Firmware speed extrusion multiplier = f(target extrusion rate)



  • Problem:
    I am used to print at average speeds of about 150 mm/s up to 300-400 mm/s. One problem which occurs is, that the real extrusion rate is a function of speed or target extrusion rate, which looks about approximately like "real extrusion rate" = "target extrusion rate" ^ factor, the factor is < 1 until the real extrusion rate is maxed out because of grinding/missing steps.

    The interesting part before the extrusion rate achieves its maximum limited by "missing steps", has nothing to do at all with missed steps etc. The forces pushing the filament are transferred not only by friction, they are transferred by non elastic deformations to the filament. So the filament "flows" and the shear rate/force is somehow proportional to the deformation. This can be seen by investigating the extruder wheel marks on the filament. The higher the speed is, the closer the marks will come together. If the filament is soft and hard to melt (small nozzle, low temps), the problem is maximized.

    As long as one keeps in the low speed range, e.g. <80-100 mm/s this effect doesn´t matter as much, although it is also there and can be seen. But when operating in higher speed areas, this leads to uneven extrusion e.g. too much at lower speeds and too less at higher speeds. I use e.g. PLAs where the extrusion multiplier is about 30 % higher at 300 mm/s compared to 50 mm/s. In general people are starting to increase temperatures when seeing under extrusion at "higher" speeds, but quality wise higher temps are no solution, because of cooling/warping issues.

    One method to overcome this problem partially is to use high accelerations, by that the average speed is higher and the average speed also doesn´t vary as much over time as with lower accelerations. So adjusting the extrusion multiplier for high speeds and high accelerations will work, but will still lead to over extrusions at lower speeds.

    Dual wheel extruders like bondtech do help, but cannot change anything to the principal behaviour.

    Solution:
    I did measure the relation between "real extrusion rate" and "target extrusion rate" for different filaments, nozzles, temperature and speeds (target extrusion rate vs nozzle diameter). But this is time and filament consuming and until now, i cannot compensate it - i just can measure it.

    1. With a filament sensor (which is also able to keep up at e.g. 200-400 mm/s), it would be very easy to get such curves (just vary the extrusion rate at given nozzle, filament and temperature setting and read/store the filament sensor output).

    2. With the measured relation between real extrusion rate and target extrusion rate one could use a simple piece wise linear or e.g. a power factor approximation to adjust the extrusion multiplier accordingly. I would suggest a 2nd "speed extrusion multiplier" in the firmware to account for this problem.

    3. Christmas is comeing and i would like to see a filament sensor and firmware which is able to control extrusion rate closed loop. I know this is just a vision, but maybe, when the target is not to control the extrusion rate every ms, but adapting the "speed extrusion multiplier curve" online under certain conditions, it is possible sooner.



  • Two things to say:

    1. I'm very interested in this relationship between speed and extrusion and your findings. when I have done some basic testing along these lines my presumption was that when I started to see under-extrusion as speed was increased, it was caused by inability to melt more filament in the hotend, rather than compression of the filament by the extruder.

    I started to notice under extrusion on E3D volcano (with 0.4 nozzle) at 180mm/s. However I did not think to boost the extrusion multiplier by 30%, I might try this later on this evening.

    1. I am impressed to hear about printing speeds at 300-400mm/s what setup do you use to achieve this?


  • The reason that the "real extrusion rate" is different from the "target extrusion rate" when printing at high speeds is that you have exceeded the rate at which filament can be fully melted. For any given hot end, this is a function of temperature and time. Because the filament itself is a poor conductor of heat, it takes time for the heat to reach the inner core of the filament. As you have discovered, cramming the filament in faster by increasing the extrusion multiplier will have limited effect other than to increase pressure inside the nozzle. This will of course lead to over extrusion at slow speeds but it will have limited benefits at high speed. You might be able force more filament out (at high speed) but it won't be fully melted. You'll find that the inner core may only be at or around the glass transition temperature so layer adhesion is likely to be poor. The other big factor is that in order to reach high speed over the short distances that we use, you need to have high acceleration. Whilst it is possible to achieve this for the print head, and the extruder will attempt to follow this acceleration, you cannot accelerate the time it takes to melt the filament so all that happens is that you get a pressure spike inside the hot end resulting in under extrusion during the acceleration phase. Then as the acceleration phase ends, the pressure that is still inside the hot end causes over extrusion. The same happens in reverse at the end of the move. We already have a mechanism for dealing with this which is "pressure advance".

