3D Printing - "It's all about design and quality output. Not how fast the printer can run."
Let the filament spool hang off-center on the small spool axle or center rod.
One of the first things I made with my first 3D printer was filament spool center inserts. These adapt the large center hole to the threaded rod that was included with the acrylic spool holder stand supplied with the printer. It seemed to make sense at the time and a good first project.
I assumed everyone used them.
I have been suffering with changing and handling those adapters through about a dozen 1kg spools of filament and scores of changes between colors. But I have seen the light. The holes in the newer spools have increased in size rather than gotten smaller. There is a reason for that.
With a perfect centered spool, the printer can give a tug on the filament that can cause the nicely balanced and free to spin spool, a chance to unroll far too much filament. It is especially a problem near the end of a roll when there are a lot of tight loops in the filament and it springs out over the side around the axel rod.
The big hole in the spool is so the spool can intentionally hang off center and be heavy at the bottom. It causes more drag while spooling off the reel and it can never free-wheel. There is always a slight tension on the filament. It also prevents the chance of a rewind tangle caused by the attendant (you).
My free-wheeling days are over…
I have read reviews on 3D printer filament where the reviewer bashes the product because the filament on the spool is tangled “from the factory”. This is misleading as the tangle is certainly 99.9% the result of user error in handling and feeding the filament. If cross under tangles could be produced (they cannot) by continuous winding, then fishing spinning reels would never work. Neither would sewing thread, lifting cranes or any process using winding on a spool. There must be severe backlash during the winding such as occurs with open reel bait casting when the reel spins faster than the line feed and the line wraps backwards on the faster turning spool.. This “backlash” is not what is reported.
Manufacturers could never create backlash in production runs. The reel is pulling, not feeding.
The likely problem is the end of the filament could “get lost” by the user on the spool and the end was “fished out” from many loose or slack turns on the spool. It is very easy to create cross under with loose windings. The last 0.1% chance is the manufacturer doing this. But not many times deep within the windings.
This error is very difficult to observe and in fact the filament will feed normally for possibly many hours if the loops remain slack (as they do) on the spool. This time delay will disassociate the user with the cause. It appears the cross under was “buried” under layers in the spool. This is not possible in manufacturing.
Tip #2 is to pay very close attention if the end of the filament snaps back into windings on the spool during handling. Check very closely for a cross under tangle when retrieving the end. Always keep the free end of the filament secure to the edge of the spools in the holes or use a clip.
Let the nozzle cool down when not printing.
ABS and PLA plastic and probably most other filaments that I have not yet used will cook off and bake into a hard crust if held at printing temperature for too long. You can clearly see this on the outside of the nozzle. It’s not as serious inside the nozzle but it will eventually cook off just the same. This is one cause of nozzle plugging.
I have printing software that lets me hold the temperature of the bed and nozzle between print jobs. (Simplify3D) This is OK for a quick restart when immediately starting a new print job. Wandering off (for lunch) and leaving the heat set on the nozzle will one, drain (ooze) plastic out of the nozzle and the entering air will oxidize and cook off the plastic, plugging the nozzle.
The odor from the heated plastic is the volitile components cooking off. When they are gone what is left is mostly ash.
Time spent reheating the cool nozzle is better spent than changing out a plugged nozzle.
This is the basis of Fused Filament Fabrication (FFF) 3D printing. It is the one major variable that must be completely understood. I have been through at least 20 spools of both ABS and PLA. There are more than a dozen other blends of filament. I haven’t “mastered” these first two yet.
The situation is there is variation in the product of even the same color, material and manufacturer. Every new role is a new experience in getting all the “tweeks” correct. Temperature and extrusion rate are the most critical.
I am currently running some black ABS that measures average 1.70 mm (undersize 0.05 mm) yet I have had to decrease flow to .85 (85%) to prevent balling and tear-out due to excess flow.
The tip and lesson learned is to not get too excited about “favorite settings” there is no constant magic number that is correct for every roll of filament. Run a couple of flat infills to test a new roll. I print a shallow box about 25mm x 50mm. It gives me at flat area to check infill and base adhesion and some sidewall to check layer bonding. I make “tweeks” until I get a good box then I go on to the real project.
My rolls are usually (all so far) consistent as a single unit but there is vast difference between rolls of exactly the same color and brand. It’s not all just the material. There are so many variables to FFF that the user must learn from experience to recognize any issue and make appropriate corrections.
