highest reasonable print speedsm

Hi all and thanks for previous answers that have helped me get my prints working.
My first machine, bought a Kossel kit that had been redesigned (poorly) for redesigned for 3 mm.
So..I'm getting PLA prints finally actually got some usuable parts.
I've optimized as much as possible: e3d hotend, close-fit ptfe bowden tube, Mk8 extruder gear.
But even running 210 degrees I cant seem to go above about 10 mm/sec, sometimes not even that, without the gear startind to occasionally chew on the filament.
So what IS an expected top speed in this situation? I thought deltas were supposed to be state of the art but I'm not even close to 50mm/sec, much less 100 as I sometimes see.
Also, how much better is the performance with ABS?
thanks

The speed limit on most deltas is extruder. Extruder cannot push filament to nozzle quickly enough.
By switching to 3mm filament, you made the situation worse. It requires more force to push 3mm filament compared to 1.75mm one.
Most reliable reports for maximum printing speed were up to 300mm/s. I saw a maximum printing speed of 400mm/s being reported too.
Even crappy 1.75mm filament extruder should do at least 100mm/s at 0.2 mm layer height, 0.5 mm nozzle.
Actually when considering printing speed you should know also what layer height and what nozzle was used because it is mostly about maximum plastic flow achievable by given extruder. The material itself is less important than how much it was heated.

I use a variation of the wades geared extruder on my Cherry Pi range. I print using a 0.4 nozzle at 210C with 1.75mm PLA and easily get 100mm/s. My extruder starts to give up at about 150mm/s. I'm pretty sure if I went to a 0.5mm nozzle I would get 200mm/s but I use Jheads and it's not easy to change. I also don't need that speed, the difference between 100mm/s and 200mm/s on a 1 hour print is only about ten minutes. I rather not stress the rest of the components.

Edited 1 time(s). Last edit at 12/09/2014 08:07AM by AndyCart.

Shadowphile,
I went through similar issues with it when I was commissioning my printer very recently, but I am not an expert with PLA.
I also used 3mm PLA when I started, since my heated bed was not ready. In the end I found my all-metal 0.5mm hotend was not suited to PLA and would jam.

There is an interesting article here on extrusion rates. I don't know the nozzle size you are using.

IMHO if the filament is grinding there are a few things to check:
- Crank the tension on the extruder as high as you can. Check the tooth prints on the filament with a magnifier.
- Try hand feeding the filament directly into the bowden tube by hand at different speeds, and with retracting... does it jam that way? (Mine did - that's how I knew what the basic issue was)
- Pull the filament out and check to see how much melted plastic has flowed up.
- Make sure the teeth of the drive gear are clear of filament grinds
- Test printing without retraction enabled.
- Max the cooling on the hot end thermal break
- My 3mm PLA was very very stiff, and would have to bend in various ways when the print started, and would carry that bend. Can you re-position the extruder to reduce bends?
- Have you lubricated the filament?
- Put some marks on the drive gear and on the filament with a sharpie so you can clearly see how the gear is moving, and how the filament is moving or slipping.
- Is the driver current set too low? The motor will not have enough torque - it will stop turning or slip back. You can just adjust the driver a bit higher.
- Is the driver current set high? The driver will cut out (thermal protection). You may hear a clunk sound when it re-energies. You can tell when this happens if the extruder pauses while printing - by putting some tension on the filament and see if it pulls out then. Just adjust it a bit lower - no meter required.
- Is the flow rate calibrated?
- Have you tried a sample of different filament?

I switched to ABS and all the problems went away - after I tightened the extruder a bit more and lowered the driver current. I've printed as fast as 200mm/sec on a spiral sliced vase at 0.18 mm. Note that I am using my direct drive high torque Nema 23 extruder.

A question for the community: is there a troubleshooting guide that goes through these common issues? The troubleshooting guides here do not.
I propose that the Troubleshooting link be on the main page, and perhaps a sticky at the top of the forum page.

---

My printer: Raptosaur - Large Format Delta - [www.paulwanamaker.wordpress.com]
Can you answer questions about Calibration, Printing issues, Mechanics? Write it up and improve the Wiki!

Hey Thanks Paul, thats a great troubleshooting list. It's nice to know I am not even close to my printer's limits. I know ABS is supposed to be easier but I would prefer to have access to working PLA, plus I dont have a good thermal setup for that yet.

First check your temps, that is where I would start.
The temp reading is not an actual number but a reference as they are highly inaccurate, what reads 210 on your printer, could be 120 on another. Disconnect your extruder, run the hot end up to temp and see if you can push filament through with your hand. If you know it's not clogged, and cannot get it through, up the temp. I had one printer that would read 360c when it was actually only 180c. Find where it goes through decently, that's your 200c area. If it's way off, so be it, you can recalibrate it later or simply use that number, it's only a point of reference. It's nice having an accurate number so you can relay to others what works for you, but it's not necessary.

Don't just crank down on the extruder (sorry, Paul)
It can crush and ovalize the filament, causing it to bind up in the bowden, leading you to further crank it down, and making it worse... It creates a vicious cycle. Keep in mind though that this is much more common 1.75mm filament than 3mm.