    IMO, once you get close to the melt rate of the filament, then varying the rate at which filament is being crammed into a hot end isn't the answer. Pressure advance helps but the only solution is to melt the filament faster by increasing the surface area of the melt chamber but there is still a finite time limit.

    You might find these blog posts interesting https://somei3deas.wordpress.com/2017/06/22/exploration-of-print-speeds-with-a-diamond-hot-end/ and https://somei3deas.wordpress.com/2017/06/25/duet-pressure-advance-experiments/


  • administrators

    @vp:

    …The interesting part before the extrusion rate achieves its maximum limited by "missing steps", has nothing to do at all with missed steps etc. The forces pushing the filament are transferred not only by friction, they are transferred by non elastic deformations to the filament. So the filament "flows" and the shear rate/force is somehow proportional to the deformation. This can be seen by investigating the extruder wheel marks on the filament. The higher the speed is, the closer the marks will come together. If the filament is soft and hard to melt (small nozzle, low temps), the problem is maximized.

    That appears to confirm that the problem lies at least partially in the mechanism that the extruder uses to push the filament, i.e. the teeth of a hobbed shaft gripping it. Your proposal for increasing the commanded extrusion rate with speed is interesting (actually it's the pressure that it needs to be compensated for rather than the raw speed).

    A complicating factor may be that the filament may be getting compressed in the extruder drive, so that it comes out of the drive with more average cross sectional area than when it went in.

    Another solution may be to use a different extruder drive, such as the type that pushes the filament between two belts.



  • I know E3D are experimenting with fins in the filament path to try to increase the effective surface area inside the nozzle to increase thermal conductivity to the filament. It's whether these then cause jams, friction etc… Their discussion with Tom from youtube was very enlightening on this topic https://toms3d.org/2017/09/29/live-qa-e3d-online/ although I distinctly remember them saying that 3mm seemed to be less problematic at very high melt rates than 1.75mm which is slightly counterintuitive.



  • Hi,

    a lot of comments, i try to catch them all. Here is my approach to high speed printing and extrusion problems:

    1. I am not interested in a "hardware" solution. I want to push the software as far as possible. Hardware can be changed in addition to software.
    At the time being, you won´t find anything better than dual wheel extruders - and they also just reduce the problem. Belts doesn´t work for high speeds, they slip because they try to achieve grip only by friction….. it did not try belts, but reports in the web are very clear.

    2. The extrusion limit is by far not defined by the thermal extrusion capacity of the nozzle (it is no problem to use a longer nozzle….). The behavior i have described above is for sure due to the increased pressure which is needed to achieve higher extrusion rates. In general, with normal e.g. e3d v6 nozzles and small diameters, this pressure will be more or less produced by the last 2-3 mm at the tip, where the plastic is squeezed through the e.g. 0.4 mm hole. This has nearby nothing to do with "it is not completely molten". For sure the viscosity is a function of temperature, but again, it is never a problem to use longer nozzles.... The flow is laminar at the tip, so as rough approximation the pressure built up there will increase about linear with speed. But especially PLA (and in my opinion PETG also) is known to get very sticky when melting. A very sharp transition zone from rigid to liquid helps. This kind of "friction" there is therefore getting more important at higher speeds. Users which uses MaPa (low friction nozzles) report that they are able to reduce the pressure. Because laminar flow doesn´t care about roughness, the decreased pressure with low friction nozzles can only be caused by the reduced friction before the 0.4 mm hole.

