Default Rostock

[Polygonhell]

I just set the baud rate to 57600 in the device manager and in repetier host. It didn't change anything. Is there some way to set the baud rate in arduino?
EDIT:

More bad news for me. After I changed the baud rate, I can't communicate with the rambo anymore through repetier host. I changed it back to what I had before, but it isn't working. IS there any way I can just completely reset the rambo board?

The baud rate in repetier must match the boards rate compiled into the firmware, and has nothing to do with the setting in device manager.
You probably need to set repetier host to either 115200 or 250000 baud.

[Generic Default]

It just started working again. In the repetier host manual tab, there were 6 commands that were waiting to be sent to the printer. The thing just started moving on its own and now it finished the commands and is idle.

To update the firmware, should I have the rostock turned on or off?

[Polygonhell]

It just started working again. In the repetier host manual tab, there were 6 commands that were waiting to be sent to the printer. The thing just started moving on its own and now it finished the commands and is idle.

To update the firmware, should I have the rostock turned on or off?

Should make no difference, you just need to have 5V going to the logic, either via USB or via the 12V on the board.
You cannot upload to the board while repetier or any other program is connected though.

[Generic Default]

Nothing seems to be working, and I can't calibrate my machine until I can upload firmware to it. Would anyone else be willing to try downloading the latest version of arduino and uploading the firmware to their rambo just to see if you get the same error as me?

I hope I didn't get a defective rambo board.



-
Also, I'm getting excellent results with trimmer line. Anyone who wants to experiment with it on a well calibrated machine should know that it makes indestructible parts. I just printed a 2 inch long trigger-like piece and I can't even break it by torquing the ends as hard as I can with pliers.

[cambo3d]

quite possible, seems a couple people here have been experiencing this.. timeout error lately.

[chosenken]

I seem to be facing this same issue as well. Just got (almost) every thing done (don't have the extruder yet or the hot end completed, broke the thermometer...) and I am trying to flash the firmware and see if I can manually move the arms.

I have the Arduino 1.0.5 software and the latest Repetier firmware. When I try to upload the firm ware in Arduino I get the following messages:

Code: Select all

Binary sketch size: 101,612 bytes (of a 258,048 byte maximum)
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_ReceiveMessage(): timeout
avrdude: stk500v2_getsync(): timeout communicating with programmer

I am able to connect to it in Repetier host when I set the baud rate to 250000. I see that it is on RepetierMAX 0.80, so maybe I don't have to update. Though it is continually in Dry Run mode. Maybe since I don't have the Hot End temp set up yet.

I should note that I am trying this with the power on and from a Mac.

[Generic Default]

Looks like we're getting the exact same error. You should be able to move the arms and stuff through repetier host. I haven't actually flashed the firmware yet but I can control my rostock.

[chosenken]

I can't make my printer move in Repetier host yet. Every time I send a command to it, it replies it is in dry run mode. I wonder if this is because I don't have the hot end completed yet, so it isn't getting a temp reading for the hot end.

EDIT

After fixing the pins on the motor connectors, its alive! First time it has moved! And only took 10 days! Haha, was slow building this week, kept breaking things.....

Haven't tried flashing the new firmware yet. I moved to Windows and it seems to work "better" at the moment. I'll try tomorrow when its not 3:00 AM.

[Generic Default]

I just tested glue stick on glass to print with the nylon. It works great, especially with a raft. I made a 1 inch ball bearing with 9 airsoft BBs inside. The thinnest part of the outer bearing race was only 1 thread thick (0.5 mm, light translucent) and it spun nicely until about 30,000 RPM, at which point it blew up in my face and sent BB's flying in a radial pattern. So make sure your races are a tiny bit thicker when printing ball bearings. Other than that, it worked great. It's motivating to know that my printer isn't even calibrated yet but it's printing functional mechanical parts.

[Generic Default]

The problem with not being able to upload firmware to the rambo was that seemecnc messed up the bootloader when they programmed the board. So I sent it back and they fixed it, and now two weeks later I'm printing again. I adjusted the delta radius in the firmware so that my rostock stays at the correct Z height across the bed. Now I'm trying to make the magnetic arm parts so I can get decent print quality. I printed the ball cups and they failed with terrible tolerances and huge seam-lines from badly made gcode. I've tried slicer and cura so far.

Also, every time I try to open the EEPROM menu in repetier host, the EEPROM table starts filling from top down, then repetier crashes. So I can't really adjust any EEPROM settings. Occasionally it does work, but even then the rows and columns are mixed up and some values are missing.

1) Which slicing program is best for cylindrical objects?
2) How do I fix the EEPROM?


Thanks!

