Armored Turtle / Box Turtle AMS Filament changer

[PFarm]

On 2/13/2025 at 3:17 PM, mvdveer said:

Mine is not as per the BOM as I designed this prior to the BOM being released. Here is the official BOM. The last line of the BOM will drive the cost up exponentially - Polyfilament Dryers x 3

 I purchased a Sunlu S4 for 99.00 Cnd. to dry the four filament rolls before and after printing. At that price it was hard to pass up.

 

Edited February 15 by PFarm

[mvdveer]

2 hours ago, PFarm said:

I see you printed the dual-sensors were you able to configure the dual sensor in the AFC software? Also, do the corners have LEDs in them?

The dual sensors are not needed. At the time of my build, the bottom sensor could not be used. With the latest AFC software, it can be used, in a manner similar to the first sensor. I chose not to go that way. I only use the top/first sensor - this tells AFC when the filament has reached the toolhead, and then starts the load sequence. Seems to be what most of the builds on discord have been doing.

However, you do not need any sensor as AFC can use "ramming" to determine whether the filament has reached the extruder. I still like the idea of a sensor though. (Not even using sensorless homing)

No, I don't have corner LED's, as I could not figure out how to mount them. I might look at that again once I start building the LDO kit.

[mvdveer]

2 hours ago, PFarm said:

purchased a Sunlu S4 for 99.00 Cnd. to dry the four filament rolls before and after printing. At that price it was hard to pass up.

Sure is. The idea of the polymaker dryer, is to use only the dry box to mount on top of the BoxTurtle enclosure, according to the CAD. Unfortunately you cannot purchase the Dry box separately. And at AUD130 per unit, I think I will stick to the dryers I have, if I need to dry filament.

[PFarm]

11 minutes ago, mvdveer said:

 And at AUD130 per unit, I think I will stick to the dryers I have, if I need to dry filament.

 

I hear you at 106 Cnd per unit I agree! So they want to use the bottom half (heater) of the Polydryer? 

Edited February 15 by PFarm

[mvdveer]

1 hour ago, PFarm said:

I hear you at 106 Cnd per unit I agree! So they want to use the bottom half (heater) of the Polydryer? 

Yes. Called the dry box. So wasting a filament holder if you already have those. This is also mounted on the top of the enclosure. Heat rises, not falls - to me it doesn't make sense, but I am not a designer.

[mvdveer]

You'll also need a way to rotate the screen as it will be "upside down"

[PFarm]

7 minutes ago, mvdveer said:

Yes. Called the dry box. So wasting a filament holder if you already have those. This is also mounted on the top of the enclosure. Heat rises, not falls - to me it doesn't make sense, but I am not a designer.

 

 Are they fan-driven?  Even if they are doesn't make much sense to try and drive hot air down.

Edited February 16 by PFarm

[mvdveer]

1 minute ago, PFarm said:

 Are they fan-driven?  Even if they are doesn't make much sense to try and drive hot air down.

Must be fan driven - looking at the animation on the Polydryer site.

Sure doesn't make sense

[PFarm]

5 minutes ago, mvdveer said:

Must be fan driven - looking at the animation on the Polydryer site.

Sure doesn't make sense

It would make more sense to mount one down low on each side or front and back. Maybe make the enclosure larger and mount them inside.

Edited February 16 by PFarm

[PFarm]

On 1/5/2025 at 12:37 AM, mvdveer said:

Tuning of the Box Turtle, as it is still in Public Beta, was dependant on discord help and contrary to prior experiences on Discord, the Armored Turtle guys were non judgemental and very helpful.

Millennium Machine guys on Discord are very helpful as well with no smart-ass comments.

Edited February 16 by PFarm

[therm_virtual0y]

On 1/25/2025 at 12:55 PM, mvdveer said:

Had a lot of trouble getting the snappy turtle cutter to work reliably. The discord channel also indicates this is where users struggle. I had intermittent lane feeds that did not complete. The issue was that the macro for snappy pulled the filament past the hub, thus if there was stringing after it retracted from the hotted, this would confuse the sensor in the hub, or worse cause a partial blockage. Which means, you still had to do Tip Forming which in my opinion defies the purpose of the cutter. (This code in the macro has since been fixed).

I decided to abandon the Snappy Turtle and built a Galileo 2 Filematrix cutter. (The Box Turtle guys have consolidated all the files for use in the Box Turtle build in a GitHub called FilamAtrix)

Installed the toolhead and everything seems to be working as intended now. Since the V2, on which this is to be mounted has just been completely refurbished with CNC motion system, and Bowden lengths have changed, its, I will need to start the calibration process again and then attempt the serial print stl which has 45 tool changes in it.

There has also been a major update to the AFC_Klipper Add on software (AFC Klipper Add On) that runs the Box Turtle, so I'll need to update that first, then calibrate.

Will keep you posted

I built the filametrix G2E mashup and have had nothing but trouble with the cutter.  The depressor is mounted sturdy and is doing its job but the cutter seems to work great and then once in a while during the print, a cut piece comes out with the filament and then when it tries to load back into the tool head, it brakes off all the way and jams the extruder and MMU.  

