Printing Cylinder Head object issues

Hi all,
I am trying to print a cylinder head as part of a rotary engine build. Unfortunately the print leaves a lot to be desired, I printed the object on its back so the fins of the cylinder head are printed vertically. I tried printing the cylinder head on its top but the fins just became a congealed mess.
The first issue you can see from the rear view is the supports confused the build process and the fins seem to have merged together producing an unsightly mess. The side view shows the fins are not smoothly produced but seem to be blobs of plastic rather than a smooth edge. Thirdly the face view shows the circular fins have become dipped creating a trench.
To better understand the difficulty for the printer the fins are only 0.75mm thick and the gap is 0.5mm. The printer is a Fisher Delta with a heated bed set at 60 degrees and the nozzle at 200 degrees using 1.75mm filament.

Has anyone come across this issue and hopefully a resolution? I have seen similar 3d prints on line but not the particular engine i'm building. Any help in resolving this issue would be gratefully received.

Many thanks
Gary

Try printing more than one and use a cooling fan!

You have an over heatng problem there, or, to say it better, you are printing faster than the plastic can cool before the next hot layer is put on top of it. If you are useing PLA (I guess for your temperature), you need to use a more agressive layer cooling. Put you fan at 100% and if that is not enough... you need a better fan system or as previously stated: print more than 1 object so each layer has more time to cool. As a general rule, for pla, you need at least 10 seconds to let it cool before putting another layer on top without worrying about heat.

Many thanks for your help, the fan forms part of the head assembly and is running at 100%, therefore to cool the printed object faster I will need to apply additional cooling. Can I use an external fan or does it have to be directed to the part more accurately. If I can use external fan and suggestions?
Regards
Gary

I would have thought you will also have issues due to the size of the fins. What size nozzle are you using?
0.4mm is a typical size which the slicer calculates as a real width of around 0.68mm. If your fin width is only 0.75 we hope your slicer realises and just prints a single run at the 0.68mm. However it has to stop at the end (outside of the fin) which inevitably causes a blob (because the nozzle is stationary and there is pressure in the chamber) . So your fin will be thicker at the edge. It also will be affected by any movement in the hot end (inertia). And at the bottom surfaces you are printing unsupported.
Looking at the picture the gaps look bigger than the fins so I assume it is just doing a single run.
Obviously if you are in to detailed parts like this you can go for a smaller nozzle.
On the support structure you may find that it is a mess because it doesn't scale down to the detail of the model. By that I mean it is usually set with a larger spacing as an after thought so any support "tower" will not necessarily align with each fin.
As has been said - temperature is a big factor for fine detail.

MCcarman, you are somehow wrong: With a 0.4 noozle, you can print fine details upto 0.4 wich s the noozle size. 0.68 is the MAX width you can extrude witha 0.4 noozle (in fact you can extrude more, but not in a consistent way)

Going back to you Gary, I dont understand what you mean with the fan being part of your head. You have to have 2 fans in your extruder: the cold end fan, and the layers fan. The layer fan is meant to avoid exactly the problems you are haveing. If you already have your layers fan running 100% and that is not enough, you have to check that the fan shroud is propoerly pointing to the extruded material (so it starts cooling right away after going out of the noozle.)
After checking that, try again and if it still not enough reduce printing speed to something around 30/40mm/s and if that it is not enough then you need a better layer cooling fan / fan shroud. Axial fans are really not good for cooling because once you install them on a fan shroud there will be some flow restriction and axial fans hate that and are not able to push the aire. Whenever you can install a radial fan system, you will have plenty of air with that.

Hope it helps. In case you need, post pictures of your layer fan / shroud and extruder

Thanks MCcarman and Tinchus for your assistance, I am a complete novice when it comes to 3d printing but learning fast. The unit I purchased is a Fisher Delta printer which required assembly. The fan is a small axial 50mm fan and directs air to the nozzle; so I assume this air is to cool the filament as it is extruded from the nozzle.
I am using Simply 3d software to produce the g-code and supports which are currently set at 5mm. As mentioned I can reduce the flow rate and produce a second cylinder head along side. The nozzle is a 0.4mm unit.
I can apply artistic licence to the part as it will be fitted to a semi scale RC aircraft. However, I have seven cylinder heads to produce for the rotary engine and they will need to look the part. It was mentioned that a smaller nozzle could be used for this type of detailed work, are they readily available?

Again many thanks for your help

p.s Forgot to mention that I am always open to suggestions and the suggestion of additional cooling is of interest and would be happy to receive any detail relating to this.

Many thanks
Gary

Thanks Tinchus. I wondered why the 0.68 was quoted as I always assumed 0.48 when designing my parts. I assume you meant you can print detail down to 0.4mm. I would have thought that anything less than that requires under extruding which would give unpredictable results.
Can you help explain how this width calculation works as its often shown in the Gcode files on this site which implies its the width being used. Does the 0.68mm override any extrusion multiplier - if I specified 200% with a low layer height that would result in a width of 0.8 would this be restricted somehow to 0.68.
In this case the problem stays the same as a 0.75 fin really wants a profile of 2 runs of 0.375 but there is probably still going to be a blob at the ends as the nozzle/bed stops..

