Inconsistent extruder feed, jams - tried new hotend with no luck

I was having some problems with inconsistent feedrates and extruder jamming at 'high' feedrates (10mm/s) when manually extruding.

I replaced the hotend with a hexagon since I thought it was the culprit, same problem. Temp = 200C. It extrudes OK at 5mm/s. Any idea why this might be happening? It seems to flow okay for about 4cm and then it jams for a bit, flows ok for a bit and jams again. I assume it isn't the hotend this time since the hexagon should be pretty good?

I'm getting poor print quality, especially at low layer heights, I think the inconsistent feedrate may be the culprit. Any help would be greatly appreciated!

I'm using the mixshop extruder [mixshop.com] with a hexagon hotend, 1.75mm PLA

Edited 2 time(s). Last edit at 06/19/2015 05:48PM by geemoto.

The hot end can only melt plastic so fast. If you exceed its capability, the temperature will drop and you'll be trying to push almost solid filament out of the nozzle. What sort of power is the hot end heater rated for?

Quote
the_digital_dentist
The hot end can only melt plastic so fast. If you exceed its capability, the temperature will drop and you'll be trying to push almost solid filament out of the nozzle. What sort of power is the hot end heater rated for?

Its a 12V 30W ceramic heater. I assumed 10mm/sec was pretty slow? I guess it's possible the PID controller parameters are not correct (if it is a heating issue)

Is that the print speed or the filament feed speed?

if you change the hot end you definitely need to run the PID autotune before trying to print.

Quote
the_digital_dentist
Is that the print speed or the filament feed speed?

if you change the hot end you definitely need to run the PID autotune before trying to print.

Yep, I ran the PID autotune and input the new settings. That is the manual extrusion speed that I use when I'm testing. During the print it doesn't bind.

I noticed that when I extrude at different speeds, the amount that gets extruded is inconsistent. That's what led me to believe that this may be the cause of the poor print quality.

I assume that the stepper/stepper driver probably aren't at fault for inconsistent filament feed?

Edit: I just checked the temperature of the extruder as I extrude at 10mm/s, it drops ~5C at most, to 195, but nothing too drastic.

Edited 1 time(s). Last edit at 06/19/2015 07:15PM by geemoto.

I measured the filament extrusion length at different speeds,

at 1mm/s, 2mm/s and 3mm/s it seems like it extrudes 51mm instead of 50.
for 4mm/s it extrudes 49mm
for 5mm/s it extrudes 48mm
for 6mm/s it extrudes 47mm

I guess this means that my extruder gear [www.a2aprinter.ca] is slipping at higher feedrates? Is it possible the stepper motor/stepper driver are not functioning properly? I remember being able to manually extrude way more quickly in the past.

Check the motor current for the extruder. Maybe it's too low ...

If you're feeding 10mm/sec into the extruder and the extruder has a 0.4mm nozzle, you're driving the extruder at an equivalent print speed of 190 mm/sec.

I did a little more experimenting, it seems that the extrusion length changes significantly with or without the hotend attached. This should not be the case, right?

The stepper isn't actually 'skipping' at least not audibly...so I guess the gear is slipping on the filament?

What are you measuring exactly?

If you tell the firmware to extrude 10mm of filament, that means 10mm on the input side of the extruder.
You'll notice that the extruder drools even when it isn't extruding filament- that's because the filament in the hot end expands as it melts and pushes some liquid out the nozzle. If you're measuring the output at the nozzle it will likely include some drool which will vary depending of how quickly you measure.

Quote
the_digital_dentist
What are you measuring exactly?

If you tell the firmware to extrude 10mm of filament, that means 10mm on the input side of the extruder.
You'll notice that the extruder drools even when it isn't extruding filament- that's because the filament in the hot end expands as it melts and pushes some liquid out the nozzle. If you're measuring the output at the nozzle it will likely include some drool which will vary depending of how quickly you measure.

Measuring the input!

The other thing I'm contemplating is that the PLA absorbed humidity, swelled, and now isn't flowing as smoothly through the hotend =/ It's hard to say affirmatively what the problem is

Edited 1 time(s). Last edit at 06/20/2015 02:03PM by geemoto.

Problems with poor quality for thin layers may be due to back pressure on the extruder. That indicates you're running at the limit of the extruder's torque, which would also explain slip at higher extrusion speeds. That means either a bigger motor or higher current/driver are needed, or there is something wrong with the driver you're using.

Quote
the_digital_dentist
Problems with poor quality for thin layers may be due to back pressure on the extruder. That indicates you're running at the limit of the extruder's torque, which would also explain slip at higher extrusion speeds. That means either a bigger motor or higher current/driver are needed, or there is something wrong with the driver you're using.

Maybe I'll try swapping out the driver and motor and see if it helps. Thanks!

It's mostly strange because it was printing perfectly at one point.

I swapped the driver, still no help. I think it really is a problem with friction in the hotend/backpressure. It is strange because with this same nozzle I had no issues previously - do you think switching to a geared extruded may help? I could try to switch to a wade or greg's hinged extruder style.