An Mk8 hobb is good for well over 200mm/s even on small nozzles, if it's tight and not pushing, something else is wrong.

Just a note...
E3d PLA performance depends a lot on the version and what you are printing, most using them specify exactly which model they have for this reason. All metal hot ends in general vary greatly in performance, reliability and how fickle they are with various materials, and even from person to person. We tested one all metal hot end (not E3d), in which the PLA would cool off and stick to the inside of the hot end, we then had to either heat it to 300c, or give the extruder a healthy shove to break it free the next time it printed, eventually, we figured out that the easiest fix was to retract the filament completely after each print.

So how fast can your Kossel go?
You have a number of factors against you, first, you went to 3mm, which is harder to push through a nozzle, worse, the E3d is not known as a high speed nozzle, it has a lot of inherent back pressure compared to some others. It's a good hot end, it's just not meant for speed, for speed you want a larger melting chamber, which is exactly opposite of the E3d design philosophy. There is also your 3mm filament, it's so stiff and heavy that it can cause your effector to tilt as it moves across the bed, this is particularly a problem as it starts swaying back and forth at speed and all metal hot ends tend to be HEAVY. Another issue is that the Mini Kossel frame itself is simply not that stiff and only gets worse as you pile on heavier moving assemblies. Regardless, you should still be able to break 100mm/s if you ignore quality and wear and tear.

Just to be clear, I bought this commercially as a kit instead of proven model because I wanted tinkerability. But it came with a number of (bad) mods and I'm stuck with 3mm.
That being said, it want to try the flexible materials which virtually require 3mm for Bowdens anyway. But I does seem like the original Kossel was a 1.75 machine.

This has all been very juicy info and I thank everbody!
..Stuck thermocouple down throat of heater without filament, within 10 degrees of 210
..why thermistors inaccurate? I was going to upgrade to a TC but read thermistors more accurate in their range.
..I've noticed that my 10mm test cubes are not equal dimensions on x and y axes. Could assume some kind of distortion on the frame/effector mechanics, I will look at the Bowden tube forces on the effector.
..I noticed in the past that after cleaning out the heatbreak I could push the filament through pretty but it would stiffen eventually. Since Ive concluded that almost all filament binding is due to molten plastic getting up past the thermal line in the heatbreak, then retractions can be bad. I can run at 60mm with no retractions or up to 1/2mm. At 3/4 mm my filament started jamming. In the past it was set to 1.5mm so that would explain a lot.

Edited 1 time(s). Last edit at 12/11/2014 08:02PM by shadowphile.

If molten plastic is getting too far up the nozzle, that suggests to me that your hot end cooling is not good enough. Do you have a good hot end cooling fan? Do you have guides to direct the airflow at the fins on the hot end?

Edited 1 time(s). Last edit at 12/12/2014 03:38AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

dc42, I am using the stock e3d design, which comes with an injection-molded fan mount. The closest cooling fin disk is not even warm to the touch.
I did measure the tube ID and it is 3.2mm. The filament is about 2.9, give or take. This seems kind of loose, but maybe it is supposed to be.
It seems like if the filament is retracted far enough it will pull the molten plastic up into the heat-break past the melt-line where it will cool. The heat-break ID finish is pretty good, especially after inserting a micro-bottle brush with toothpaste inside and spinning it with a hand-drill. No, it is not causing circular ledges, was very careful that not the result.
BTW, my hot-end is e3d Ver 6, for 3mm Bowden. I just which it was PTFE all the way to the brass nozzle, then probably no issue.
I CAN achieve 100 mm/sec on a small part with features although the results start getting sloppy, but it does not sieze up with 0.5mm retraction. I need more retraction though because I am still getting ooze-strings trailing across my print.
On a good note I accidentally bridged 0.2mm layers over an inch long last night with almost no snag. Only took about 3 layers before the bridge lines became perfectly flat. This is good because I am learning to HATE support structures.

Quote
shadowphile
I just which it was PTFE all the way to the brass nozzle, then probably no issue.
.

You can drill out the stainless bolt more if you wish allowing the ptfe tube go all the way to the heater block. This is what I have done with the custom all alloy hotend I designed which in theory means no binding in the hotend

Edited 1 time(s). Last edit at 12/12/2014 08:51PM by d3delta3d.

---

My delta build blogspot [d3delta3d.blogspot.com.au]

Custom Delta printer
300x500 build volume
magnetic effector
Smoothieboard controlled

Quote
sheepdog43
Don't just crank down on the extruder (sorry, Paul)
It can crush and ovalize the filament, causing it to bind up in the bowden, leading you to further crank it down, and making it worse... It creates a vicious cycle. Keep in mind though that this is much more common 1.75mm filament than 3mm.

No worries, rather - thanks for the correction and clarification. There's a happy medium of max bite but not crushing the filament. That's why he needs to inspect it.

Shadowphile,
Glad you were able to speed up a bit.
.3mm it a bit much clearance according to this. I wonder if this can that be corrected with a different PTFE liner? If so, where from?