What cutting blade are you using?

Where is your depressor mounted?

My cut happens pretty slow so maybe a faster movement?  I don’t know.  Spent hours and hours on it without completing a print other than first layer patches.  
I built the PicoMMU and it seems like it can work well if the loading and unloading process was dialed in.  I need to take a rest from it at this point but not too long because I’m close for sure 

[mvdveer]

3 hours ago, therm_virtual0y said:

What cutting blade are you using?

These are the ones I got and cut it to size with a Dremel:

No4 Cutting Blade on Amazon. Has been working without a snag. I do a double cut each time - don't know if that makes a difference.

variable_cut_count                : 2

I just received a batch of these from Fysetc on Aliexpress, already cut to size,  but have not tried it as yet.

3 hours ago, therm_virtual0y said:

Where is your depressor mounted?

Mounted front left of the printer as recommended on the Filametrix github

Mount on Gantry Front Left                                     GE2 Filametrix with dual sensors

   

 

                                                          Both shown together

 

                                               

 

Here is my cutting macro. (As per AFC configuration)

[gcode_macro _AFC_CUT_TIP_VARS]
description: Toolhead tip cutting macro configuration variables
gcode:  #leave empty

# This should be the position of the toolhead where the cutter arm just
# lightly touches the depressor pin
variable_pin_loc_xy               : 6.5, 66    # x,y coordinates of depressor pin

# Accel during cut. This will overwrite the global accel for this macro. Set to 0 to use global accel
variable_cut_accel                : 0

# Direction to make the cut move (left, right, front, back)
variable_cut_direction            : "left"

# This distance is used to move toolhead to cut filament
# and to create a small saftely distance that aids in generating momentum
variable_pin_park_dist            : 8.0       # Distance in mm

# Position of the toolhead when the cutter is fully compressed.
# Distance the toolhead needs to travel to compress the cutter arm.
# To calculate this distance start at the pin_loc_xy position and move
# your toolhead till the cutter arm is completely compressed. Take 0.5mm off this distance
# as a buffer. 
# Ex pin_loc_x : 9, 310  fully compressed at 0, 310 set cut_move_dist to 8.5

variable_cut_move_dist            : 6       # Distance in mm

# Speed related settings for tip cutting
# Note that if the cut speed is too fast, the steppers can lose steps.
# Therefore, for a cut:
# - We first make a fast move to accumulate some momentum and get the cut
#   blade to the initial contact with the filament
# - We then make a slow move for the actual cut to happen 
variable_cut_fast_move_speed      : 32        # Speed mm/s
variable_cut_slow_move_speed      : 10        # Speed mm/s
variable_evacuate_speed           : 150       # Speed mm/s
variable_cut_dwell_time           : 50        # Time to dwell at the cut point in ms
variable_cut_fast_move_fraction   : 0.85      # Fraction of the move that uses fast move 0.0 - 1.0
variable_extruder_move_speed      : 25        # Speed mm/s for all extruder movement

# If the toolhead returns to initial position after the cut is complete.
variable_restore_position         : False     # True = return to initial position, False = don't return

# Distance to retract prior to making the cut, this reduces wasted filament but might cause clog 
# if set too large and/or if there are gaps in the hotend assembly 
# *This must be less than the distance from the nozzle to the cutter
variable_retract_length           : 20

# This can help prevent clogging of some toolheads by doing a quick tip from to reduce stringing
variable_quick_tip_forming        : False

# Number of times to run the cut movement
variable_cut_count                : 2

# Retract length and speed after the cut so that the cutter can go back 
# into its origin position
variable_rip_length               : 1.0       # Distance in mm to retract to aid lever decompression (>= 0)
variable_rip_speed                : 3         # Speed mm/s

# Pushback of the remaining tip from the cold end into the hotend
# *Must be less then retract_length
variable_pushback_length          : 15  #25        # Distance in mm
variable_pushback_dwell_time      : 20        # Time to dwell between the pushback

# Safety margin for fast vs slow travel. When traveling to the pin location
# we make a safer but longer move if we are closer to the pin than this
# specified margin. Usually setting these to the size of the toolhead
# (plus a small margin) should be good enough 
variable_safe_margin_xy           : 30, 30    # Approx toolhead width +5mm, height +5mm)

 

 

[therm_virtual0y]

I am using PicoMMU but filametrix tool head but like the code from Armored Turtle and will see if I can implement the double cut.  Will probably help.  I have my depressor mounted the same place too.  Looks good .

what are some of your filament change numbers?  I have a print I want to work on but it has 190 changes.  Also the purge amounts are not very dialed in so I’m getting cross color contamination.  I’m using orcaslicer and the purge amounts are calculated every swap.  Wonder if there is a tutorial for this?   

[mvdveer]

3 minutes ago, therm_virtual0y said:

 I have a print I want to work on but it has 190 changes.  Also the purge amounts are not very dialed in so I’m getting cross color contamination.