0.68 is the max width for an extrusion to be made with a 0.4 noozle. You can actually extrude more than 0.68 but not in a predictable and controlled manner. So, the formula noozle size X 1.7 is what you use to calculate the max width of extrusion for certain noozle. For a 0.4 is 0.68, for a 0.3 is 0.51 and so on. In some slicer you are goin to se that number (assuming you have a 0.4 noozle) for many reasons: 0.68 is teh number that will give you the strongest bond between layers. Also 0.68 used for the infill is what is going to give you fastest infill (the % of infill is calculated as a %, so if you set a 10% infil, it will be calculated as the 10% of the object volume, as example, lets say that if you use 0.68 you would have 10 lines of infill, but if you set infill width to 0.4 you will have 16 lines of infill)
Also, you can not have unlimited extrusion through the noozle, there is a limit and that limit expreseed in mm3. And that limit is also going to restrict the max width you can have in an extrusion (for example, there is no way you can get a 1 mm width extrusion useing a .4 noozle, no mattter how much pressure your extruder can apply on the plastic.)
Adn regarding the extrusion multiplier.... you are mixing concepts. 0.68 is the width of an extruded line, the extrusion multiplier, to say it somehow short, is the man in charge of getting that width done.
And going back to the 0.75 fins: all slicers (of my knowledge), will try to solve that problem as best as posiblem for them. In slic3r for example, in auto mode, you will get a 2 lines extrusion, of 0.4 each line, makeing a wall of 0.8, with no blobs at the end ( I have rinted a lot of obkects like that).

Hi Tinchus, I hate to come in with contradictions when you're being helpful, but I think you have a couple of things a little mixed up here. The max extrusion width isn't a hard limit the way you suggest, but more of a guideline. The upper limit depends on the nozzle geometry and specifically on the outer diameter of the nozzle tip. If you look at the drawings for the E3D nozzles (as a well known example), they give a table of sizes for the different nozzle geometries and the tip OD is column B: [wiki.e3d-online.com] They are using 2.5 x the bore for the tip OD. I've used widths of 2x bore and larger without problems.

You're right that max flow through the nozzle is an issue, but again it's not so much a hard limit on usable width as an issue of practicality. If the width and the height of the extrusion are sufficiently large that you run into the max volumetric extrusion rate then the linear speed of the print head will have to be capped to match it. Newer versions of Slic3r allow you to specify the max extrusion rate and will limit print speeds to respect it. So pushing very wide and thick extrusion through a very small nozzle bore is counter-productive and somewhat pointless, but using wide thin layers may well be useful.

However, this is all a little off-topic for the OP's problem. I agree that it sounds like the way forward is additional part cooling. I'm still not clear on the fan arrangement - perhaps we could have a photo? A smaller nozzle so that the minimum usable extrusion width can be smaller to match the object detail size might also help, but since parts of the model in the photos look fine I don't think it's the main issue.

But I never said is a hard limit (may be excuse me but english is not my mother language so...), as I said, max width is the RECOMENDED limit you should not go over it for aceptable predictable results. Talking about noozle geometry is going inside an issue wich ahs been discused somehow on the net, mainly regarding heating issues. And from the point of view of an end user, geometry of the noozle is somehting "transparent" to him. From the end user point of view, nobody will calculate extrusion or max width, etc measuring the noozle sizes, specially because not all used noozles are original E3D. In real life, you make an extrusion calibration, and the result will be the necesary steps neeeded to get certain exact extrusion width, that is all it matters. And if you use slic3r (others slicers also do the same and uses the same formulas) the extrusion width is not calculated useing the noozle geometry, apart from the noozle size (the size of the little hole),nothing else of its geometry is taken into account. The exact information and getting rid of my somehow bad technical english is here: [manual.slic3r.org]
And regarding the max flow: no, it is imposible to go over certain max amount of flow, not matter hor hard the extruder pushes and how slow the linear speed is. Teh plastic is not water, its viscosity is high and once it is out of the noozle start to get solid and that will restrict the amount of plastic it can come out, at certain point, if you extrude ALOT and go really slow, you will end with a blockage.

If you take a look at CURA you can experiment with this: yo can set manually an extrusion and speed, at one point cura will warn you with red letter saying you are extruding (or trying to ) above certain max level, wich is exactly the max flow that noozle can handle.
If yu have extruded 2X time the noozle size, may be is posibe, but I really doubt you can make 2 lines useing that extrusion and that those lines have the same widht (when I say teh same I mean exactly the same size).
To the general and practical purposes of an exact estrudeed print these are the facts (no my opinion, is the info you can find on the tech notes on the net): min extrusion width= noozle size, max extrusion width= noozle size X 1.7. Any values above or below that are posible, but not accurate and stable (in fact slic3r will fill gap spaces usein smaller values tahn the reccomended ones)

The problem of that print Im almost sure is because of lack of cooling, either because it is printed too fast or because the layer fan is not enough or properly pointd/working.
I would recommend you the COBRA fan shroud, it is AWESOME, you can find it in thingiverse.

Hi Tinchus, I think we're mostly in agreement, and I apologise for pulling this thread off-topic. I did a quick test tube at 2.5X nozzle size, and rather than adding more to this thread I posted it here: [forums.reprap.org]