[Broose]

Which slicing program is best for cylindrical objects?

I ended up using Kisslicer for the ball ends. It starts each layer at a different point on the cylinder (I think this is because of wipe), so it averages out the errors and the surface is a little rougher but you don't have those seam lines. Here's a couple I made early on with the two programs

[Generic Default]

I still can't figure out how to rotate objects in pitch or roll in kisslicer, so I've been using Cura and Slic3r for everything so far.

Slic3r just plain sucks; every gcode path it has generated so far, regardless of the settings I put in, have WAY to many random retracts. It retracts the filament, lifts up, moves to random points around the print while oozing nylon, then sets back down. I printed 24 of the ball cups for the magnetic arms and not a single one held tolerances of less than 0.05 inches. It doesn't make the correct wall thicknesses on thin wall prints, and it travels paths in the least logical order possible. The wall thickness should depend on the nozzle diameter setting I set, but I'm getting +- 0.3mm thickness variation between prints without changing settings.

Cura doesn't hold tolerances either, and it doesn't retract at all. It does make smooth enough surfaces to print some semi-decent ball cups for my magnetic arms. I had to print each ball cup individually as separate gcodes just to get useable pieces, and even then I had to dremel each cup for a few minutes just to get them to work.

So as of now I have magnetic arms on the rostock (I don't know their exact length and I have no way of accurately measuring), and I also have cork sound reducers on each motor.
I think one of my main problems is that I can't figure out retraction. I'm currently using a 0.35mm nozzle with the stock seemecnc hotend, and retraction just doesn't seem to work. I think 3-5mm retraction is normal for ABS and PLA, and nylon should be around 6mm. But I've tried everything up to 12mm retraction and I still get serious oozing with slic3r. It seems really inconsistent.

I ordered my rostock on May 16th, now it's July 3rd and I still don't have decent prints. It is normal for it to take this long?

[jesse]

In my testing, Cura 12.12A produces the best g-code. It is possible for Cura to do retracts.

[Generic Default]

I calibrated my rostock extensively for the last few days, and now I'm getting better results. I got the retracts in slicer to work (at least for now) by turning off Z lift and not staying in perimeters for retract moves. I made a few good prints, but one slight problem I'm having now is that the bed is calibrated at near the aluminum extrusion towers and at the center, but near the edge of the build plate between the towers it lifts just a little bit. I'm guessing it can be fixed with the settings in the firmware. I don't think it's the delta radius option because the nozzle stays flat near the towers and in the middle. Are there any other parameters that would make it curve up between the towers?

[Broose]

one slight problem I'm having now is that the bed is calibrated at near the aluminum extrusion towers and at the center, but near the edge of the build plate between the towers it lifts just a little bit. I'm guessing it can be fixed with the settings in the firmware. I don't think it's the delta radius option because the nozzle stays flat near the towers and in the middle. Are there any other parameters that would make it curve up between the towers?

It sounds like your work surface isn't completely flat. Do you have a glass plate on top of the Onyx? How thick is it? I had some trouble with flatness until I got the Borosilicate glass plate that SeemeCNC sells, and that seems to hold the surface flat, even with the warping of the Onyx. I also have a 1/8 inch aluminum plate as a heat spreader sandwiched between the Onyx and the glass.

If you do have a flat surface, and if the points you are measuring between the towers are further away from the center than the calibration points near the towers, then any delta radius errors you have remaining would be magnified out there and you may need to tweak it slightly.

[Generic Default]

I do have the borosilicate glass plate, and I think I reduced the uneven surface problem, at least enough to be able to print some larger parts. I switched back to my 0.5mm nozzle for faster prints but now I'm having problems with tolerances again. The OD of a cylindrical part I'm trying to make is 1.125 inches, and an eccentric hole in it has an ID of 0.442 inches. But after printing it, the OD is 1.10 inches and the ID is 0.402 inches. So the outside is too small, and the inside is also too small. I sliced it with Cura and used 0.56mm as the nozzle size. The problem is that it increasing the nozzle size should make the outside and inside larger, not smaller.

Has anyone else had this problem? I have magnetic arms, cork dampers, and a 3 inch computer fan to cool the rambo. I doubt the latter two would mess with the tolerances. I have my magnetic arm diagonal rod lengths set in the firmware.

#define END_EFFECTOR_HORIZONTAL_OFFSET 41.73
#define CARRIAGE_HORIZONTAL_OFFSET 35
#define PRINTER_RADIUS 199.25
#define DELTA_DIAGONAL_ROD 274 // mm

I just changed the diagonal rod length from 274.5 to 274 because my magnetic arms are 283.515 from ball end to ball end. Subtract 0.375*25.4 and I get 274 for the arm length. Maybe that half a milimeter was messing up my horizontal scaling factor....