I am using blobifier for purging, not ORCA settings - can highly recommend this. Have not had any colour contamination. The most changes on a single print I've had were 97. With the BoxTurtle, did not have to intervene once. 

[therm_virtual0y]

Nice. Thanks for the info

[Penatr8tor]

12 hours ago, mvdveer said:

The most changes on a single print I've had were 97. With the BoxTurtle, did not have to intervene once. 

That's pretty impressive. 

[NurseKateDesign]

On 2/15/2025 at 6:57 PM, mvdveer said:

Yes. Called the dry box. So wasting a filament holder if you already have those. This is also mounted on the top of the enclosure. Heat rises, not falls - to me it doesn't make sense, but I am not a designer.

I don't want to a bad mouth but in physic hot hair it supposed to go up. By putting trhe polydryer up  tyhe dryer of the enclosure will not be effective. And the cost wow, just for the enclosuise with the dryer it's around 400 to 500$US.

[NurseKateDesign]

On 2/15/2025 at 7:11 PM, PFarm said:

It would make more sense to mount one down low on each side or front and back. Maybe make the enclosure larger and mount them inside.

I had a discutioni with 2 ldo employe on discord  and they are working for a more effective enclosure dryer, similar to the filamentis enclosure. The heater will be a 24 volt 300watt heater with 2 fan will shoot the hot hair under the filament the circulation should be facilitate bu the round top

[NurseKateDesign]

On 2/15/2025 at 8:59 PM, PFarm said:

Millennium Machine guys on Discord are very helpful as well with no smart-ass comments.

I'm happy tpo see that I'm not the only one thinking that.

[Penatr8tor]

On 2/15/2025 at 5:57 PM, mvdveer said:

Heat rises, not falls - to me it doesn't make sense, but I am not a designer.

You can't break the laws of physics but... you can cheat.

What I mean by that is... if you have a fan that can circulate the heat around the chamber, you can minimize the effect of heat rising and pooling at the top. There will always be a temperature variation between the top and bottom but not enough to matter. As long as you circulate the air in the chamber... you will distribute the heat enough to complete the job, which is to dry the filament or more accurately... keep the filament as dry as possible while printing.

[PFarm]

@mvdveer You self-sourced your first Turtle Box was the cost halved (or close) of the LDO cost? I'm seriously thinking of self-sourcing mine. You mentioned having issues with your Moon steppers were they purchased on Aliexpress? If so which seller did you purchase them from? I'm looking at getting them from Trianglelab. Both my printers run Moon steppers and I like them. Purchase whatever I can for Isik's Tech. Any other recommendations?

Edited February 20 by PFarm

[mvdveer]

1 hour ago, PFarm said:

@mvdveer You self-sourced your first Turtle Box was the cost halved (or close) of the LDO cost? I'm seriously thinking of self-sourcing mine. You mentioned having issues with your Moon steppers were they purchased on Aliexpress? If so which seller did you purchase them from? I'm looking at getting them from Trianglelab. Both my printers run Moon steppers and like them. Purchase whatever I can for Isik's Tech. Any other recommendations?

The cost was half, ( in total around AUD250) and that included sourcing expensive misumi extrusions, which is really not needed. The most expensive was the AFC lite board as importing it to Australia had a hefty postage bill attached to it. 

I did not purchase the motors from Isik's Tech, but from Aliexpress, but in retrospect Isik Tech motors would have been preferable (Apparently quieter)

I bought the 80x8mm rods on amazon very cheaply.  The D2HW switches and ECAS couplers from Aliexpress - cheap as.

The small bearing from Amazon - again very cheap

The moons motors were from Aliexpress - I bought in bulk, and not from a reputable store (was cheap). thus I think there may have been an issue with them not being genuine moons. (Though never had an issue with Moons motors before)

Stick to the turtleneck buffer, don't be tempted to go with the TN2, all it does extra is LED's which you really not need this. And it will drive up the cost.

Speaking of LED's - make sure you buy RGBW. I used some that I had that were used in the Stealthburner and had some trouble configuring them. 

[mvdveer]

For wiring the LED's I got these of Amazon, which I will highly recommend.

[mvdveer]

14 hours ago, PFarm said:

You mentioned having issues with your Moon steppers

I have just had a look at the BOM for the Box Turtle and the steppers I bought. Sad to say but I bought the WRONG steppers - so it is not the brand, but the model of the motor. But although the motors are branded Moons, the Aliexpress site states it is BTT motors - Mmmmh??  So maybe not a bad idea buying from Isik Tech's site. (Not bad at $11US a motor). But the same motor on Aliexpress sells for about $8.50US and free shipping.

I think LDO motors are a bit of an overkill.

[PFarm]

@mvdveer I appreciate the info everything from Aliexpress comes to 175 CND including the servo for Blobifier, and LED wires you suggested, just need the AFC board from Isik's Tech. For from $455 for the LDO kit. I already have leftover 2020 & 2209 drivers. I went with the DC and stepper motors that are in the BOM.

Edited February 21 by PFarm