    Extrusion problems (not all but many) at high speeds come from varying extrusion, which comes from varying speed. The higher the acceleration the smaller is this problem, because the speed is more equal. The visco part of the liquid acts like a low pass filter. Any moves which are quicker than its "cut off" frequency will not go through the nozzle. Over extrusion is not nice, but the print will not fail, under extrusion is "game over".

    3. Transition zone. Gets more important with speed. Should be as short as possible to avoid friction, cool the cold end as much as possible and e.g. use titan throats.

    4. Extrusion capacity is not printing speed. To find out what the extrusion capacity of a certain hot end is, there is no need to print anything and waste material. Just heat up the hot end and command the extruder to extrude with a given speed. Increase the speed until missing steps or in my case grinding occurs - this would be the very hard limit. Do the same procedure and measure the real extruded filament length to find out how much you should have to compensate the extrusion multiplier as a function of speed.
    But when printing under extrusion will start much much earlier, because the visco part of the liquid will produce a phase shift (time lag). This phase shift leads to slower pressure build up (or change) as needed - and this cannot be compensated with the existing pressure advance. If pressure advance would also adjust the "time lag", it could help.

    All of these problems are no real problems at 50 mm/s, but the higher you go the more problems will occur…. and at bout 150 mm/s (real printing speed, not maximum allowed speed in the slicer) and above it determines the difference between constant extrusion capacity and real life extrusion capacity.
    My real hard limit is cooling. At a minimum layer time of about 6-8 seconds i can print "normal" overhangs with 150 mm/s at 0.2 mm layer 0.4 mm nozzle and also sharp corners without problems. But some when the cooling problem is the limiting factor. All you can do is to reduce the amount of plastic to let it cool quicker => small nozzles (0.4 mm) and small layer heights. Or to use filaments which have a high glass transition temperature and a low melting point, like BDP filament from extrudr.

    But when it comes to "stupid" top and bottom layers, when cooling is no issue, one can utilize the extrusion capacity and 300 or 400 mm/s is not a big deal - for sure not at 210 °C....this happens in my case at about 235-240 °C with PLA which can be printed at lower speeds at about 200-210 °C

    5. Myth about nozzles. Forget the nozzle material and forms, all metals which could be used including titan have a thermal conductivity which is multiple times bigger than any plastic. All what matters is the active surface, which is defined by the length of the nozzle and the filament diameter. 1.75 mm is much better than 3 mm filament, for pressure and thermal reasons. Fins will not help to improve the "pressure drop/thermal flow" relation with a laminar flow. It would help a lot to increase the diameter right after the transition zone, but than retraction won´t work anymore.

    I am thinking about to split the "nozzle" into 2 parts. An entry zone, where the temperature is much higher than the target nozzle tip temperature and the nozzle tip zone, where only the final temperature is controlled. Because of the higher delta T at the entry zone, the filament would melt much quicker, so the transition zone would be much shorter. To make this happen, one would have to separate these zones thermally, that mean no metal connection…. that is tricky for DIY, if all together should get smaller.

    "Diamond" hot ends have their certain advantages, but increased printing speed is it for sure not. Because the biggest pressure drop is created at the very last mms (e.g. the 0.4 mm part), it is very important to keep this part as short as possible. Because of the "Diamond" design, this part is multiple times longer than compared to a e.g. v6 nozzle and by that the pressure drop is also multiple times bigger. Because the pressure distributes equally, i doesn´t matter if 1 or 100 extruders push against it, all have to push with the same force. The only difference is, that in theory (without stepper drivers) the netto power (not torque) which is needed per stepper would be reduced. But this doesn´t matter at all. So in the end - when the target is high speed printing - a Diamond hot end just adds weight, which is not what we want to.



