Poor or no extrusion

Problem

This could be due to a number of reasons:

  • Initial construction problems (if the nozzle has never successfully extruded)
  • The nozzle is partially or fully blocked
  • Extruder motor does not move much but makes a squeaking noise.
  • Extruder motor rotates, but the gears do not.
  • Extruder drive motor and gears rotate, but the filament does not feed.
  • The extruder gears squeak, rub and/or get stuck as the big gear turns.
  • The Bowden tube comes out of the brass unions

Testing: Hot end or extruder problem?

Hot end testing:

  1. Disconnect the Bowden tube from the extruder. If there is no filament in the hot end/Bowden tube, insert a piece. Check that it runs through the tube and into the nozzle smoothly. If it does not, see ‘Construction problems’ below. Check the filament diameter is within tolerance (1.75mm, +/-0.1mm).
  2. Heat up the hot end to operating temperature (200C for PLA), and push filament down the Bowden tube, by hand, to extrude. It should extrude filament with a small amount of pressure, and the extrusion should be smooth. If not, see ‘nozzle blockage’ below.
  3. Check the temperature of the aluminium cooling block – it should be cool to the touch, which the heatsink and fan are designed to do. If it gets hot, filament will melt further up the nozzle than it is designed to, and the pressure needed to extrude increases. See ‘Construction problems’ below. You may have to disassemble the hot end if there is a ‘plug’ of filament in the hot end.

Extruder testing:

  1. Disconnect the Bowden tube. With no filament in the extruder, and the extruder motor turned off, turn the large gear by hand. Does it stick at any point? If so, most likely the gears are a little tight.
  2. Load filament into the extruder. Turn the large gear by hand again. Does it turn smoothly and easily?
  3. Hold the filament tight, and turn the large gear. Does the filament slip? Check the hobbed insert is clear of filament in it’s teeth. Check the filament diameter is within tolerance (1.75mm, +/-0.1mm).
  4. Using Pronterface or the web interface, command the printer to extrude (you may have to send M302 to allow for cold extrusion). Does the motor move correctly?
  5. For all of the above, see ‘Extruder problems’ below.

Solutions

Construction problems

  • Hot end cooling: Check that the hot end fan is on ALL THE TIME. If the hot end fan turns off, heat can travel higher up the nozzle, and the force of extrusion increases, eventually stopping extrusion. The hot end fan MUST run all the time (it should be wired to the +12V directly), and there should be good contact of the heatsink to the heatsink block.
  • Hot end cooling: Make sure the heatsink is installed so that the fan can blow air through it!
  • Hot end construction: Check that the brass tapered nut is tight against the heater block on the nozzle. Tighten with spanners – more than finger tight! This will ensure the threads make good contact with the nozzle, and heat transfers well.
  • Hot end construction: If the ptfe nozzle liner is not cut square, or cut too short, and there is sufficient gap that fills with molten filament, again the force of extrusion increases. Cut a new piece of ptfe tube, 8mm long, with square ends.
  • Hot end construction: There should be space between the heater block and the cooling block; the heater block should not be up against the cooling block.
  • Bowden tube: If the Bowden tube is tight into the brass unions, and the filament has difficulty moving through the tube, this increases the force needed for extrusion. The tube should be 10mm into the brass unions, then run a 2mm drill into the ends to clear them. Push a piece of filament through to check it is smooth, and to clear out any debris.
  • Extruder: If the teeth of the hobbed insert have slipped on the filament, there may be pieces of plastic in the teeth, which the filament will slip on. Remove the filament, take out the big gear with the hobbed insert, then check and clean the teeth of the hobbed insert – a small wire brush is good for this.
  • Extruder: Check the idler bearing has a washer on it between the bearing and the motor, or the bearing will have difficulty turning

Nozzle blockage (also for changing filament)

To ensure the nozzle and melt zone are free from contamination, follow these steps:

  1. Heat nozzle to operating temperature (200C for PLA)
  2. Extrude a little filament, like 10mm (if possible) by hand or via the interface, then set temperature to 100C
  3. Wait for the temperature to drop to 100C, then reverse filament until it comes out of the extruder drive (about 380mm). You can do this at 600mm/min, or by hand if you wish.
  4. This should pull out the filament from the melt chamber, hopefully down to the nozzle, along with any contamination.
  5. Cut the contaminated end from the filament, and drive or feed the filament to just before the hot end.
  6. Set temperature to operating temperature
  7. Command the filament to extrude short lengths, 5mm at 200mm/min, until it squirts out of the nozzle.
  8. Repeat steps 1 to 7 as necessary, usually at least a couple of times. If there is not improvement after 5 times, you may need to disassemble and clean the hot end.
  9. If you have to disassemble the hot end, try to clear all of the filament out of the hot end before reassembly. You can heat the nozzle up with a lighter (hold it with pliers) to soften the filament, and pull it out. If the filament has broken off inside the nozzle, heat it up and gently remove it with a 2mm drill.

