@droftarts Cool - looks like that 3 colour printer predates the Prusa MMU1 by quite a lot.

@Leonard03 thanks for your feedback and for testing all these firmware builds.

Just a brief update on this conversion. I could never get a good consistent first layer, the mesh bed levelling just did not seem to work as expected. Other threads discussed this but it's possibly down to the z probe offset from the nozzle and less than perfectly square construction. After spending far more time than I would have liked trying to work around this, I decided to swap the print head and z-probe out.
I already have a printer using one of the dyze designs horizon z probes and this had solved bed mesh levelling issues on that printer so went with one of those along with an e3d revo roto extruder/revo hot end with a high flow hardened nozzle. Having the nozzle act as the z probe get around any issues with an offset z probe.
I modelled up a new mount to fix to the x axis gantry and had this cut out of aluminium along with an adapter to attach the extruder to the horizon probe.
This solution has solved the mesh levelling and now the first layer is consistent. The total weight of the new print head is within 10g the same as the original.

new_mount_1.jpg

new_mount_2.jpg

new_mount_3.jpg

@stuartofmt

Looks promising - thanks.

Frederick

Time for an update! Files published here: https://github.com/jcwebber93/DuePrint3/
Includes

Interface Board KiCad Project Detailed BOM Wire List (from-to) RRF configuration files

Major changes:

Received, built, and tested the v1.1.0 interface board. I recognize its not that complicated of a design... but it works! Chief issue - functions that are driven by the OUT_X headers are powered on once power is applied to the system. This is bad in that, once power is applied, but before the 6HC fully initializes, the chamber and extruder heaters blip on. If the system is powered on, with the Duet not yet loaded with an updated config.g, the heaters will turn on. I need to address this, pretty major oversight. Added a small 48V power supply to power the 6HC (and by extension the stepper motors). Gives a bit of extra speed. Swapped out the Geckodrive G320x for a STM32 Nucleo-64 running SimpleDCMotor (derived from SimpleFOC). Several weeks back, I lost extrusion control. Extruder motor control is PWM signals (Stratasys control board, G320x, Nucleo-64, etc) -> series resistor -> 74AC14 hex inverter w/Schmitt Trigger -> L298 -> DC motor. I found I had burnt up a series resistor, as well as the 74AC14! On the Dimension 1200 PDB, the series resistors are 121ohm, while on the uPrint PDBs, they are 1k ohm. Based on some additional bench testing, I believe the G320x in this application outputs a peak voltage (and/or current?) that is much too high for the 74AC14s used on this PDB, as isn't intended to be used as logic control. I repaired the board (one damaged pad later...), but still needed a method of consuming STEP/DIR and outputting a signal to drive the extruder motor. Came across the SimpleDCMotor extension of SimpleFOC, which was quite straightforward to implement.

sIMG_3466.JPG

DuePrint3 v1.0.0.JPG

Not too bad for an ABS Benchy printed at ~200mm/s at 305°c hotend, 75°C chamber! Plenty more tuning to do.
50ddee67-4859-46d3-a738-9f0215c3293e-image.png

Hi,

Well I spent a couple of days striving for a set of Z axis end-stop sensors I was happy with.

Tonight the bed did it's first "auto level" using the end-stop sensors.

Worked just fine.

Next step - design and install the hotend and related bits and pieces - like a pair of Z probes.

Frederick

managed to get things printing at a decent speed it was issue with the slicer not set same as the config for the duet...running pretty good speed and nice and quiet thanks for all the advice

@Thatguywithathing Good question! From a long term sustainment perspective, keeping the majority of the Stratasys parts is a risk. If something breaks, or a consumable (nozzle, or liquifier tip as Stratasys calls them, etc) is needed, head to ebay and hope you can find it. Over the years when working on some other 3D printer projects, I've picked up a few random spares off ebay, so for me the risk is low (for now!).

In my case, some pros:

Limit switches, end of axis switches, door switch and head toggle switches already exist, as well as their wiring. Can reuse Dual extrusion system is repeatable and reliable. Can reuse. (I also quite like dual/multiextrusion, be it this, IDEX, etc. Would like to maintain this as a dual extrusion system). Infrastructure to power/switch chamber AC heaters already exists, can reuse 120V heaters for the model, support material are present, as well as the ability to drive them. Existing nozzles (liquifier tips) work well in heated chambers, no concerns about clogging or heat creep Cost - reusing as much as possible is likely the most cost efficient approach (assuming we start with a functional printer).

Added Capability

I added a small 48VDC power supply to power the Duet 6HC. This grants some additional speed for the X,Y, and Z stepper motors vs. stock 24V.

Cons:

Existing tool head is large & heavy. Yes, it offers great dual extrusion capability, but a smaller print head (or somehow 2x for IDEX) could zip around the chamber and offer excellent prints.