  • Interesting points vp, and you do sound as if you have some knowledge in this area. A couple of things I'd like to pick up on, as they seem contrary to my experience so far (which could be because I have been making wrong assumptions or not).

    a) Transition zone, I am now using water cooling for my heatsink which is never above 28 deg C even in a heated chamber. I should expect to be able to extrude a higher volumetric rate of filament all other factors being equal?

    b) Are you printing 300-400 mm/s with a 0.4mm nozzle? As above I ask as I have not managed to exceed 9.1mm3/s via volcano/0.4 without significant under-extrusion, although I have not tried "boosting" the extrusion multipler under these conditions. Although I did try increasing the temperature which had as you point out only minimal effect. (210 to 230 for PLA).

    c) Have you seen or tried Prometheus hotend where you can alter the length of the melt zone? Are you suggesting that short with a sharp transition would yield best results?



  • The empirical data that I have collected is at odds with your theories so we'll just have to disagree. I can't bring myself to believe that time in contact (with a hot surface) is not a consideration.

    Oh, by the way what you say about the Diamond geometry is completely wrong. The 0.4mm section is only 3mm long for each filament and then there is a combined 2mm section so a maximum of 5mm which is about the same as a V6 and not multiple parts longer as you state. Of course, the diameter isn't always 0.4mm -mine are mostly 0.5mm but I also have a 0,9mm version.

    Anyway, we won't dwell on that. You are obviously very enthusiastic about your theory. Personally I just can't buy into it.



  • @ DjDemonD.

    ad a. my understanding is, that when filaments get soft, they get "sticky" (especially PLA). The shorter the transition zone, the shorter is the "high" friction zone. Increasing cooling leads to a sharper/bigger temperature gradient, which will make the trans zone shorter - this is good. On the other handside, if your cold end cools e.g. 20 W more than usual, your hot end has to provide the power - beyond a certain point this might get tricky. But i would expect that your constant thermal extrusion capacity is increased when cooling more. With air i had never the problem that i did cool too much. What should not happen is, that your transition zone is shifted inside the hot end - which is possible (at least in theory).

    When priniting PLA, PTFE inliners which go directly to the hot end (through the throat) also allows about the same speed as with a full metal titan throat. So i guess it is really about the friction in the transition zone.

    ad b. I have 2 printers, a slow one with an e3d v6 setup and a fast one with the following hot end (ubis style).
    https://www.aliexpress.com/store/product/E3D-Upgrade-parts/1654223_32697889176.html.
    With e.g. extrdr MF PLA 240 °C i have no problem to extrude with a 0.5 mm e3d like nozzle 1.75 filament with 10 mm/s, this is about 24 mm³/s or equals at 0.6 mm layer width at 0.2 mm layer height and 200 mm/s. And this is just a "e3d v6" hot end - no volcano. But i cannot print that fast with e.g. retraction. Any disturbance of the "constant extrusion capacity" will give underextrusion. At top/bottom layers this is no big problem (no retraction).

    With a volcano nozzle and the same hot end (no long e3d heating block) i manage > 300 mm/s at same conditions. If i reduce the layer height to < 0.2mm e.g. 0.1 mm 400 mm/s are no problem.

    ad c. i did/do combine a volcano nozzle with the above ubis style heaterblock - the heater block is not located towards the nozzle tip, it is located towards the cold end, before the throat. So i did experiment. Placing the heater block towards the cold end, just gives a very sharp temperature gradient == short transition zone and i prevent, that the transition zone is shifted to the nozzle. In terms of heat transfer the contact area has to be "maximized" (but not more as needed otherwise friction is increased). If the transition zone is large (or already shifted into the hot end), it just means that the surface temperature is smaller than possible - this means you lose thermal capacity. I did not try the Prometheus hot end, but various nozzles up to 35 mm long.

    @Deckingman:
    thanks for "enthusiastic", i am just a nerd….
    If i do a quick google for me it seems that the last part is much longer than at e.g. e3d nozzles. Maybe i found the wrong drawings.

    The time in contact is for sure important, but not (always) the limiting factor. For a given viscosity, flowrate and nozzle you need a certain "pressure". As long as your thermal capacity is big enough, it doesn´t limit you. But e.g. if someone uses a 0.2 mm volcano nozzle the length of the noozle is for nuts, it just increases friction. If you try to print really fast with e.g. 0.5 mm increasing temperature helps - why ? Going from 210 to 240 °C gives only about 30/190 = 17 % more deltaT (which would give you exactly 17 % more speed), but the viscosity will drop by a factor of e.g. 2 or more.... and by that the pressure is reduced. Increasing the temperature wouldn´t help as much if the thermal capacity is the limitation. The pressure built up seems to be the limit.