Extruder problems

If the extruder motor does not move as expected, but makes a squeaking noise or just vibrates, it may mean it does not have enough torque to drive the extruder feed mechanism, because it is stuck or jammed.

  1. Check that the nozzle is not blocked (see solution above)
  2. Check that the idler bearing can rotate freely (there should be an M3 washer between the bearing and the motor)
  3. Check the diameter of your filament is not too wide (over 2mm in diameter will not feed through the extruder)
  4. If the motor vibrates rather than turning, even with no load on it, the stepper driver chip may be damaged or the motor connections may be faulty.

Extruder gear rotates, gears do not

  1. It is unlikely the small gear will rotate on the motor shaft. If it does, contact RepRapPro support for a replacement.
  2. On the big gear, check that the hex head bolt is not rotating in the hex hole. If it is, again, you will need a replacement. As a temporary fix, you may be able to use epoxy glue or superglue to get the hex head to hold again.

Extruder drive motor and gears rotate, but the filament does not feed. There are a number of potential reasons for this:

  1. The teeth of the hobbed insert have plastic in them. This will cause the teeth to slip on the filament. Clean the teeth with a pointy tool.
  2. The nyloc nut on the back of the large gear has come loose, and the hobbed insert is unwinding
  3. The filament may be too thin, or it is trying to grip on a section where filament has been worn away. Remove filament and check diameter.

The extruder gears squeak, rub and/or get stuck as the big gear turns:

  1. There may be printing artefacts on the large and small gear, or they have been printed too ‘full’, so that they mesh very tightly. You can use sandpaper to improve the fit, or contact support to send you new gears.
  2. The large gear may be rubbing on the surface of the extruder body, or catching on the lower part of the small gear. Put a second M3 washer between the large gear and the first bearing, on the M3 hex bolt. This will lift the large gear away from the extruder body.

The Bowden tube comes out of the brass unions

  1. If the PTFE tube pushes out of the brass union, it is probably not screwed in far enough into the union. It should have about 10mm of thread. Remember to drill, with a 2mm drill, into the brass union with the PTFE in place, or there may be a tight spot that the filament can’t push past.

 

Filament doesn’t stick or parts warp

Problem

  • If the first layer does not adhere well enough to the heatbed, there is a chance the component(s) will warp during printing.

Solutions

Bed surface: Some people are lucky, and seem to be able to print directly onto the glass bed, and the PLA sticks. Most, it seems are not so lucky; for them we provide a roll of Kapton tape. Kapton can be applied to the glass surface in strips – try to keep the air bubbles out, and put the strips as close together as possible. Kapton is durable: we use it in the production of kits, and will last at least a couple of months of 24/7 printing. Usually it peels up before the PLA won’t stick to it. Blue painter’s tape can also be used. PLA doesn’t stick as strongly to it, and the surface isn’t as flat or durable as Kapton, but it is more widely available, and often in wider widths.

Cleanliness of build surface: The bed surface needs to be completely free of all oil and grease (including finger marks), otherwise your prints won’t stick to it. Set the heatbed to a temperature of 45C and wait for it to settle there. Clean the surface with nail polish remover (containing acetone, glycerine, and as few other ingredients as possible, and definitely ”’not”’ lanolin or any other oil or grease) using a lint free cloth. Set your heatbed to your print temperature ready for printing. Other products that also work include pure acetone, isopropyl alcohol, white methylated spirits, white vinegar. All of these dissolve oil and grease before evaporating. Don’t use Windex/Windowlene or polish; they often have a non-stick component!

Setting Z zero: At the Z ‘home’ position, where Z=0, the nozzle should be just touching the bed. Follow the instructions laid out in the Commissioning and Calibration instructions

Bed temperature: For PLA, try a setting of 50-60C. If you go too hot, the PLA will stay liquid and can be pulled away from the bed by the cooling of subsequent layers. Too cold, and it won’t stick. For ABS, the bed temperature needs to be much higher, at around 100C. See the ABS printing guide later.

