My Geeetech Rostock 301 Lessons learned

Please Login to Comment

This is a collection of my observations and hacks for the Geeetech Rostock 301 printer kit

I finally replaced my Mark8 extruders with geared extruders. Major mod, had to drill new holes in top plate to fit 3 3d printed geared/belt extruders and then increase the height of the filament spool supports to clear the much bigger extruders. But...WHAT A DIFFERENCE! The extra torque is just whst this printer needs. Prints are much much better. The thing is, you cannot print fast with this thing no matter what you do so don’t try. With this geared extruder the recommended steps per mm of filament is around 800. But this high a value causes the geeetech to freak out! So, i set it to 200 steps per mm and then tell repetier to multiply the filament factor by 4, yes, 4. Totally not kosher but works. Also i print at 7mm/s. Yes, patience is a virtue. But prints are great!

I have now wasted weeks of my time in trying to get get one single decent print out of the Rostock 301. No success, whatsoever. I even changed to metal extruders, Wade extruder, E3d V6 hotend, increased voltage for stepper motors, cooling fans, and numerous other "tips and tricks". This printer is just not capable in doing its job. My other printers (Flashforge Creator Pro and Tevo Little Monster) are producing parts as expected, or even better. This Geeetech thingy does not. I will give it away to the Fablab of my university and I hope one kid will be able to make something useful out of it. I am fed up with it.

Here is my Bowden tube cutting jig made from a block of wood. I use this to make sure my Bowden tube ends are square. Critical for this printer. https://jumpshare.com/v/LuACKxeXfFpSLyl4bQyI?b=0

Why don't you print that, instead of carving it out of wood?

Thats a good idea, i was just laxy and had a block of wood and a 4mm drill bit but i will create an stl model yo print.

My checklist before each print:
1) boot up computer
2) start repetier host s/w
3) click on repetier connect
4) go to manual tab and click on home to home printer
5) use z axis control to take carriage down to 195mm in z. I do this because the printer acts weird processing commands when home switches are in contact (bug?)
6) put either painters tape or kapton tape on the bed print area. Spray with hair spray.
7) turn on hotend and bed heaters to temps appropriate for material. Recall i leave the printer in a state where filament is retracted 100mm from the hot end. Also stsrt slicing your object or load gcode to print
8) ensure bowden tubes are clear, sometimes they get clogged right at the ends in the hot end. Sometimes i need to unscrew the brass pneumatic tube fittings because i cannot pull the tube out even depressing the release on the fitting.
9) manually push the filament to the point of resistance
10) extract 10mm lengths of each extruder until you see the filament flowing out nicely
11) position the cariage to z=0 and check the bed leveling, adjust as necessary so a piece of ordinary copy paper slides with some resistance.
12 start your print. And cross your fingers...

ABS is a pain with this printer, so far 0 successful prints with ABS. PLA is pretty reliable, with all the known issues like color mixing etc.

This whole printer is a PAIN, but problems with ABS are mostly due to incorrect bed temperature. The sensor is located underneath the heater and actually measures the temperature of the heating wire that is also facing down. The hotbed is made of aluminum and is cooling very fast. So when the sensor shows 120 degrees C it is actually about 65 at most, which is not enough for ABS.

You can move the sensor away from the heater using a piece of thermal insulator like cork, so that the sensor measures the temperature of the air between the printer base and the heater. That way it will show temperatures more resembling those of the hotbed surface and the firmware will heat the bed appropriately.

Moving the bed sensor as you suggest is an awesome tip! Thank you!!!! This will let me print successfully with the abs filament i have. I will also havevto get a higher smoke point oil for bowden tube lubrication. Reall sppreciate your advice!

I forgot to mention that you will have to adjust the P-I-D regulator values for bed temperature after you move the sensor.
I used Ziegler-Nichols' method to do the adjustments.

Excellent. I will look that up. When you moved the sensor did the temperature control exhibit "ringing" or did it have difficulty getting to temp?

I don't understand the term "ringing", but if you have any problems with the temperature after moving the sensors, that is most probably because of wrong setting for the PID regulator. It was set up for the original location and parameters for the new location differ dramatically. You may see the temperature oscillating around the desired value or not reaching it at all. All those are the symptoms of a misconfigured PID regulator.

To print ABS I use Kapton tape snd hairsprsy. To prevent shrinkage and corners pulling away I configure a brim of 6 mm. Still have not made ot all the way through an ABS print but getting close.