  • @vp:

    If i do a quick google for me it seems that the last part is much longer than at e.g. e3d nozzles. Maybe i found the wrong drawings.

    I'd say that you probably misinterpreted the drawing. It's an easy mistake to make because at first glance the labelling does show a 24mm long section at 0.4mm diameter but when you look closely, you see that this is counter bored to 2.0mm for 21 mm of that length, leaving just a 3mm long section of 0.4mm diameter. Trust me, I own enough of them to know that this is the case. But of course, the drawings on the web are all generic and if you order a Diamond with a larger diameter at the nozzle tip, then of course the 0.4mm diameter section at the bottom of the melt chamber is also increased. I have 0.5mm and 0.9mm versions as well as the "standard" 0.4mm.

    As a Diamond hot end user who has proven that it is indeed possible to print at up to 300mm/sec with a 0.5mm nozzle by exploiting the fact that there are multiple melt chambers, without increasing temperature or extrusion multiplier, you'll understand my scepticism about your theory. As I said, no hard feelings but from my own tests, I can't bring myself to accept your hypothesis. 10/10 for enthusiasm though. 🙂



  • Thanks for your feedback.

    The length of the 0.4 mm hole of an e3d v6 0.4 mm nozzle is according to this drawing https://wiki.e3d-online.com/File:DRAWING-V6-175-NOZZLE.png 0.6 mm - that is slightly longer than the diameter and needed to give the flow a chance to develop an even speed profile.

    According to this drawing http://reprap.org/mediawiki/images/7/73/Diamond_Nozzle.pdf the length of the 0.4 mm hole is multiple times longer than 0.4 mm - there is no value given, but just compare the diameter 0.4 mm to the length of the hole…. it can also be seen, that the longer part of 0.4 hole doesn´t have to take the whole flow rate, the flow rate there will be total flow rate/#nozzle. Unfortunately it is the wrong way to try to reduce the pressure by splitting the flow rate and multiplying the length of the 0.4 mm section, we have laminar flow, pressure drop is proportional with speed not with speed^2.

    1. We have laminar flow. The pressure needed is proportional to the speed. The cross section and speed is proportional to the diameter^2. Therefore the pressure drop per length of a 0.4 hole is (2/0.4)^2 = 25 !!! bigger than compared to a 2 mm section. So we don´t have to care about the 2 mm length too much in a "laminar flow" thinking. But it makes a huge difference if the 0.4 section is longer than needed. The minimum length is limited to the development of the speed profile (a too short 0.4 mm section would give an uneven flow).

    2. Experience tells us, that anything which reduces friction, reduces the needed pressure. A laminar flow doesn´t care about roughness, the roughness is covered by the laminar boundary layer. If reducing the friction reduces the pressure, this effects cannot be related to the "hydro dynamically" flow, there must be something else. I call it "friction" and this friction has to come from some where where the filament is not flowing like a liquid, so the laminar boundary layer has no effect. Therefore i assume it has to come from the transition zone. This friction part will be proportional to the surface which is proportional to the length of the transition zone. Everything which reduces the length of the transition zone will reduce the friction and the needed pressure.

    The problem is now, that at a constant nozzle temperature increasing the thermal active surface (nozzle length * filament diameter * pi) will lead to a reduction of the temperature difference between metal (nozzle) and filament. From that follows, that the transition zone gets longer because the thermal flux rate (W/m^2) gets smaller and the friction increases. The best nozzle length is just as long as needed to melt the material. Anything in addition will just increase the friction. So if the speed is increased, the part (%) of the pressure which comes from friction is reduced and the part (%) of the pressure which comes from "flow" is increased, up to the point where flow part limits (or the thermal capacity).