If it used to work and now it doesn't, something changed. You have to figure out what has changed. You swapped the driver and it didn't fix the problem- did you swap the cable and the motor, too?. You changed the hot-end- did the old hot-end work? Is the nozzle partially clogged?

I doubt the nozzle is clogged since I just bought it two days ago. This hotend seems to work better, the old one would jam immediately if trying to print <0.15mm layer height, the hexagon won't jam at low layer heights, but the print quality isn't great.

I'll try swapping the motor today and see if it helps, that's a bigger job and I didn't think it was likely the motor was the problem, but I guess it's worth a shot.
Edit: I realized the motors I have aren't the same size - do you think it's likely the motor is the culprit? I'd have to order a new one to test it.

Edited 1 time(s). Last edit at 06/21/2015 10:30AM by geemoto.

It's more likely a problem with the cable to the motor than the motor itself. Extruder carriage cables flex a lot when the printer runs. It is easy for the cable to become intermittent due to excessive flexure, especially if you have a cable chain with a small radius of curvature on the extruder carriage. Your problem could be caused by an intermittent motor wire or heater wire. You won't be able to tell by looking at the wires- the metal conductor may have broken inside its insulation.

Quote
the_digital_dentist
It's more likely a problem with the cable to the motor than the motor itself. Extruder carriage cables flex a lot when the printer runs. It is easy for the cable to become intermittent due to excessive flexure, especially if you have a cable chain with a small radius of curvature on the extruder carriage. Your problem could be caused by an intermittent motor wire or heater wire. You won't be able to tell by looking at the wires- the metal conductor may have broken inside its insulation.

Gotchya, that makes sense. I may just take the opportunity upgrade the extruder to clough's itty bitty double extruder.

Why do you think the heater wire may be the culprit? The temperature seems to be controlled quite well

If the thermistor went intermittent, it would trigger a fault in the controller and you'd know about it.

If a motor wire went intermittent, it might cause some temporary slipping that would create the intermittent extrusion you mention. Also, an intermittent motor wire can kill the driver because it is the same as disconnecting the motor while the thing is powered up- a big no-no. I used to use a small bend-radius cable chain on my extruder carriage until a motor wire opened up. After that I just hung the cable down from above, and then finally switched to a flex ribbon cable. The picture shows the sort of location dependent defects that I was getting when the extruder motor wire in the cable drag chain opened up.



If a heater wire went intermittent, the temperature would vary and when it cools a bit it would be hard to extrude plastic- it doesn't take much of a temperature drop to thicken the plastic enough to cause extrusion problems. You mentioned seeing a 5C change in temperature- that doesn't sound good. PID control should keep the extruder within +/-1 or 2C, even at 200C. If the thermistor reports a 5C change, the change inside the nozzle is likely to be bigger. If you don't have an intermittent heater wire, the heater may be underpowered for the print speed you're attempting. A 30W heater can only melt plastic so fast...

Finally, plastic dust stuck to the filament drive gear in the extruder can cause periodic variations in the extruded line. Make sure the drive gear is clean.

A geared extruder can produce more downward force on the filament and I believe that's why there seem to be fewer reports of problems with extruder jamming with 3mm filament (all 3mm extruders have gears) than with 1.75mm filament (most are direct drive). Keep in mind that if you use gear reduction to multiply torque you limit maximum extrusion speed. You're trying to run at an equivalent print speed of 190mm/sec which is pretty fast, and are maybe running into the maximum speed your extruder/hot-end can handle. If you put an extruder with a 5:1 gear box on the extruder your top speed drops to about 20% of where it is now.

Edited 1 time(s). Last edit at 06/21/2015 11:34AM by the_digital_dentist.

My cables are in a spiral wrap coming down from the top, so there isn't too much tension on them. A lot of things to consider I guess. My prints aren't nearly as badly as the print you posted, it seems more like 1 or 2 layers are missing in extrusion.

I feel it has something to do with the motor/stepper driver/hobbed bolt. When I remove the hotend, it is quite consistent in #steps/mm, but when I put the hotend back on I require significantly more steps/mm. This is quite worrisome, indicating the pressure somehow affects the #steps/mm. I'm not sure if the load on the stepper is causing it to miss steps, or if the load on the hobbed bolt causes it to slip on the filament

Is it likely that using knockoff RAMPS/Arduinos could be causing problems?

Edited 3 time(s). Last edit at 06/21/2015 01:15PM by geemoto.

Here is an example of what the prints look like, you can see some layers seem like they are 'missing'

Quote
geemoto
My cables are in a spiral wrap coming down from the top, so there isn't too much tension on them. A lot of things to consider I guess. My prints aren't nearly as badly as the print you posted, it seems more like 1 or 2 layers are missing in extrusion.

