I had a similar problem-- Closer examination revealed the motor was turning, but the hobbed gear wasn't. The set screw was just close enough to be able to rotate the drive gear under light load, but not heavy load. So it would extrude just fine at the higher temp I used for the first layer, but after that, it would start slipping, and not push any filament.

May be completely unrelated, but may be worth checking.

 

I designed a mount for a dial guage for the direct drive carriage that I am planning on using at first. Since I have a way to level the bed, the auto level really isn't need. I might add one later though. However, on my Printrbot, there is no manual bed leveling so the sensor is a must. I'm not as worried about the small Z movements on it since it has a single Z motor. But having two on the C-bot, I worried that all the extra movements may cause any movement errors to accumulate over time and cause one side of the bed to change more than the other. Probably an irrational fear, but still something I was concerned with.

 

Oh cool. You're welcome I 'spose!

Good points. Oh, i'm not suggesting you recommend some bracing structure to all the builds. It just got stuck in my mind when i think adamcooks mentioned his printer has a bit of play in it when he twists the frame. Some sort of support could be a good addition to these monster builds a few people have done. I think i remember someone mention fastening the printer to the wall to combat it as an alternative... I think that's cheating!

And thanks! Very excited to get building. Now if only my T-nuts would clear customs. A 1 month wait is a joke!

 

Well all my printed parts came in today! Unfortunately they are in two different shades of blue and the replicator 2 that they were printed on seems to have a big problem with warping at the corners of the bigger pieces. I'm chatting with the gentleman I paid to get the parts done and I had a question for the rest of you if you would indulge me. The warp is in the z and on the order of 2mm or so at some of the corners of the idlers and the motor mounts. In my mind I'll be able to force the frame square with the aluminum extrusion regardless of whether or not the pieces are a bit out. What worries me more is the carriages which are bowed on the inside face in about 1mm or so. Do I have the clearance or am I going to be rubbing against the extrusion? With the standard build what kind of spacing is there from the extruder carriage to the extrusion?

Well this certainly reinforces the drive to get a c-bot going. Let's hope I can get it working as well as the rest of yous.

Cheers!

 

I'd say nip this in the bud and go back to your printer and get them to sort it out! Especially if you paid them for it

 

I have a self built i3, and am now looking to do something larger with more than 1 extruder, this certainly looks good!

 

Ha, that's awesome Looking at the z18 print, you can see how the antlers get a weird faceting on top: It looks like they ran some sort of decimation on it first for some reason.

 

I think I see a few broken antlers as well. I'm kind of excited about seeing the C-bot 2, there have been some really neat variations since the original. If you update the frame calculater, increase the front to back a touch. I had originally planned on dual extruding with another stepper on the rear of the platform and added 35mm, I should have added 75mm to allow for full coverage and a additional stepper. I did miss the boat on the open-builds extrusion sale. Re-sizing my frame will give me 6 420mm 40x20 extrusions, maybe I use those for the horizontals on a delta, I do have a collections of boards these days.

On the board front, there is a new board designed for Sailfish firmware. It will be produced by Carl Raffle. It would be an excellent candidate for any smaller C-bot, due to the built in allegro 4492's max output of .8A. It would have no issues driving steppers in the 200mm^3 build envelope size. For anyone that doesn't know, one of the major performance increases that sailfish provides is due to the .x3g file format. We use GPX to post process our gcode files into nothing but stepper pulses, eliminating any onboard processing by the 8-bit AVR. This virtually eliminates the pause zit. There are a few more bonus like JKN pressure advance, thermocouple support, and really useful LCD menus.

@AK Eric since moving your linear ways and lead screws apart have you noticed any more binding? The advice i had unanimously been given by Dan,Ryan and Jetguy was to place your ways and screws as close together as possible. I would love to hear your realworld experience with this. They also said that cantilever beds as large as yall's would be recipe for disaster, the impetus for the TripleC, and that appears to have been disproved. Have you printed anything big and off center? When you arrange parts do you center them between your screws and ways?