Bad product can exist but I am certain most issues are user correctable and many are user caused.
Experiment with settings. See what happens with over and under feed. Move temperature and observe the results. Vary speeds. Be scientific, do one at a time.
Don’t worry about the cost of material for testing. It is insignificant to spoiling a major build. (shown)
A Good Clean Nozzle is Critical
A nozzle does not have to be plugged to be “bad”. Nozzle flow can appear to be normal but in fact the tip could be partially plugged. For me, changing a nozzle is one of the last things I want to do. However, in many cases I have discovered it should have been the first thing!
The first clue is that shells don’t bind together on their sides and infills are sparse and stringy. Adhesion problems with the first layer is also a good indicator. These problems often appear when nothing as far as flow settings have been changed and the printing has been problem free for quite some time.
Many newbies want to blame the filament. In my experience that is really the LAST thing that can be causing the problems just described. The diameter can be measured and flow rates tweaked. Most cases the filament may be a bit undersize. If the adjustment doesn’t change the low flow issue then it is a sure indication it is the nozzle.
The 0.04” nozzle may be acting like it has a 0.03” or even a 0.02” hole. This equivalent bore is likely to be a bit askew. If the flow test done well above the surface is not a straight down flow but tends to spin or do other weird flows, then for sure the nozzle needs changed.
A nozzle drag across a print bed or across a previously printed portion can be the cause. I find the outside appearance of the nozzle is a good clue. All nozzles eventually get a bit “crusty” looking after long use. It may be some of the visual debris that gets too close to the nozzle hole.
Why it happens is not too important unless it starts happening extremely often. Bad looking nozzle and defective printing means change out the nozzle.
Spare nozzles are reasonably cheap. Keep a half dozen on hand. Order more when you get down to one. Drilling them out doesn’t seem worth the effort to me. The nozzle is a critical precision flow device. A burr or an incorrect angle can only cause more grief getting a good print.
Nozzles should always be changed when they are HOT, at printing temperature. That’s the reason I dislike changing them. My fingers are not heat proof…
My observation is that modern society is very competitive. For some people life is all about winning. It's as if everything they do is a competition, and winning is obtained by being the fastest. Success equates to speed. That's grand and noble encouragement for the slow and lazy, who don't try their best. But best is not always about speed. If the goal is quality, many times speed is the last consideration. There is an adage about craftsmanship, in the form of this question, "Do you want it done fast or do you want it done right ?"
3D printing is like sex. It is mostly about doing it correctly. That means finding the right speed for the best quality.
Uh... Back to 3D Printing, there are two speeds to consider; rapid and print. They both directly relate to the velocity of the print head. This is the speed to which there seems to be the competitive urge. Other factors influnce overall print time are not a measure of velocity. For example, infill percentage, number of top and bottom solid layers, number of shells are some of the major factors. They determne how long the "race" (print job) will run but not how fast to go.
Rapid speed is how fast the hardware can move when not printing. It is dependent on the design limitations of the hardware. Some factors that determine rapid speed are mass, inertia, friction, power, accuracy of control. I am not going to explain these. Manufacturers usually specify the limits for a machine's rapid speed. There is another adage worth considering. it says' "Just because you can, doesn't mean you should."
Fast rapid speeds are good when moving far distances. Less time doing nothing. But using top rapid speeds all the time are hard work on the hardware. The closer to the limit speed the more wear and tear on the hardware. It also makes no sense to travel at 200 MM/sec when you only have to move 20 MM. Starting and stopping require ramping speed up and then down. So it will never travel 200 MM/sec in that 20 MM move. If your work-space is only 200 MM wide then 100 MM/sec is still very fast. If a printer is "banging around" with hard starts and stops, then the rapid speed is probably way too high. Also ramping may be inadequate. Adjusting ramping may let long transits run at higher speed, but that is a whole 'nother technical subject. Just selecting a lower rapid speed is much easier to implement.
The most important speed is the printing speed. Because a printer can rapid-travel at 200 MM/sec does not mean you can print at that speed. Well, maybe you can, but the thought expressed by the adage is, consider if you should. For me, the goal for most prints is to obtain desired quality as reasonably fast as possible. I realize I don't have the same "reasonable" expectations as another person. Also not the same acceptable quality standards.