The only other thing that I'm confused about in the firmware is the #define CARRIAGE_HORIZONTAL_OFFSET 35

What exactly does that do, and how do I measure it? I have the solidworks files from my magnetic arms and a 12 inch digital caliper to measure.

By the way, thanks to everyone here who has helped me through all this stuff! I think I'm at the point where I can start printing really nice mechanical things, I just have to get these last few problems fixed.

[Broose]

You should calibrate DELTA_DIAGONAL_ROD if your parts are undersized on X and Y axes. One way to calibrate is by printing a good size part like 100mm X 100mm and measuring the actual vs. programmed dimension. That also takes into account part shrinkage for the filament material used. It sounds like with the actual vs. desired dimensions you've given above, you should decrease DELTA_DIAGONAL_ROD.

I just used some 12 inch calipers and measured the distance from the nozzle to the Z tower with two different Y locations. If the actual dimension is too large, then increase DELTA_DIAGONAL_ROD. If it is too small, then decrease it, reload firmware, and test again. After you have that calibrated, I would recheck DELTA_RADIUS to see if you have to adjust the offset.

Let me try to explain the CARRIAGE_HORIZONTAL_OFFSET. If you draw a line from the centers of two adjacent balls on the effector that both connect through the rods to the same cheapskate, the perpendicular bisector of that line goes through the center line of the effector (a vertical line basically through the center of the hot end if you've got a single extruder). The distance from the centerline to the intersection point is the CARRIAGE_HORIZONTAL_OFFSET. Here is a diagram which is not so clear for this dimension (http://forum.seemecnc.com/viewtopic.php ... ram#p10025). In any case, CARRIAGE_HORIZONTAL_OFFSET is only used to calculate DELTA_RADIUS, so if you have DELTA_RADIUS calibrated, I wouldn't worry too much about it.

[Generic Default]

Thanks for explaining that Broose!

Unfortunately my PEEK tube just broke when I tried to unscrew the 0.5mm nozzle to switch back to the 0.35mm nozzle. I didn't realize how terribly designed the seemecnc hotend is until now, but I guess it's too late. The PEEK couldn't even take the torque of its own threads, and I was being gentle! I always printed at 240 degrees, so I don't think that weakened it. I thought that if I ever broke my hotend, it would be because of a dumb mistake like smashing it into the glass plate, not normal use...

I ordered my rostock May 16th and I still don't have a working, calibrated printer. I've spent at least a hundred hours over the last month and a half building, fixing, calibrating, ect, and nothing to show for it. And I've put ~1100 dollars into it (will probably be over ~1200 by the time I have it working again). At this point I'm starting to regret choosing the rostock.

I'll probably get the E3D all metal hotend, but I'll have to get a replacement PEEK tube so I can print a mount for the E3D. If there are any problems with the E3D or if there are better metal hotends, please let me know!

EDIT:

I just noticed that the PEEK tube is glossy black on the inside and the threads are broken off so that the bottom part is stuck inside the lower aluminum piece. I hope I can get it out...

[Broose]

I'll probably get the E3D all metal hotend, but I'll have to get a replacement PEEK tube so I can print a mount for the E3D. If there are any problems with the E3D or if there are better metal hotends, please let me know!

If you are running the hot end at 240 all the time (for nylon, I presume) that is at the upper end of the operating temp range for PEEK and I wouldn't be surprised if it weakened it, causing your failure. The E3d all metal hotend is completely awesome for ABS and Nylon. I've had some problems printing PLA with it. Mine stalled with the EZStruder and I had to go back to a geared extruder to get it to work with PLA. IMO the SeemeCNC hotend or a J-Head would probably work better for PLA than an all-metal hotend.

[Generic Default]

13 month thread bump, maybe a record for the SeeMeCNC forums?

Anyway, I think my Rostock is pretty good at this point. I'm getting tiny, detailed, high quality prints like I originally wanted. I didn't want to leave my build thread unfinished on a bad note, so I'm updating it.

That's the most recent picture I have of it. I upgraded a lot of stuff on this printer;

-PTFE magnetic brass tube arms
-Tri hotend for triple extrusion
-Top mounted Steve's extruder
-Top wiring
-Smoothieboard

Newer Rostocks are probably good to go from the start, and I highly recommend anyone reading this thread who is looking to get a 3d printer to go with a Rostock Max or Orion. Good machines, good company. Before I got mine, I was in the mindset that the machine you buy is the one you'll be stuck with. Remember that once you have a 3d printer, you can use it to make another 3d printer, or just upgrade itself to perfection!

Happy printing!

-Generic Default