Getting Extruder to work

I'm building a clonedel, and after resolving all kinds of mechanical and firmware problems I'm now at the stage to get the extruder to work.
The clonedel includes a geared extruder, and my hot end is a PTFE barrel with a M8 brass rod with 3 mm bore and 0.5 mm nozzle. Heating is from a power resistor in a small brass block screwed onto the rod. I recently added PTFE sleeving the whole length.
My problem is that there is hardly any extrusion coming out of the nozzle. I once clamped the hot end into a vice to try and press filament through manually and I need to exert significant force to get tiny bits of extrusion at the other end.
At the moment I've got three theories:
- my hot end doesn't get hot enough, either from the heater or not enough heat is conducted into the melting chamber. When I had the heater in a vice I was able to pull the hot material back out of the heater, and it seemed to be quite soft.
- my extruder isn't generating enough force. Though it doesn't get stuck and it grinds a notch into the material, so I don't think it is the motor. It seems to have enough force to rewind material back out of the extruder when set to reverse
- I picked the wrong material to start with. I bought a roll of white abs from reprap kit a while ago. Does anyone have experience with that material? Do I need more heat or pressure?

You should be able to easily push through by hand.
ABS needs at least 230°C.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

> You should be able to easily push through by hand.
How much force roughly? More like tooth paste? Or more like cookie dough?

I do think that the heater reaches that temperature, though I had to try and calibrate the thermistor against a temperature-controlled soldering iron because I didn't find any sensible relationship between the datasheet and measured values. I tried it with higher temperature settings with the same result.
What is people's experience with the heater block/power resistor method? How significant is the loss of heat across the threading? Is it worth adding cpu thermal paste onto the threading?

Edited 1 time(s). Last edit at 06/15/2011 05:28AM by Jens_R.

Some possibilities:

1) Is the path that filament takes through the extruder fairly straight? If the filament is being bent significantly by the hobbed bolt, there will be too much friction for it to work properly. With the extruder unpowered, and the hot end removed, you should easily be able to run filament through it by turning the large gear.

2) Is the extruder's steps per mm calibrated correctly? If you're using Skeinforge 40 or later, you'll need to change the firmware to reflect changes to the way that Skeinforge tracks filament usage. Basically, the old steps per MM were based on MM of plastic coming out, while the new ones are based on MM of plastic going in (Should be about 36x different).

The path is quite straight, though the filament comes bent from the roll. It's quite a standard geared extruder, so I didn't really expect problems in the first place.
I tried the PTFE barrel and hot end in a vice, pushed the filament through by hand and I meet quite some resistance, so maybe it is not hot enough? Or there are additives in the filament that reduce its plasticity?

I have not calibrated the extruder steps, yet. I've got options for both <40 and >=40 versions in config.h, but the problem is that with both settings very little gets extruded (related to motor motion).

Where is your thermistor mounted? I had my thermocouple on my block for a bit until I realized the transfer across the threads to the barrel wasn't great. You really need to take your temp reading from as close to the tip as possible. I was seeing about a 20 degree difference. The other common problem is a gap that will be larger then your melt chamber between the ptfe and the barrel. The filament seems to expand larger then the melt chamber, cool down and then becomes very stubborn as it won't fit back into the melt chamber. If you push your filament in until it extrudes, let it cool down and pull it all apart you'll be able to see the large lump if that is your culprit.

The bore needs to be 3.5mm or greater. Filament tends to expand as it gets closer to the melting pot. If your bore size is 3mm, and you have filament that is already 3.1mm, it's not going to go far before seizing against the walls.

My hot end filament path is

Thermal Break: 40mm of PTFE@3.5mm Bore

Top of Melting Pot: 4.25mm of copper@3.5mm Bore

Bottom of Melting Pot: 4.25mm@3.0mm Bore

Nozzle Tip: 0.5mm of copper@0.5mm Bore

A good way to evaluate your filament path is to insert the filament into the path with no heat when it is clear (not been used, and full of melted plastic). The filament should travel the entire length easily with no binding. Next, heat up the filament using a hair dryer, and try again. There should be no difference in resistance. If there is, you should open up your bore slightly.