My Cartesian machine printing speed is rated up to 100 MM/sec. I use that for rapids. I have run rapids on it at 150 MM/sec but it does nothing to reduce print times. My best print speeds are around 40 MM/sec. Print quality deteriorates at 50 MM/sec due to ringing (vibrations) showing in the print. I have experimented extensively and I know where the sweet spot is for my machine.
I will also say that my delta type printer can run and is set at a faster print speed of 50 MM/sec because the moving Bowden print head provides far less weight and inertia (to overcome) at higher print speeds. The delta does not move the heavy print bed surface at all and therefore the printed object is never being moved. That means less vibration and ringing "ghosts" produced in the print. Truely a great design for fast printhead movement. However, there are other limiting factors about a delta to evaluate.
Tip #6 is, don't think faster printing is better. 3D printing is all about the technique.
Head/Nozzle Temp Control Critical
Every experienced FFF (Fused Filament Fabrication) printer owner/user knows temperature is the most critical factor in producing great looking prints. All the other factors are important too, but the entire process is all about the melting, extruding, and cooling of the plastic material. Temperature is the key.
My printing machines are low cost, but reasonably good designs. However, one critical part that needs improvement is the temperature sensing of the nozzle/printhead. An aluminum block is heated by a thermal resistive element inserted into the block. The nozzle screws into the block. The filament feed tube screws into the opposite side. These two parts must meet inside the aluminum block and be pressed firmly together. There is considerable hydraulic (melted plastic) pressure on this connection. Tight connection is required to prevent leaking of molten plastic.
All these components must be accurately maintained at the desired melt point within plus and minus one (1) degree. I have worked with much more unnecessary variation, sometimes 10 degrees, until I discovered the solution.
With my machines, the problem is the aluminum block that connects everything as a heating block, simply has a hole drilled for the insertion of the thermistor temperature sensor. The thermistor is held in place only by the wiring. The hole itself is too large. A very poor design.
I have had the thermistor come out of the hole due to flexing of the wiring in operation. I fixed the wiring flex with a tie wrap but the poor fit and sensor contact remained.
I had some red colored high temperature silicon seal (The kind in a squeeze tube) I used to insulate wiring connectors on a printer hot bed. I tested the red high temp sealer to see if it could withstand the printhead temperature (200 – 250 C) It has been holding well.
I applied a small dab of the sealer on the sensor hole and wiring. On testing a printing project, this simple fix immediately provided the stable print head temperature control I wanted. One-degree temperature control when printing.
The little blob of sealer should be easy to remove if I ever need to change the block or the thermistor. So easy a fix for critical temperature control.
I need to find more uses for the big tube of sealer left over… Ha!
The symptoms are terrible looking erratic, hit-and-miss extrusion flow. Gaps appear in flow and uneven heights between rows. In severe cases the extruder may start “clicking” as the drive wheel slips on the filament trying to force the flow.
At first, I though it was signs of a plugged nozzle, and sometimes it is. The cure is to replace the nozzle. I don’t believe trying to clean a nozzle has any long-term benefit. I might consider it if it was my last nozzle.
There is another culprit that will generate these symptoms. Unless the printer has an “all metal” hot end, there is a PTFE plastic tubing inside the hot end feed tube. I have discovered if I have been carelessly experimenting with high nozzle temperatures, I can cause this PTFE tube to shrink in diameter. I have experienced a reduction of 50% and possibly more. This severely restricts nozzle flow. I have experienced the PTFE tube shrink small enough to be forced into the nozzle itself.
It was this low flow that tempted me to increase nozzle temperature. It was the dog chasing its tail. The higher the temperature, the more the tube would shrink.
The cure is to replace the entire feed tube/pipe, and also the nozzle while I am at it. On my Cartesian machine it is an all thread pipe. Replacing the pipe a bit of a chore but it is the only solution. The PTFE tubing can't be replaced by itself on this machine. On my delta, the Bordon tube feeds down into the hot end pipe, so I haven't had the issue there.
It feels so wonderful to have a proper flow again.
My delta may be considered an all-metal hot end. I am not certain. But I know why they exist. To avoid this PTFE tube shrinking problem, especially with materials that need high extrusion temperatures. I don’t know if all metal tubes have other issues of their own, but it is a conversion worth considering, especially if running high temp materials.