Hi, I am Primož from Ljubljana, Slovenia.
My knowledge of German is on a level of 2 year old German kid. Therefore I will rather write in English (I hope this is not against forum rules). I often look at this forum, but because I don't understand much and google translate does a very poor job with translation in English, I mostly just look at pictures.
I will post here some pictures with my modifications of my printer. You must understand, that most of these mods were meant to be only temporary solution and have some flaws (mostly they are hard to assemble/set up). But since they work, I never made improved versions.

First: Filament tension mechanism
I changed it, because original was a bit flimsy and I thought this is the source of filament grinding.


Second: Extruder drive gear
As the previous mod didn't change filament grinding, I replaced drive gear with MK8 gear.


Third: Extruder fan
I realised that problem is in extruder. The part that is called cold end wasn't cold. So I mounted a fan on upper part of extruder. Fan is powered through external AC/DC converter, currently at 7,5V. If I put it at full power (12V), then it cools too much.


Fourth: Surround nozzle for part cooling fan
One side of my printed parts was not as good quality as the other side. Usually it was curling upwards.
(extruder fan and surround nozzle went trough different versions, sometimes separate, sometimes combined in one unit. Pictures are of current version )


Fifth: Z endstop
Z endstop had broken and it was replaced. Because I was afraid that it will break again I designed different mechanism. Instead of bending the PCB of endstop, a coil spring provides the force to press endstop and allows for displacement during printing.




Sixth: Re-gluing silicone heater to ceramic bed
I noticed that bed doesn't get as hot as before, and then I saw that heater is sagging in the middle. I re-glued it with silicon resistant to 300°C (silicone for engine gaskets, I had it at hand and it had a little better spread-ability than other high-temp silicone that I know of)

Seventh: Enclosure
I put it in old wardrobe in a garage. I insulated a wardrobe from inside with 30mm Styrofoam. I also have 100W incandescent light bulb, that provides light and also heat in cold days .


I hope this might help some of you, or inspire you to make better mods than mine. If you have any questions, ask in English if you can or in very simple German, and I will try to answer.

If i may suggest something? I made very good experiences with not using the Filament guide from the side of the printer, instead from somewhere above or aside above., Then you won't need that long PTFE Tube and get less friction.

Hi Primoz

I like your Surround nozzle part cooling fan system and this is one part that I have realize in the future.
Do have no problems with the higher temperature expant values of Z-holder (before aluminium now ABS?).

Best regards Frank

@Wessix
I am using stock tube with 6mm hole diameter, but I inserted short part of PTFE tube with hole diameter of 3,175mm. This centers the filament and prevents it from grinding on the edge if the 4mm hole in aluminium part. Otherwise I am quite happy with this tube and I don't have any problems because of that.
Some other designs of filament guides have a problem, that the length of the filament between guide and extruder changes during printing. This means that extruder is sometimes pulling on the filament and sometimes pushing it back. In my theory this alternating force can lead to more grinding/slipping of the filament than (more or less) constant pulling force through a tube.


@RFrank
Thanks. Maybe you should consider using an radial fan (blower). I think it would work better and would also be easier to make. I used 0,2 or 0,3mm aluminium and glued it together with the same high temp. silicone adhesive as I used with heated bed. That's why it is all black and ugly. (here is a picture of crossection).


It is actually PLA. I planed to print it again in ABS, but I never did, because it works as it is. I haven't noticed any problems because of higher temp. expansion.
Here is a bit better picture. The blue part is a lever that pivots around a screw.The spring must be preloaded (I used some spring from a pen). But it was very hard to assemble, as you have to screw the yellow part to the plate first and then you can insert blue lever. The hardest thing is to insert a spring and then the washer and nut.

Hi,

I used a different switch with a spring giving about 2 mm security, so my construction is pretty much easier to do. But thats not what i really want to say.. Having very much trouble with ABS I decided to follow your suggestion and bought a MK8 gear. The effective diameter of this gear is one mm smaller than the RF1000 one. How did you correct this ? Its simply the multiplier or did you change the firmware (thats no problem for me, i simply would like to know).

PeterKa

The one that I bought had a diameter around 7mm (measured with calipers over the teeth). At the time I wasn't comfortable with editing the firmware, so I used extruder multiplier. Currently is 1.25 (varies with different filaments from 1.22 to 1.27).
Even though I was changing the firmware later on (I lowered homing speeds and lowered stepper driver current) I never changed that. So I still use extruder multiplier.
I am sorry if I mislead you, maybe I didn't wrote clearly enough, that changing the gear didn't help much. The biggest difference for me was mounting the fan for cooling the cold end.