 

Bed levelling and G32 issues

Problem

  • Prints stick to bed in some areas, but not in others
  • Bed equation fits points show a strange shaped bed, eg [35.0, 0.0, -0.136] [35.0, 180.0, 0.963] [210.0, 180.0, -0.231] [210.0, 0.0, 0.577]

Testing

Make sure the bed is as close to mechanically level as possible first; see the manual bed levelling section on the Calibration page.

The example G32 output is saying the bed is ‘saddle’ shaped, rather than flat, though this is more likely to be due to the movement of the axes, or incorrect measurement, rather than the bed NOT being flat. However, to check for a flat bed, put a flat edge (eg metal rule) across the bed diagonally, and see if there is light under the ruler, either in the centre, or the centre is holding the ruler off at the corners.

Check that your G31 setting in config.g set to something like ‘G31 Z2.3 P600’. The important thing is to have the P-value set to around 600; this is where the sensor is most accurate. G32 uses the P-value from config.g as it’s trigger value for bed levelling. If you have set it to something like ‘G31 Z0 P900’, you have misunderstood setting the Z homing, and both your Z homing and your bed levelling will be inaccurate.

Do a repeatability test; if you run G32 multiple times, do the readings stay consistent? This would prove if the sensor is working correctly, and there is something else effecting it. We have found that there is an offset difference across the whole bed between the first and second time you run G32 (which we’re looking into), but after that it is consistent.

Solutions

Generally, incorrect bed measurement is not due to the accuracy of the sensor; it is just triggering when it hits the value set by your G31 command in config.g. More likely, it’s a mechanical problem, and the sensor is not being given the same conditions sensing at each corner. This could be caused by:

  1. Wires pulling on the back of the probe changing the reading at the different points.
  2. Probing missing the target, or on the edge of a target, effecting the reading. Make sure the probe is measuring at least 5mm inside the edge of each target.
  3. Poor or sticking movement of Z axis (lubricate the linear bearing and z-threaded-rod; see https://reprappro.com/documentation/ormerod-2/maintenance/#Regular_maintenance
  4. Either, or both, of the X and Y axes may be twisted:
    • If the X axis is twisted, it changes the sensor height relative to the nozzle as it moves along the X axis. Look along the back of the x-axis-plate, and check it isn’t twisted, as shown in the last couple of steps here: https://reprappro.com/documentation/ormerod-2/hot-end-assembly/#Final_assembly_and_mounting
    • If the Y axis is twisted, the two Y smooth rods won’t be parallel, so the bed is rising and falling as it moves back and forth. Levelling this is usually a case of twisting the y-axis ends one way (or the other…), and tightening them in place once straight.
    • If both X and Y are twisted, you’ll see an error in the probing.
  5. Movement to the extreme of the X axis causes the hot end to lift slightly, effecting the reading. The shape of the x-axis-plate causes the hot end to lift slightly at the far end of travel. Reduce the X high probe position by 5mm.
  6. Are you using the new probe in ‘modulated’ mode? When you send G31, you should get a response such as ‘500 (550)’, ie two number with one in brackets. If you only get one, you’re using the probe in ‘unmodulated’ mode, so ambient light could effect the reading.

In the end, if you have mechanically levelled the bed to within a tight tolerance (+/-0.1mm), you may not need to use the G32 automatic bed compensation.

 

Hot end parts hit print

Problem

  • Fan/nozzle duct is closer to the bed than nozzle
  • Proximity sensor is closer to the bed than nozzle
  • Parts that are being printed are knocked off the bed

Solution

This can have a number of causes, in construction, and set up/adjustment. The bottom of the cooling nozzles and the proximity sensor should be about 1mm ABOVE the tip of the nozzle. To achieve this, a number of things need to be set correctly.

