.

Last of these sources of wires is the power supply. This is, of course, completely prebuilt and sealed. It has chunky power cables that need to carry substantial current to drive the motors and heaters as well as a lower voltage supply for the electronics. It is fixed to the frame with four bolts.

So, speaking of electronics, we now add the place where all the wires will route to, the electronics assembly. This lives in a printed plastic box with a door that gives access. The circuit board is fixed into the box and the box secured to the frame.  Then the fun begins. The cables all have to be grouped and gathered together and fixed to carefully positioned points along the frame with cable ties. The description and photos in the manual are good, but some parts are still fiddly. There are two wrapped cable groups that have to be fitted into printed clamps and then the remaining loose cables are routed into the base of the box. All of the cables, except the power cables, are all terminated with plugs of various shapes and sizes and I didn't have much trouble connecting it all. The cables in the photo look a real mess - I could have made it tidier but then it would have looked like it belongs to someone else.

The last step in assembly is to push together the spool holder, clip it to the top bar and add a spool of filament.

Print head

The next section in the manual for assembling the Prusa i3 Mk3 printer is called the E Axis. The 'E' is the extruder, what I think of as the print head. It is very compact and made of many parts, some printed plastic, some electronic and some prebuilt. There's a motor that draws the filament into the extruder. There's the extruder, known as the hotend, which melts the filament and forces it through a nozzle. There are two fans to ensure the extruder doesn't overheat. There's a filament sensor to check the feed hasn't run out and there's a sensor, known as a P.I.N.D.A., which senses the distance to the print bed to ensure the nozzle doesn't hit it.

All of this lot have a lot of cables. They are all carefully threaded to a printed half tube which then get enveloped with a robust cable wrap. The whole compact package is fixed to the X axis assembly which uses the captured nuts that have precariously sat in their little pockets.

The photo shows how compact the head is, but gives no read idea of the complexity.

X then Z

Now the basic frame is complete the next stage is to build the X axis assembly. This moves the print head along two steel bars, driven by a motor using a toothed belt. The first stage leaves the belt off and was fairly simple. One fiddly bit is the way bolts (or are they screws?) fix into the plastic mouldings. The steel M3 screws (bolts?) would chew through any threads in the plastic, so Prusa have created little slots that intersect the holes for the bolts that a square, flat nut gets pushed into. It is a simple captured nut, but is allows the bolt thread to engage steel not plastic. The assembly that will eventually hold the print head has five of these captured nuts in it, but there's no bolts in them yet so as the fitting gets moved around it is easy for the not-yet-captured nuts to fall out, usually needing a search on the floor to find the escapee.

The second part of the job is the Z-axis assembly. The previously assembled Y and X axis units are moved by toothed belts each driven by a single stepper motor. The Z axis is driven by two threaded bars each with its own motor. I think this is to allow very fine control over slight unevenness in the print bed. The threaded shafts were already fitted to the motors and assembly was easy. Now the X axis assembly can be added, being careful not to lose those loose nuts, then the toothed belt was added. Tensioning these toothed belts seems a bit of a black art, but it seems that the calibration process at the end of the build should check that they are at the correct tension.

The section of the manual looked quite long for the Z axis assembly, but it turned out that there were alternative ways of working for older or newer parts so the relevant part was much shorter.

Now with the X and Z axis assemblies in place

The next section is the E axis. A 4th axis? We'll see.

Y-Axis is fiddly

I am assembling a Prusa i3 Mk3 3D printer kit. The assembly manual breaks the process into sections and the first one is Y Axis assembly. I pulled together the parts labelled for this section, the photo shows most of them, but there's some steel rods and a motor that's not shown.

The first part puts the basic frame together. It seemed fairly straightforward until the check that the frame sits level. I needed to put the frame on a flat surface and check that all four corners sit on the surface with no wobble. I had wobble. The manual suggests loosening the bolts and pressing down on the frame while retightening the bolts. That didn't help.

One thing that the manual suggests is that Prusa can help, so I looked at the comments section of the online manual. Lots of other people seem to have the same problem and always the front right and rear left corners wobble. This must be a design or manufacture issue, but there were suggestions to fix it. I swapped the positions of two extruded parts and loaded about 8 kilos of load to flatten the frame before tightening the bolts, being careful to do them up equally. That seemed to work and the wobble is now less than the thickness of a sheet of paper. Prusa say that's fine.

The rest of the build of the Y axis section was just as the manual described, so now I have a frame to build the rest of the printer on.

Before the build

I took a look at the assembly manual and it seems well laid out and easy to follow. There are LOTS of steps, this is going to take many hours to assemble, but it seems well documented. Repeated in various places is the message that you are not on your own with email, website and live chat and a forum all available. There is also a full, up-to-date manual on the Prusa website. The printed manual I have includes an update sheet that means I need to check the website version for sections 5 & 8 because the parts supplied have been changed since the manual was printed. I haven't read it cover to cover in detail yet, but I will read each section completely before I start any assembly. I'm really looking forward to the assembly in its own right.

I opened all of the boxes to see what they hold. It looks like a lot of parts. Most are in bags and each bag or box has a clearly printed label on it. Each label has a number on it to indicate which section of the assembly manual the contents get used in. I should be able to lay out the parts needed for each part of the build which should make finding each part easier, which helps when there's so many parts.

The plastic connector parts are all 3D printed, of course. I chose black parts, orange was an option. There is a bag of spares with one of each kind of bolt, nut and so on, just in case.  The first section of the manual is an introduction then the second is about assembling the Y-axis mechanism. It is 33 pages long, so I'm going to read it carefully and find all of the parts mentioned before I assemble anything.

Below are are the box contents spread out, but I intend to revisit the sections as I assemble them.

Aluminium framework, the heated bed and the print bed

Aluminium extrusions and plates

Steel rods and sleeves

Motors

3D-printed plastic parts

Power supply

Tools, fixings, belts, tie wraps and an LCD display 

Various premade parts, circuit boards, the extruder, fans and cables