 

@adamcooks "have you noticed any more binding?" :
Not sure what you're referring to, I didn't have any 'binding' in the past. I mainly did the leadscrew rearrangement to try and keep the build platform from what I visually interpreted as 'bouncing' on larger prints as weight was added to the platform over time. Note, this was just me staring at it and not liking what I saw (the 'bouncing') as it printed, I honestly didn't do an A|B comparison of the same print before and after leadscrew rearrange, nor was I ever that upset with the results from the original design. Like mentioned a few posts back, I just get a weird squeaking now from the ACME block as the platform raises, but everything seems to work a-ok. Would I do it (front\back leadscrew) again? Absolutely. Haven't seen any downsides yet, and the bed seems faaaar more stable now than stock. Triple-leadscrew would probably be the best, but this can't be that far behind IMO.

 

FYI, since I switched to the direct-drive setup, I decided to mount the 1.2mm volcano nozzle on it. Here is my "calibraton Groot"... pretty happy with the 600 micron quality:
Blog: “I am (calibration) Groot”

 

Yep, single rear (big, 24cfm) fan. It's an interesting tradeoff in mm/sec I'm finding with the big volcano nozzles: Big nozzle = more filament = more to heat = harder to heat. I'm running at 30mm/sec, and if I push the fan up over 50%, the nozzle has a hard time keeping temp up, even though I have it surrounded with both ceramic insulation and kapton tape. But if I don't run the fan high, it doesn't cool the fine details fast enough. I tried running it at 45mm/sec, and I couldn't use the fan much past 25% before the temp started dropping. An interesting conundrum...

 

Any ideas??
This print is little strange and I have no idea what happened.
It is obviously an issue with the drive, as the slic3r shows a straight print, but what?
It starts to go wrong about 1/3 up and starts printing at an angle.
A previous print of a wrench was fine but was only about 1/4 the height of this.


front


back


Actually, it is twisting, the back is leaning to the left as is the front.
The part was printed facing forward.
The length center was about in the center of the Y axis.
The the width center was at about 3/5ths X axis.
--> |.. I.|
front
It appears as if the bed rotated CCW as it printed the top 2/3rds - which is can't.

 

here is a pic of the platform. It just cant twist that way. Even if it could I wouldn't see that kind of change!


My jerk settings appear to work up to 100mm/sec - no ringing or blobs.
They r :

#define DEFAULT_AXIS_STEPS_PER_UNIT {200,200,800,299.9}
#define DEFAULT_MAX_FEEDRATE {500, 500, 8, 25} //** (mm/sec)
#define DEFAULT_MAX_ACCELERATION {9000,9000,100,12000} //**

#define DEFAULT_ACCELERATION 5000 // X, Y, Z and E acceleration in mm/s^2 for printing moves
#define DEFAULT_RETRACT_ACCELERATION 12000 // E acceleration in mm/s^2 for retracts
#define DEFAULT_TRAVEL_ACCELERATION 9000 // X, Y, Z acceleration in mm/s^2 for travel (non printing) moves

// The speed change that does not require acceleration (i.e. the software might assume it can be done instantaneously)
#define DEFAULT_XYJERK 20.0 // (mm/sec)
#define DEFAULT_ZJERK 0.4 // (mm/sec)
#define DEFAULT_EJERK 5.0 // (mm/sec)

I am also running @ 40mm/sec or less. What do u recommend?

I never saw this before. I did tweak jerk settings, but have not seen this issue since I changed it.
It did improve blobs and ringing.

What could be different about this print? This is the first one with this length and height printed in ABS since I changed it. The rest were PLA.

 

Actually, the defaults are much lower now. They were higher in the past.
these r current RC 2 settings

#define DEFAULT_MAX_ACCELERATION {3000,3000,100,10000} // X, Y, Z, E maximum start speed for accelerated moves. E default values are good for Skeinforge 40+, for older versions raise them a lot.

#define DEFAULT_ACCELERATION 3000 // X, Y, Z and E acceleration in mm/s^2 for printing moves
#define DEFAULT_RETRACT_ACCELERATION 3000 // E acceleration in mm/s^2 for retracts
#define DEFAULT_TRAVEL_ACCELERATION 3000 // X, Y, Z acceleration in mm/s^2 for travel (non printing) moves

these r old ones from release 1.1
#define DEFAULT_MAX_ACCELERATION {9000,9000,100,10000} // X, Y, Z, E maximum start speed for accelerated moves. E default values are good for Skeinforge 40+, for older versions raise them a lot.