Here is the link to the pictures of the actual print of that Heart Bracelet:

It shows the stringing and then after cleanup. I give it a C+, maybe a B-...

I bought a Geeetech Rostock 301 in February 2017. Here is what I have done to get it running.
1) I had to replace the defective mixing hotend. It would not extrude for anything. I bought a couple extra replacements (around $40.00 each) and replaced the one that came with the printer. Before that, I had ordered just the brass part but even though it seemed to work, I had to drill out the thermocouple hole and the threads were never quite right for the finned filament tubes and it leaked, so I just replaced the whole thing. I made sure I tightened everything when hot.
2) I replaced the extruders with metal ones I got from Banggood.com per a post I had seen. That was a good move.
3) I set the reference voltages on the stepper drivers to 1.09V for the extruders (they are bigger motors) and 0.85V for the axes steppers.
4) I put some open cell foam with several drops of olive oil in a small piece of 1" PVC pipe that I make the filament run through (sits on top of the input to the extruder) as a lubricator and a dust filter. This made a HUGE difference in performance. I have seen no adverse effects from the small amount of oil but I did see small drips of oil once when I started the printer up and started heating the extruder. Still no bad stuff happened. Olive oil because it has a high smoke point.
5) Of course, I level the bed and I also adjusted the radius settings to get rid of the "concavity" as when leveling I noticed that the center was looser (piece of paper slid more easily) than at the extreme limits of X and Y. I level at temperature (205 to 210 PLA hotend and 55 bed). There are some good internet references for how to do this if you search for concavity and convexity delta printer adjustment.
6) one of my diagonal rod sets had some play in it at the U-joint bearings so I put a rubber band around them to keep them from rattling around during moves.
7) I put 8mm inside diameter, 1 inch or so outside diameter ball bearings on the rod for the filament so there is no resistance to pulling. That and the lubrication and dust filtering I think has made the most difference. To keep the ball bearings separated properly to support the spools, I used small lengths of heat shrink tubing between the ball bearings, the tubing is a little larger than the 8mm support rod.
8) I tried printing at 20mm/sec and it seems to work but I hear the extruders clicking. So I will go back to 10 - 15 mm/sec. I print at 205-210c with a 55c bed. I use blue painter's tape on the bed and use hairspray to prevent the print from dislodging and giving me a "birds nest".
9) I set the retraction to 5mm / 40mm/sec but I still have stringing so I am continuing to work on that.
10) I put some custom gcode to run after the print is finished to set the temperature to 170c, go home, make some very slow (10mm/sec) moves up and down from Z=210 to 220 and back to waste time while the hot end cools down (Geeetech firmware does not understand a G code that would wait for the temp to go DOWN to 170), then I retract the threee extruders 100mm each. Then the heaters all shut down but I keep the steppers enabled, otherwise the carriage slides down and damages the print.

With the above I was able to print a 3X rendition of the "Heart Bracelet" example that Geeetech provides a link for in their user manual. It comes out reasonably well, but I have to do clean up of the stringing and there is some bleed of color as it switches extruders. I used Slic3r and I have heard people like Cura better so maybe I will try that sometime.

Hope this helps!

I did almost all the same except that for 1) I forced the seller to send me a replacement. The stock diamond hotend didn't print becase it had a 1.4mm diameter for the filament instead of 2.0mm as per RepRap's Diamond Hotend drawing. Even their push rod wouldn't go in the stock hotend for me.

I haven't yet tried lubricating the filament/tubes. Will give it a try. Anyway 15mm/sec speed is waaaay below the expectations for a delta printer that are said to be known for high speeds. Besides, color mixing is actually impossible, so I'm thinking about giving up on this printer and selling it or salvaging it for parts.

Concur, it is a dissapointment. When i print a mix each color "prefers" its side of the nozzle. The effect is interesting but not not what i wanted. I was ablevto print st 20mm/s with the lubrication.

Beside bad mixing, the 3-in-1-out hotend also adds a problem that when you print in single color the unused filaments may get overheated and the pressure of the filament in use will most surely clog the unused filaments' tubes. All in all, it is just unusable. One of my thoughts about this printer (in addition to selling or salvaging it for parts) is replacing the hotend with a single color one. Selling it is the preferred way to go though. I don't want to waste more time on it, I want to print.

Great summary, Chris, thanks a lot. I am still waiting for my first successful print, but I need to take care of some other stuff right now.
When I get back to the printer, I would like to try your custom gcode, as your ideas sound very good. Would you mind sharing that?

Thank you,

Here is the ending gcode...