  1. When constructing the hot end, the brass Bowden tube union that screws into the aluminium cooling block needs to screw ALL THE WAY IN. If it is not, the nozzle will be lifted slightly, in relation to the sensor and the fan duct.
  2. The hot end need to sit on the x-carriage so it is vertical. If it leans forward, the fan duct will be lower. Change this angle by adjusting the bearing on the back of the x-carriage. If there is not enough adjustment, you can elongate the adjustment slot that the bearing sits in, or replace the bearing with a larger one (the standard is a 9mm bearing, replace with a 10mm 623 bearing if available), or put a ‘runner’ on the back of the x-axis plate for the bearing to run on – some people have used a hacksaw blade.
  3. If the hot end changes angle as it moves along the X axis, the x-axis-plate may be twisted, so the whole x axis arm has a twist along it’s length. Look along the arm, and rotate it until it is straight. Tighten the 8 screws in the x-motor-mount, and the screw in the x-idler to hold it in position.
  4. If the proximity sensor is too low, check that the MDF spacer is inserted between the nozzle-mount and aluminium cooling block. You can add an extra couple of washers if you need more clearance.
  5. It’s also possible that the y carriage bed is a long way out of level with the x-axis. You may need to adjust the corner screws of the heated bed to get it more level with the x-axis.

 

Axis sticking problems

Problem

  • Axis doesn’t move smoothly
  • Motor stalls when moving (sometimes okay at low speed, doesn’t move far enough at high speed)

Solution

  1. Make sure rods are clean and linear bearings run smoothly. A little light oil (like 3-in-1 oil) will help lubricate the bearing seals. Make sure you don’t drop oil on the bed, or prints will not stick! Smooth rods can be cleaned with wire wool or kitchen scourer to remove stubborn lumps.
  2. On the Y and Z axes, where two linear bearings are used, they may ‘cam’ against each other as the bearing clamps are tightened, and the axis may not feel smooth as you move it. Slacken the screws that hold the bearings and bearing mounts, and then re-tighten while moving the axis back and forth. This will let the bearings settle in the right place, and ensure the axis runs smoothly.
  3. On the Z axis, make sure the leadscrew is clean and well-lubricated. The motor may be stalling due to resistance to turning. Also check that the z-gear and z-driven-gear mesh cleanly; if they are tight, the motor may stall when trying to move.
  4. On the X and Y axes, check that belt alignment is correct, and the belt is not rubbing unduly on belt guides or anything else.
  5. Check stepper motor current is not set too low in config.g. As standard this is set to 800 milliamps by this line (sometimes Y is set to 1A):
    M906 X800 Y1000 Z800 E800 ; Motor currents (mA)
  6. Check there is no mechanical obstruction to the movement of the belt, or bearings on the smooth rods.
  7. Check that the linear bearings are not contaminated or damaged. Do the bearings run smoothly on the rods individually, when removed? If the bearings still does not run smoothly, there may be contamination in the bearing. You can clean these out quite easily; washing them in petrol is usually easiest, and then put a little lubrication (see http://reprap.org/wiki/Lubrication#Linear_bearings for recommendations) in the bearing before fitting.
  8. If the bearing appears to have a manufacturing fault, contact support for warranty replacement.

 

Wobbly Z walls and non-circular circles

Problem

  • Vertical walls are not accurately printed on top of each other
  • Variability in layer height causes vertical walls not to be smooth
  • Circular objects print out square

Solution

Generally we lump these problems together under the term backlash. This can happen on any of the axes, or a combination of them.

  1. Check belts are tight enough. On the longest free length, i.e. the side not attached to the carriage, pluck the belt. It should make a just audible, low pitch, twang. Tighten or loosen as needed.
  2. Check pulleys are not loose on stepper motor shafts (X and Y axis) – hold the motor shaft with pliers, then try moving the carriage, while looking at the pulley
  3. Check that the axes are moving freely: see ‘Axis sticking problems’ above
  4. Check extrusion is consistent: see ‘Poor or no extrusion’ above

 

Stepped layers

Problem

Partway through a print, the next layer appears to have slipped by a millimetre or two (or much more) causing a step which should not be there. This can be caused by:

  • Axis belt slipping where it is attached to the carriage.
  • Print head snags on part of the print, usually the print curling up or lifting off the bed. This can cause the belt to skip on the pulley, or the motor to stall.
  • Axis snags on something. For example, the wiring catching/getting in the way of movement. This can cause the belt to skip on the pulley, or the motor to stall.
  • Stepper driver overheats and temporarily shuts down.

Solution

Belt slipping in carriage

  1. This usually happens on the y-carriage. To test, hold the motor pulley tight and try and move the carriage. If it slips, secure it in place more positively. The quickest fix is to put cable ties around the two ends of the belt on either side of the attachment to the carriage, and tighten them.