    3. There is no doubt that you can print > 300 mm/s with a diamond hot end, but what i meant is just you don´t need it. You don´t need that much "surface area" this is not a limitation for 0.4 mm nozzles. When it comes to bigger nozzles e.g. 0.8 mm the thermal capacity will be limiting much earlier, because the pressure drop is much smaller (0.4/0.8)^2 = 1/4 and the extrusion rate (== needed thermal capacity) at equal speed is much higher (0.8/0.4)^2 = 4 compared to a 0.4 mm nozzle. But in real life using bigger nozzles the real limitation will be cooling and not extruding, it is not hot end constrained then.

    4. To sum it up. There are 3 major constraints (flow, friction thermal) and all can be limiting, but the thermal capacity is not a real problem using small nozzles.
    A small nozzle will be limited by the pressure drop due to the flow, a big one by the thermal capacity. In between the friction will also be important.



  • This is starting to become tiresome.

    How do you explain the following?

    Using a single input on the diamond hot end, with a 0.5mm nozzle, I can get up to about 120 mm/sec at my normal print temperature before I see signs of under extrusion having an adverse effect on print quality. This is consistent with what most people see with a standard E3D V6. According to your theory, this should not be possible because as you say the diamond has a "restriction" that is multiple times longer than the E3D V6.

    Furthermore, we have established that the "restricted part" of a Diamond hot end consists of a 3mm long section followed by a 2mm section giving a total of 5mm@0.4mm diameter (although the diameter depends on nozzle size). So now when I employ all 3 inputs, the "restricted zone" becomes 3 x 3mm (3mm for each of 3 filaments) plus the "common" 2mm length. However you look at it, the "restriction" is far more for 3 filaments than a single one. Yet by employing all 3 inputs whilst keep the temperate and extrusion multiplier the same, I am able to attain print speed almost 3 times as high. Clearly the limiting factor in this case is NOT friction, nor the pressure drop.

    The reason why I am able to print at higher speed using 3 inputs compared to a single input, has to be that I am employing 3 melt chambers. This has two advantages. One is the greater surface area of filament to hot surface and the other is that for a given speed, each filament spends 3 times as long in the melt chamber compared to a single filament in a single melt chamber. Thus the main constraint to printing at high speed is the melt rate of the filament and not friction or flow rate through the nozzle.

    Forget all this stuff about Laminar flow too. It may sound good but it's not relevant over the tube lengths we are talking about. In any case you seem to have completely over looked the fact the molten filament is a non- Newtonian fluid.

    As I've said, 10/10 for enthusiasm but your theory is ….... I'm trying to think of a polite way of saying "just wrong".





  • @deckingman:

    Using a single input on the diamond hot end, with a 0.5mm nozzle, I can get up to about 120 mm/sec at my normal print temperature before I see signs of under extrusion having an adverse effect on print quality. This is consistent with what most people see with a standard E3D V6. According to your theory, this should not be possible because as you say the diamond has a "restriction" that is multiple times longer than the E3D V6.

    Furthermore, we have established that the "restricted part" of a Diamond hot end consists of a 3mm long section followed by a 2mm section giving a total of 5mm@0.4mm diameter (although the diameter depends on nozzle size). So now when I employ all 3 inputs, the "restricted zone" becomes 3 x 3mm (3mm for each of 3 filaments) plus the "common" 2mm length. However you look at it, the "restriction" is far more for 3 filaments than a single one. Yet by employing all 3 inputs whilst keep the temperate and extrusion multiplier the same, I am able to attain print speed almost 3 times as high. Clearly the limiting factor in this case is NOT friction, nor the pressure drop.

    The reason why I am able to print at higher speed using 3 inputs compared to a single input, has to be that I am employing 3 melt chambers. This has two advantages. One is the greater surface area of filament to hot surface and the other is that for a given speed, each filament spends 3 times as long in the melt chamber compared to a single filament in a single melt chamber. Thus the main constraint to printing at high speed is the melt rate of the filament and not friction or flow rate through the nozzle.