I built a mendal repstrap and without a doubt the most difficult part of this build was to getting the extruder to work. I was determined to build all components and in the end purchased the nozzle, PTFE tube and the long stainless barrel from Makerbot. I built a tray that holds the power resistors (I used the inexpesive square type) that the nozzle and tube are thread into. The key is there can be no where for the PTFE tube to expand into, including vertically (it has to be trapped by the nozzle at the bottom and the extruter frame at the top. I found that you really can't bench test this because any way you try to hold it will cause excessive heat loss. One more comment on the Makerbot nozzle. I didn't really think there would be much benifit in the teflon coating but it helps to keep the outside of the nozzle clean. Again unless you have access to a decent metal lathe you will find it difficult to match this purchased nozzle.
My opinion, I sure the die hard self fabricators will disagree.

I found that my extruder problems came down to 2 issues.

1) The hot end wasn't mounted firmly enough, and was increasing friction when it moved out of place.

2) I wasn't running it hot enough.

Given that you tried to calibrate the thermistor by hand, I'd expect that the second issue is what's causing your problem.

Look through the Temperature.h file, and try to see if you can find a table which matches your thermistor's part number, or manufacturer/resistance.

raldrich Wrote:
-------------------------------------------------------
> 2) I wasn't running it hot enough.
>
> Given that you tried to calibrate the thermistor
> by hand, I'd expect that the second issue is
> what's causing your problem.

This is a really strange thing. I've got a datasheet for the thermistor, but the temperature/resistance values are really a lot off from the measured values. Not a few degree different, but a completely different order. The datasheet said it should have 1.455 kOhm at 160 degree Celsius and 0.6396 kOhm at 200 degree Celsius, I'm measuring ~4 kOhm at 160 degree and ~2 kOhm at 200 degree. In the end I tabulated a lot of measurements in 5 degree steps and fit a new temperature curve over that. Unfortunately the thermistor itself is not labelled (or it is so small that I can't see), and I can't tell if there has been any mix-up with a different type. I guess the next step is buying from a different vendor to see the difference.

Progress: This is with a new heater with both the resistor and thermistor closer to the barrel. The thermistor is nearer to the bottom.
This is an older barrel with a ~0.8mm hole and an older ptfe barrier that I cut open to check the threading and fitting of the barrel. I still need to grab the filament with pliers, but it seems to extrude easier than before.

Edited 2 time(s). Last edit at 06/19/2011 03:40AM by Jens_R.

@Jensr By chance are you using black ABS from MakerGear? They had a bad batch about a month ago. I contacted them, after having problems similar to what you describe, and they confirmed the culprit to be "funky ABS"

jcabrer Wrote:
-------------------------------------------------------
> @Jensr By chance are you using black ABS from
> MakerGear? They had a bad batch about a month
> ago. I contacted them, after having problems
> similar to what you describe, and they confirmed
> the culprit to be "funky ABS"

No, it is white from reprap kit, and you can see a bit in the picture in the last post. I was a bit worried if some additive used for another application could cause this, but nobody mentioned any problems with this plastic so far.
The new heater seems to work better, and increasing the micro-stepping seems to sort out some mysterious problems probably caused by rounding. The next thing to try is improving the grip on the M8 bolt. I originally tried the cutting-disc-on-dremel method to cut notches into the threading, but I guess I'll try the original method with an M3 tap now. I'm a bit afraid of ruining it, though.

I'm getting first prints!
Key changes:
- I grudgingly paid the higher delivery cost at a different vendor and ordered the Honeywell mentioned on [reprap.org] .
- I bought a roll of PLA. That seems to extrude much better, so I'll make the first batch of upgrades and replacement parts from PLA
- I've got a stepper with higher torque on order to run the extruder with. I'll try ABS again once it arrives and I have parts for a new extruder.