#define DEFAULT_ACCELERATION 3000 // X, Y, Z and E max acceleration in mm/s^2 for printing moves
#define DEFAULT_RETRACT_ACCELERATION 3000 // X, Y, Z and E max acceleration in mm/s^2 for retracts

I have no idea what "raise them a lot" means at this point.

BTW, I have posted a few pics of my build in the past. A lot like Alan's - but a little simpler.

 

The fact that the first third of the print is OK, definitely sounds like overheating to me.

Have you calibrated your stepper drivers? Do they get any air flowing across them? Has a heatsink come loose?

I would guess one stepper driver is overheating, or set to a higher current value than the other.

By the way, huge thanks to Carl (and Adam) for this design. I should be starting the build of my version of the triple c variation next week.

 

Thx, I will chk that, but I have not had any issues with overheating since I built it (months ago).
The motors r just hanging in free space, no heat sinks or fans.
the driver current has been set
If it was overheating how would that produce a linear twist of the model?
If one motor was continually losing steps it would do that - in my case, the one on the left as my motors r on the back of my unit. That would cause it to run more on a diagonal.

But if it is I would expect a more random diagonal movement, otherwise it has to be losing steps at a fixed, consistent rate for the whole top 2/3rd of the print. Maybe more like a driver failure - doing 16 microsteps instead of 32. But that would yield a 22.5° twist - this is about 5°

How can it consistently lose 10% of the steps?

 

Well, the real test would be to start a new print from "cold" (ie, after the printer's been off for awhile). If it immediately starts skewing the print, then it's mechanical. If it works fine for awhile, then starts skewing, then you're looking for something that's getting hot and acting unusual in that condition.

 

It may be that my motor currents got out of adj.
How often does one need to check those?
I am certain they were set accurately in the past and I have not touched them.
I am using the 8825 bds set @ 0.75 for 2 amp steppers

 

Anyone using printed t-slot nuts? Could save a decent amount on fasteners

Just for illustration: M4 T Slot nut by daGHIZmo

 

Trying to put some finishing touches on my printer and wanted to get some input on speeds. What are some good values for print speed, acceleration, and jerk?

EDIT: Also, what sort of numbers are you using for steps per mm on all axis and the direct drive extruder?

 

What size nozzle are you using? I was doing 60mm/sec on my .6mm volcano, but am pretty sure I can go higher once I strengthen the printer frame more \ attach it to the wall. I run my .4mm Makerbot Replicator 1 at 120mm\sec pretty regularly (with sailfish firmware).
Note, right now I'm runing a 1.2 volcano nozzle on the direct drive, and I'm printing at 30mm\sec: I could print faster, but I'm I'm having a hard time maintaining temperature based on the large volume of filament being melted + the cooling fan.

Jerk: Never changed it from the Marlin defaults:
#define DEFAULT_XYJERK 20.0 // (mm/sec)
#define DEFAULT_ZJERK 0.4 // (mm/sec)
#define DEFAULT_EJERK 5.0 // (mm/sec)

I dropped my acceleration to 800 and it helped with some ringing it was having. It was 3000 by default. I did notice slower overall prints when I did this.
For steps per mm, that entirely depends on what stepper drivers you're using + the nuanced characteristics of your machine, so what other people have don't really matter since you need to tune them manually regardless. But since you asked, this is what I'm using on my DRV8825's, XYZE
#define DEFAULT_AXIS_STEPS_PER_UNIT {199.7403, 200.5415, 804.90995, 300}

Good luck getting it tuned in!

 

I always go for aesthetics myself #2 ftw!

 

I was just looking for numbers to get in the ball park to start the tuning. I know the steps for E is WAY too high right now. I am using a .4mm nozzle (1.75mm) and a Duet 0.85 controller with Allegro A4982 (I believe) at 16 micro-steps.