Nozzle hitting printed part

  1. The printer should generally have the power to overcome hitting a part while printing, and the hot end can flip up a little. However, if printed parts are curling up, particularly on overhangs or bridging, reducing the extrusion temperature 5°C at a time will usually help.
  2. If the parts are curling up from the first layer, see ‘Filament doesn’t stick or parts warp’ section above.

Belt skipping on pulley

  1. Check belts are tight enough. The actual tension required comes with experience, but should be at least tight enough to produce a low frequency, just audible ‘twang’ on the longest section of belt. Over-tensioning the belts can also be detrimental, as the motors will have to work harder.
  2. Check that the belt is running smoothly and in line, and the edge of the belt is not snagging on the motor and idler ends. With the motors off, check the axis moves smoothly – if not, see the ‘Axis sticking problems’ section above.
  3. Check all wires, pulleys and belts whilst printing and reposition/realign anything impeding the smooth movement on all axes.

Stepper motor stalling

This is a result of the motor not having enough torque to move the axis (temporarily, since the print continues at the new position).

  1. Check that the motors are being supplied with sufficient current to meet the demand; check the setting in config.g
  2. Use a secondary cooling fan to cool the electronics if they are getting too hot (this should only happen when printing in very extreme environments).

 

Printing ABS

Problem

  • Concerns about printing ABS
  • Heated bed takes a long time to get up to 100C, or never reaches it
  • ABS doesn’t stick

Advice

Important: all the supplied printer parts are made from PLA. Long term exposure to the heat of ABS printing will cause some of them to fail. If you plan to print a lot of ABS, you should first reprint some of the parts from ABS, specifically (and in this order) the x-carriage, z-runner-mount, extruder-body and the rest of the extruder parts, nozzle-duct, fan-duct.

When printing ABS without upgraded parts, at the start of the print leave the x-axis high above the bed (at least 100mm) so that it is not getting hot while the bed heats up.

Heated bed

Check the voltage of the power supply, particularly under load. Ormerod PSUs should supply around 12V, but may be supplying a voltage below this. The PSU supplied with Ormerod 2 is adjustable, so you can increase the voltage a little; there is a voltage adjustment screw next to the screw terminals on the end of the PSU. This should allow the heated bed to heat up quicker and get to a higher temperature. However, don’t go beyond 13.5V, or the heated bed will draw too much current. The ATX PSU supplied with Ormerod 1 is not adjustable; some customers have replaced this PSU with ones that can supply more voltage.

The heated bed can max out at around 100C, due to the thermal mass of the aluminium and glass. This is a designed limit, and is generally okay for the ABS we have tested. You can increase this to around 110C by covering the bed with an insulator while it heats up. In the past, we’ve used a foil-fronted piece of MDF, which reflected heat back onto the bed, but was held off the surface by the clips, so didn’t heat up. Remove it to start printing; the first layer will then be a bit hotter, so should stick better if it’s being difficult, and the temperature will drop during printing to hold at around 100C. Another improvement suggested by a customer is to put aluminium/kitchen foil between the heatbed PCB and MDF insulator. This also decreased warm up time. Be VERY careful not to short the main power connections through the aluminium foil! ABS shouldn’t need heating to more than 110C anyway, as this is beyond it’s glass transition temperature; it’s like printing PLA onto a bed at 80 degrees – the PLA stays so soft it gets pulled off the bed. ABS generally does this above 110C.

Other considerations

Also, keep draughts to a minimum, and try to keep the area around the printer at a reasonable temperature – above 25C minimum. This should help to prevent the part warping as it prints. You can build a small ‘greenhouse’ to cover the printer, but be careful it doesn’t get so hot (over 45C) that the PLA parts of the printer melt!

Another source of draughts is the hot end fan. The current Ormerod fan-duct (with the nozzle pointing at the hot end) is primarily designed to help with bridging across voids with PLA filament; we have had good results by removing the fan-duct, but leaving the heatsink-duct in place, for ABS printing. Otherwise ABS can suffer from delamination (the layers splitting) during printing as it cools too quickly.

It should be noted there are plenty of other problems getting ABS to stick at any temperature, and there are quite a few workarounds; the favourite around here are super strength hold hairspray (it can contain both PVA and acrylic) or making a slurry of some ABS dissolved in acetone, applied to the bed at 50C, and wait for it to dry before printing. Joseph Prusa shows how he does it HERE (follow the pictures in the ‘Older’ direction). And then some people have no problem with ABS at all! It’s a bit of a dark art, but probably depends on the quality of your ABS filament.