M104 T0 S170 ; lower temp to 170
G28 ; home all axes
G91 ; set to relative move mode
G1 Z-15 F10; move down 15mm slowly
G91 ; set to relative move mode
G1 Z10 F10;
G91 ; set to relative move mode
G1 Z-10 F10;
G91 ; set to relative move mode
G1 Z10 F10; move up 10mm slowly...
G91 ; set to relative move mode
G1 Z-10 F10;
G91 ; set to relative move mode
G1 Z10 F10;
G91 ; set to relative move mode
G1 Z-10 F10;
G91 ; set to relative move mode
G1 Z10 F10;
G91 ; set to relative move mode
G1 Z-10 F10;
G91 ; set to relative move mode
G1 Z10 F10;
G91 ; set to relative move mode
G1 Z-10 F10;
G91 ; set to relative move mode
G1 Z10 F10;
T0 ; select extruder 0
G1 E-50 F1000 ; retract 50mm
G91 ; set to relative move mode
T1 ; select extruder 1
G1 E-50 F1000 ; retract 50mm
G91 ; set to relative move mode
T2 ; select extruder 2
G1 E-50 F1000 ;retract 50mm
M104 T0 S0 ; shut off hot end
G90 ; back to absolute move mode
G28 ; go home

Thanks a lot Chris! I'll let you know how it works,
when I get to the printer.


Here are the things i had to do to get the Geeetech Rostock 301 to work. It is not great, the primary issues are:
-Slow printing speed
-stringing during moves
-mixing is not really mixed, colors prefer the side of the nozzle they are in.
-switching colors (that heart bracelet exampl) smears so the color switching is not super clean.

Now here is what i had to do to get it to print:
-replaced the defective hot end (about $40.00)
-replaced all 3 plastic extruders with metal ones from banggood (post online had links to which ones I think another $40)
-Made foam filament filters that the filament passes through before going into extruder, they are oiled with olive oil and remove dust. (This made a HUGE difference)
-Leveled bed, used online suggestions for firmware settings for eliminating bed "convexity" or "concavity) so sheet of paper just slides with some resistance with z=0 and ranging x and y through extremes, checking various points around.
-Set extruder motor voltage reference to 1.09 volts
-Set carriage motors voltage reference to 0.85 volts
-Put rubber bands on the carriage arms pairs to hold them "together" as thete was some slop in the bearings/universal joints.
-Set retraction to 5mm w 40mm/sec
-Set print speeds to 20 (was using 10 before but with oil and filter filament flows better -we will see...)
-print at 205 degrees C
-use painters blue tape and hair spray for adhesion so part does not come loose and make a "birds nest"
-I added ending gcode in Slicer that leaves motors on so carriage doesn't fall and hit the print after, homes the cartiage, lowers the temperature of the hotend to 170, goes up and down slowly (feed rate 10) to waste time while hot end cools to 170, and retracts the filament 100mm. Then hot end snd bed temps go to 0. Motors stay ensbled until i come and shut down the printer and computer.


Thank you for this, it would have been very helpful for me if I had it before I set up my printer!

I agree with 99.99999% of your observations and I am not trying to be a troll! Here are my observations:

  • Defiantly slow. The printing speed is nowhere near what is claimed by Geeetech. I use speeds right around what you are using.
  • I have not solved stringing very well yet either. I think it is a consequence of long Bowden tubes. If anyone gets this to work well on this printer, I would love to hear it.
  • About the mixing, you are right, the colors do not truly mix but most of the women I have shown these prints to thought they were cool! I would prefer a true color mixing, however and is a big part of the reason I selected this printer.
  • I have only done a little bit of color switching. One of the slicers allows you to build a "smear tower" or some verbiage close to that. I have only briefly experimented with it, but it seemed to help, but it is still not acceptable.

Compared to you on what I had to do to get it to print:

  • My hotend seems to be OK, glad I didn't have your issue.
  • Replace extruders with metal ones- yes, my printer would not work worth poop before I did this.
  • foam filament filters with olive oil. Ummm, I have not had to try that yet. I may try at some point but I am pretty OK with the prints I am getting now.
  • Bed leveling, now there is where a novel could be written. The biggest frustration was leveling the bed "perfectly" at all three towers and in the center. So far, so good and not all that difficult. The problem comes when you try to print anything "between" the towers toward the outer edge of the printer. Printing directly from a tower (i.e. X, Y, or Z) to the center is wonderful, but if you look at the geometry equations, when trying to print directly away from any tower (and between the other two towers), the Z axis resolution gets progressively worse the further past the center and directly away from any tower you go. For example printing away from the X tower to the center gives very good resolution, but the Z axis error as you move past the center toward the edge of the print bed between the Y and Z towers gets unacceptably bad. This COULD be somewhat corrected in firmware, but this printer does not correct for the reduced resolution, and the print head moves away from the bed between the towers at the out edge of the print bed. This is inherent in the Delta printer design, but if the ratio of arm length to the print head movement in the X-Y plane is large the error becomes tolerable. The arm length to X-Y movement is too low on this printer, and cannot be "tweeked" out by any of the settings. I wasted a lot of time trying to fix the problem by the settings. I now just make sure that my prints do not go out too far past the center of the bed between any pair of towers. To see this problem, just print any calibration print like this one: https://www.thingiverse.com/thing:776346 and if you level the bed per instructions will get a very good lay down of material around each tower, but the Z axis will be too high between each tower for adhesion. Googling "delta Z axis calibration" will bring you more than you ever wanted to know, but confirms that this is a design factor with Delta printers, and I do not think Geeetech is handling it well on this printer.
  • Set extruder and carriage motor voltages up: Yes! Geeetech should do this at the factory! It should not be left to the buyer to search forums to correct their lack of production effort.
  • Put rubber bands on carriage arms: I have not found this to be helpful, your mileage may vary..
  • Retraction- I have not found acceptable settings yet, but I have found a lot of them that don't work well! :-)
  • Print speeds: This printer must be made to print slow to work well. That is pretty OK with me. If the print is 8 vs 12 hours is not a big worry, I just set it up, watch the first couple of layers, then watch on web cams to make sure it isn't making a bird's nest.
  • Painters blue tape, YES! But I use the stick school glue instead of hairspray. I really like the kind of glue (Elmer's makes one) that goes on purple, then goes clear when it dries. That way I can see my coverage when I apply it, and when I go to wash it off of the print, it turns purple again in the water so I can see where I need to clean more. Blue painters tape did not work for me with ABS. I used the school glue and the print adhered well to the tape, but pulled the tape away from the print bed as it cooled on the outer corners!
  • I have found that print head and bed temps vary by color, make, and type of filament. 205C head and 40C bed temp works pretty well for most PLA filaments. I have a borosilicate glass plate that I use for all of my printing now https://www.amazon.com/gp/product/B019D3VZYE/ref=oh_aui_detailpage_o05_s00?ie=UTF8&psc=1
    (including PLA) and just apply the School glue directly to it and it works well. I have been printing A LOT of ABS and I really like it as it is not as brittle as PLA. I have been using 250C head and 100bed temps for ABS. I am experimenting with lowering the bed temp as it tends to overflatten the first layer if too hot, but if too cool, the outside corners tend to lift. I also tried the wood filiments , and it worked OK, but if left in the head it clogged it when the head cooled.
  • I just set my end Gcode to turn off the lights (I added them) and heaters and park the head at max Z and at the back of the bed (and leave the motors engaged to the head doesn't crash into the print) so it will not ooze the last "snotite" on the print. What is your thinking on moving up and down while cooling? Seems like it would just cause wear on stuff for no big gain? Not trying to troll, but trying to understand, why just not move the head away from the print and let it cool naturally?

Again, thanks for your thoughts, and I hope I have added and supported yours!

Delta Calibration & Validation Tool

Not trolling, really good info that is appreciated. I used the foam olive oil filters becsuse the printer is in a dusty garage and i had to unclog the hotend every time and many times clogs caused print failure. You are right about the up and down wear from my end code but there is no "wait for temp to fall to X" and i needed to cool down a bit them do a big retraction to pull the filament back up the tubes. The area right around the tube to hotend joint is susceptible to jams when i let the filament cool down while pushed all the wsy in. Letting the filament cool a little ensured that the plastic was pulled out of the hotend. When i did not have the motion to waste time the program would end and the filament would be pulled back while still liquid. Thanks again for your comments!

Ah, a dusty garage! I didn't think of that. The olive oil wipe is actually a good idea and I have seen that type of wipe of clamp on Thingieverse since, but your post was the first time I had heard of it.

Yeah I cannot take credit for the dust wipe and lube but like to pass on the tip. I have since upped the retraction to 8mm and the stringing is getting better. I was able to make a fully functional geared extruder i got off of thingiverse. I probably won't use it but just wanted to see if i could make a functional mechsnical part.