    The point is that you need something which has multiple times of the surface than e.g. a single easy and light volcano nozzle to print at the same speed a volcano nozzle is able too. I never claimed that you cannot print 120 mm/s with a diamond hot end….. i print 150mm/s with a single e3d v6 nozzle.

    Forget all this stuff about Laminar flow too. It may sound good but it's not relevant over the tube lengths we are talking about. In any case you seem to have completely over looked the fact the molten filament is a non- Newtonian fluid.

    Laminar or not has nothing at all to do with "newton" or "non newton" fluid. The difference regarding 3d printing of the "non newton" aspects is that the viscosity will get smaller with increasing speed - that´s all. But compared to the temperature effect vs viscosity you can assume the viscosity to be constant in the 3d printing speed range.
    The laminar boundary layer doesn´t know if the fluid is newton or not. Roughness doesn´t matter in a lamniar fluid dynamic sense for non newton as well as newton fluids. For friction roughness does matter.

    It may sound good but it's not relevant over the tube lengths we are talking about.

    Why? What else does matter (beside the friction you say it also doesn´t matter) ? If the nozzle is tooooo long, you are right, the tip of the nozzle won´t be a problem, because of the problems described above.

    Beside the popcorn i still believe that we can learn something useful from such discussions. Nobody is forced to read it and everybody is welcome to join;)





  • Is there an emoticon for "yeah whatever…"?

    @vp.
    What we have is your assumption of what might happen according to your theory(s) vs my knowledge of what does happen based on practical testing. If you want to convince me that your theory is right and that there is some flaw in my testing then you'll have to build a hot end based on your theory and present some "real world" test results. From what you say, this hot end would have an extremely short melt chamber because you claim that thermal capacity is not the limiting factor but friction is. So you can then present you findings to E3D who would be delighted to learn that their Volcano design, with it's long melt chamber is entirely the wrong thing to do.

    Another little flaw in your theory is this "you can assume the viscosity to be constant in the 3d printing speed range".

    That's not the case I'm afraid as PLA will hydrolyse (go less and less viscous). This process starts at about 170 degrees C and is a function of time as well as temperature. So at slow speeds, PLA is indeed much less viscous than at high print speeds due to time time that the filament has been heated. It's another reason why increasing the temperature helps with high speed printing.

    I'd love to continue this discussion along the lines of the complex rheology of non-Newtonian fluids and hear your views on whether you consider molten printer filament to be a yield-pseudoplastic fluid obeying the Herschel-Bulkley model, but to be honest, I'm starting to lose the will to live. 🙂



  • @deckingman:

    Is there an emoticon for "yeah whatever…"

    Hi Ian,
    This is as close as I can find…. (rolling eyes)

    (◔_◔)



  • Very interesting…. God read



  • I wish I had time to test the core suggestion that increasing extrusion multiplier non linearly will result in better high speed extrusion. It's surely just a case of printing a rectangle of the type Ian and I have both played with that allows max speed to be attained and then nudging the speed up with each layer and applying the increased extrusion multiplier. Well soon see if it works. VP what do you suggest should be the factor for the multiplier? Your other thread suggests a square relationship so % new extrusion multipler=%increase in speed2?



  • Hi Dj, see here https://www.duet3d.com/forum/thread.php?id=3728 These curves are based on one of my "high-speed PLAs", that means it is flow improved. Other filaments start to increase much earlier/steeper.

    I started with printing a cylinder in vase mode and increasing the speed and the extrusion multiplier together so far that it still worked out for me (the diagram in the link above was done differently, but that takes more time). For the beginning, a round object without corners should be fine. It should be big enough (e.g 150 -200 mm diameter) because in vase mode the layer time is really small and you will run into cooling troubles at higher speed. You need a constant flow, any acceleration is not welcome because you might get under extrusion. Or you are able to accelerate really fast. In this case, the extrusion rate disturbance is small enough to get filtered through the hot end/nozzle.

    The curves shown in the link are "constant extrusion capacity" curves, without discontinuities of the flow rate.


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.