huge hotend venting improvement

Quote
jamesf
May try your challenge this weekend!

Quote
jamesf
Also note that Reprap have changed the design in new Ormerod 2 kits, and put in single laser cut part instead of the old duct system (now part of documentation here [reprappro.com]

Quote
jamesf
I am considering having the part laser cut and going that way, but want to gain some more hands on experience as well. As Dave points out, a lot of this is down to experience and properly planning prints.

Quote
jamesf
I think a lot of newbies like me come into this thinking printing is printing - and in reality the true craft lies in manipulating method and machine from print to print depending on requirements.

Quote
jamesf
If you go look at the hot end construction documentation there are now images of both versions. The one we have, and the new acrylic design (without the printed ducts) included in kits since late October

Quote
dmould
The amount of cooling is very much a compromise, and it is certainly not a question of the more the better. Too little and the part can lose its shape, especially on layers with small areas like the roof of the house as you show, in addition bridges and overhangs do not print so well without cooling - but too much cooling and the layer bonding becomes weak (including the bond of the first layer to the print bed). There are also the factors of the hotend temperature and type of plastic that affect the optimum amount of cooling. How strong the layers need to be bonded depends on the type of part and what it is going to be used for, so an ornament would be OK with more cooling than, say, a vase (won't be watertight) or a gear shaft (will break under load).
[\quote]
Especially the part with the layer bonding is interesting. I haven't seen effects like that with my mod so far but i will take a look. So far with every print i compared the prints came out better. I also had no problems with warping so far.

Quote
dmould
Variable cooling is by far the best way to go, with the amount of cooling being varied according to the place in the part that is being printed. The Ormerod design uses the same fan to cool both the extruded plastic and the filament before it enters the hot nozzle, so while some speed variation of that fan would be possible, it cannot be slowed too much otherwise the filament would start melting before it got to the nozzle (which will create a blockage and gum up the whole works).

From what i have observered i can say that when the heatsink air must not go through the part venting there is much more cooling so i am pretty sure that my 6cm fan is
pretty oversized. I currently had an old spare from an cpu fan which is 1.5cm thick but i have seen 1cm thick 6cm fans wich have the same weight as the smaller fan.
So i think there would be some headroom going down with the turn rate of the fan. Also note that there is a small hole above the hotend which cools the hotend just below the heater
from above. I have the impression that this also helps controling the filament as the temperature gradient to the hotend gets steeper.

Quote
dmould
What is needed is a way to control the cooling of the extruded plastic independently of the cold block cooling. That could be done with a separate fan, or by adjustable baffles, or by using a different method to cool the cold end. In addition the cooling air needs to be directed pretty tightly, because if fairly large sections of the already printed part are cooled it will tend to make the part warp - it is only the top of the newly extruded plastic that needs to be cooled for a brief time. The weight of the X carriage cannot however become too great otherwise you will start getting mechanical problems, so design changes need to be considered very carefully.

I looked already at the weight of my solution. So with a slimmer fan i think it should be only a few grams heavier. The variable airflow suggestion seems pretty hard to come by with: For a complete solution you need support from the complete software stack and you need the hardware capable of controlling the airflow. But well i think it would be possible to build a flap
into my mod with some small modifications. It would be manually controlled for the first test but it would be a approach to test if variabillity makes some sense.

Quote
dmould
I certainly need to improve the cooling on my printer. At present I have removed the filament cooling completely, because I print ABS which does not require cooling for the bulk of the print. The higher melting point of ABS means that it naturally cools and solidifies more rapidly than PLA in still air (rate of cooling is a function of difference in temperature).
It also means that the part is more prone to warping. I experienced bad warping with the original Ormerod arrangement, if I were to fit your design I suspect that my prints would be splitting apart before they were 3mm thick! I do however need cooling on some parts of some prints. At present I use various workarounds - to print your house I would print more than one at a time for example, and I will often split parts into sections and glue them together after they have printed so that I can avoid having bridges or overhangs in any section.

I have not printed ABS. So i have no experience with that. But my experience with PLA showed me that the cooling from one side is far from optimal and it also cools some parts to much
and others farer away from the venting outlets cool slower so that you get worse warping than with cooling from different sides. Also note that i only have to venting outlets not directly pointed at the hotend which should give the filament some time to bind to the other layers before hit by the airstream. My hope in the arrangement of the two vents was also that by creating a whirl about the hotend that there is a nice cirular temperature gradient for smooth cooling.

I know that get perfect results if i print the house above six times that i get pretty good (not to say perfect :-) results. In fact the house in the background is printed in a batch of six and every house came out at least good some nearly perfect (with the 0.5 nozzle i have). But i think printing only one house shows the problems of the stock solution pretty well.

Thinking again for the variable cooling stuff: I think using two small fans is a worse solution than using a lighter bigger fan which has better airflow efficency than a small fan and nearly the same weight. So i think using some flaps to control the venting is a better aproach. This flap could be controlled by a sub micro servo with about 3.7g (ultra micro 1.7g) which shouldn't be a problem from a weight perspective.

Best regards
Tim

Quote
dmould
Variable cooling is by far the best way to go

IMG_4962.JPG

IMG_4961.JPG

IMG_4959.JPG