I feel it has something to do with the motor/stepper driver/hobbed bolt. When I remove the hotend, it is quite consistent in #steps/mm, but when I put the hotend back on I require significantly more steps/mm. This is quite worrisome, indicating the pressure somehow affects the #steps/mm. I'm not sure if the load on the stepper is causing it to miss steps, or if the load on the hobbed bolt causes it to slip on the filament

Is it likely that using knockoff RAMPS/Arduinos could be causing problems?

Not that likely . . . more likely is that either the current is set wrong, or the extruder does not have enough pressure to bite the filament (this can happen if the bearing backing the drive fails as well). How warm does your extruder motor get after running for a while?

- Tim

Quote
tadawson

Not that likely . . . more likely is that either the current is set wrong, or the extruder does not have enough pressure to bite the filament (this can happen if the bearing backing the drive fails as well). How warm does your extruder motor get after running for a while?

- Tim

Warm, but nothing crazy. If I were to guess, I would say ~50C. I have a fan running on it all the time so it doesn't get too hot.

I'm not using a hobbed bolt exactly, using a setup from a local company [mixshop.com] the hobbed piece sits right on top of the stepper motor drive shaft.

I'm open to changing over my extruder setup completely, but at this point I'm not convinced it'll fix the problem. I though the hotend would...but that doesn't seem to by the case.

Edit: attached a better picture of the setup

Edited 1 time(s). Last edit at 06/21/2015 08:05PM by geemoto.

All I can say is when I'm really trying to print fast (200 mm/s) (no idea what that equates to with the hot end) with PLA I'll crank it to 220-240. I mean you're trying to spit the plastic out at almost twice the speed. It has microseconds in contact with the hot end. I'm looking at getting an all metal hot end so I can hit 260+ with ABS for fast speeds.

if it jams, then you are trying to push solid plastic through a tiny hole. Not possible, that simple.

Edited 1 time(s). Last edit at 06/21/2015 08:06PM by L4nce0.

Quote
L4nce0
All I can say is when I'm really trying to print fast (200 mm/s) (no idea what that equates to with the hot end) with PLA I'll crank it to 220-240. I mean you're trying to spit the plastic out at almost twice the speed. It has microseconds in contact with the hot end. I'm looking at getting an all metal hot end so I can hit 260+ with ABS for fast speeds.

Yep, I had considered that and tried to crank the temp up to ~240C, no real improvement. But I also don't think I'm going anywhere near that fast, my print speed is <80mm/s

That looks like a very tight bend in the filament - is it like that when you print? Can you straighten the path out and make sure the filament is always free to feed into the extruder during a print?

And, if the extruder stepper motor is getting hot with a fan on it, there's a chance that the driver is reaching thermal shutdown. Do you have a way of measuring the temperature of the driver? Have you checked the Vref on the driver to see what the current is set to?

Quote
JamesK
That looks like a very tight bend in the filament - is it like that when you print? Can you straighten the path out and make sure the filament is always free to feed into the extruder during a print?

Nope, it was just like that for the picture to get it out of the way.

Quote
JamesK

And, if the extruder stepper motor is getting hot with a fan on it, there's a chance that the driver is reaching thermal shutdown. Do you have a way of measuring the temperature of the driver? Have you checked the Vref on the driver to see what the current is set to?

I don't have any good/easy way to measure the temp, but its warm to the touch. The Vref is 1.02V

Edited 1 time(s). Last edit at 06/21/2015 09:19PM by geemoto.

Quote
geemoto
Nope, it was just like that for the picture to get it out of the way.

Ah, good - sorry for the dumb question!

Quote
geemoto
It's 1.02V

A4988 or DRV8825? That would be roughly 2.5A or 2A - definitely too much for an A4988 and possibly enough to shutdown the 8825 depending on how well cooled it is. Unless you are using a different stepper driver with a different sense resistor value?

It's a polulu (A4988), after you asked the question I did some reading up on it: [reprap.org]

It seems that the standard resistor is 0.05V but some of the made in china knockoffs are 0.1V (it is likely that the one I am using falls into this category). I dropped it down to 0.8V and it doesn't seem to be any better, I'll try to drop it further and see if its any help.

Edited 1 time(s). Last edit at 06/21/2015 09:36PM by geemoto.

Quote
JamesK
...and possibly enough to shutdown the 8825 depending on how well cooled it is. Unless you are using a different stepper driver with a different sense resistor value?

Is there a way to measure the sense resist value?

Edited 1 time(s). Last edit at 06/21/2015 09:35PM by geemoto.

The DRV8825 is an alternative stepper driver with higher current capacity and up to 32x microstepping. [www.pololu.com]

Measuring the sense resistor in circuit is likely to be difficult. In fact, so far I can't even work out which resistor is the sense resistor. Anyone got a schematic?

[edit]
Ah, found the schematic at Polulu [www.pololu.com]

So it looks like the pair of large green SMDs are the sense resistors (connected to chip pins 23 & 27). Unfortunately they're not marked, so unless yours are, it doesn't help any.

[www.pololu.com]

Edited 1 time(s). Last edit at 06/21/2015 09:47PM by JamesK.