And I forgot to write what problem I had: Printer was printing fine for 30 to 45 minutes. After that filament started grinding and extruder was just spitting small blobs. The fan solved that.

Hello Primoz,

Primoz hat geschrieben: And I forgot to write what problem I had: Printer was printing fine for 30 to 45 minutes. After that filament started grinding and extruder was just spitting small blobs. The fan solved that.

I guess, several other Printers have this problems too. Mine included. I did not have many problems with PLA but with other Filaments.

{There have been some Threads which kind of describe describe similar Effects:
Druck mit Bendlay -> Extruder bei den Motoren aktiv kühlen http://www.rf1000.de/viewtopic.php?f=72&t=1470
RF2000 60mm Lüfterhalter - Modelle zum nachdrucken http://www.rf1000.de/viewtopic.php?f=64 ... hilit=PTFE
Teflon Röhrchen? Mein V2 Hottend bringt es nicht mehr! http://www.rf1000.de/viewtopic.php?f=72 ... TFE#p15100
Voltivo ExelFil ABS 3mm red - meine Nerven! http://www.rf1000.de/viewtopic.php?f=8& ... TFE#p15366}

I cool my motors and the coldend since Summer. But until now I never could spot the origin of this problem 100% sure. We got some hints, I got a solution, but the reason of this is still a bit a theory.
I once made little pinion-coolers, but dropped them, because the effect was not clearly visible.
rf1k_mjh11 then told me that it might have something to do with weakening filament because of a too hot PTFE-inlet within the Hotend.
AtlonXP stated that inside of his hotend the PTFE-piece was a bit to short - which could, in my eyes, lead to more friction between filament and the hotends inside.
druckttoll had tested the airflow and did not see a major improvement because of airflow to the hotends middle part.
Others said that they do not need this cooling fan, don't have this problem and therefore decline this additional airflow.
More problems are reported if the surrounding temperature is high / in summer.
I think the solution is a mix containing better pinion, better springs and cooling.

It is rather interesting that you solved this failures too. My failed parts looked a bit like yours!

Do you have your own plausible theory about this behaviour?

LG

Hi,

I dont feel missleaded I never liked the RF1000 gear. The simple one in my Makerbot finally works much better. So a change was neccessary anyway and the costs are ridiculous (3€), only thing was that i had to reprint the lever arm for filament pressure. This is done in between.

But you helped me with the decission to equip a fan duct. I simply replace the 25x25 fan (which accidently is broken atm) by a 40x40 and use the power supplyed for the small fan (which is regulated and can be modified by GCODE). The airconduct is divided for cold end and material. So its not too much stuff and it can be printed. There are several threads concerning this issue. But as usual, i want to fix it by my own I will show my solution when ready.

PeterKa

According to this: http://reprap.org/wiki/Hot_End_Design_Theory
You need short length where plastic is molten (temperature above melting point), so it can be extruded and the rest of the plastic must be solid, so it can transmit force from drive gear (temperature below glass transition temperature). What you don't want is plastic with temperature between glass transition temp. and melting point. Plastic in this state can not be extruded and also doesn't transmit the force. It also expands and could also buckle, therefore can grip on the walls.

Ideal extruder would have only hot end and cold end. Of course this cannot exist because of heat transfer and heat from hot zone transfers to cold zone.
Some designs have insulator like PTFE and PEEK and others like full metal hotends (E3d) use a very thin neck between hot zone and cold zone (smaller the crossection area, smaller the heat transfer ). Because it is metal it still conducts a lot of heat and you need fins and a fan. That part between cold and hot end is called heat break.

And this heat break is bad at rf1000 extruder. The heat can more or less freely travel between cold and hot end. Without fan heat creeps into cold zone. With fan on full power (12V) It removes too much heat from hot end (I use Sunon 25mm fan that was on printer for cooling parts) and plastic doesn't get melted enough. So now I have a fan at 7,5V.

I actually got extruder replaced by Conrad, but the second one worked as bad as the first one. So I thought it is a bad design of extruder. But considering that it works for some people it could also be a problem with manufacturing.

Hi,

here a simple preview drawing of the air duct. Shown are the walls (green), left and right side are still missing (under construction). But i think its obvious how it is intended to work. Its quite simple, the duct can be printed and only one fan is needed.
PeterKa