Personal Challenges:

Closed-loop DC motor control (for the extruder) - a fun personal challenge for me. DC motor control isn't something RRF (or other printer firmwares) natively support (RRF supports closed loop stepper motors via the 1HCL expansion board). Using a Geckodrive G320x with step/dir from an expansion board (could be a 1XD from Duet, or a Sammy-C21, etc) pretty much works immediately, but when driving the DC motor directly, you'll either see the driver fault out, or the motor will see excessive thermal rise - the G320x just isn't the right tool for this type of motor. I'm having much better luck with a Nucleo-64 running SimpleFOC's SimpleDC library. Custom PCB. The conversion can be done simply by running jumper wires from the Duet(s) to the PDB, but I wanted to dig a bit deeper into the world of electronics design.

Were the printer non-functional (print head assembly missing, no PDB, etc) then I for sure would have taken an alternative approach to converting it over to RRF. Your conversion is extraordinarily well done, and highlights how powerful modern controller boards are compared to the ancient boards in the stock machines!

@fcwilt I've no idea, perhaps one of the Duet folks knows? @droftarts ?

@mrehorstdmd this , i personally just want a simple system , the way him set right now , even power off bed will not move at all

I see, there are plenty of other good ideas. I like the ides with the stepper endstops from the pic @oliof posted.

Yes, @T3P3Tony, you have the biggest so far. 😊

@spiceminer

late to this thread but - beeter late than never ????

2 Caribou/Zaribo ... runing duet 2 ... my vorons run Duet3 mini + 5 though.

I dont run wolfgangs firmware releases though. Had my Zaribos on Duet before they ever really considered it.

Hope your machine is running good!

its just begging to be modded to be a corexy

@martinv said in Solution to delta printer falling arms and effector at power off:

I'm still thinking of neat things to print. Any ideas?

My delta is my go-to printer. I print everything I need on it, except where I need the slightly wider (300x200mm instead of 300mm round) bed of my tool changer. I've recently been printing tools and jigs on it to aid with assembly of a light aircraft.

@o_lampe Nozzle right now seating at +7.8mm height from bottom of XY-top, which I will need to lower to +2mm height by adding 6mm length on extruder back-plate to move it slightly lower. Then bed only need to "extend" over Z-bottom by 2mm, which is already achieved.

I'm going to print testing parts of XY this week to test it out, and see if any adjustment needed to the design.

@martinv

Hey Martin, I haven't tried playing around with the carriages yet. I don't think I'll bother with that, as I'd probably just upgrade to linear rails instead. I was contemplating that on this recent set of upgrades, but haven't decided if I'll bother.

The best change I've made so far was upgrading to a 0.9° stepper motor on the drive axes. It really made the prints come out quite a bit nicer. The switch to a whambam PC sheet also helped a lot.

Getting rid of the flying extruder in this recent set of mods has also helped the quality quite a bit, as the fast prints would make the Bondtech knockoff I had before swing around a bunch, causing artifacts at speed. I even noticed this during my autolevelling, the extruder moving would sometime make the levelling irregular, so I had to have a pause before each probe. It was annoying.

For high quality prints before, I used to stick to 30mm/s. Now 50-60 mm/s seems to be fine. I haven't pushed it much faster. I'm pretty sure I can move up to 100 mm/s and get decent quality now, but honestly most of my mods were to make the printer quieter rather than faster.

I'll let you know when I figure out my Orbiter mount. It should be in the next month or so when I get some time.

@magnets99 I’d think you could do the same thing with a BLTouch for 3D shapes, as the sensitivity and repeatability is usually around 0.01mm. However, if it’s a 2D shape that you’re trying to get the outline of, like a PCB, I’d think keeping it simple and using a scanner, or taking a photo of it, would be simpler and faster.

Ian

@o_lampe this was just for the show, to make it transportable. Using the whiteboard was convenient for demonstration purposes. Now it’s back home, I’m going to mount it to the wall, and see how it draws then. In the past, the board it was mounted on was always laid back a few degrees, which may both help and hinder drawing, by increasing the pressure on the pen.

It’s still not very precise, least accurate at the limits of its range by 1-2mm perhaps. I need to tune the speeds and accelerations a bit more, but we did find at the show that some low frequency, broad band input shaping helped with belt wobble a bit.

I could mount two more motors at the bottom, if I went back to using a full sheet board, but then I’d have to deal with winding the belts/cables onto a spool, and redesign the gondola to have 4 arms. 3 points would limit the drawing area.

It may be that something like the original CoreXY ‘Bot of the Cloth’ would make a better, more accurate, portable plotter. https://corexy.com/botofthecloth/index.html

Ian

@o_lampe that is a good observation:). There is an eye icon in the bottom right corner which opens up the usual DWC top panel (this is a feature of the BtnCmd plugin).

Screenshot 2024-11-08 at 14.33.03.png

And if you press the little button with the message bobble icon, it echoes the current polar coordinates to the console.

Screenshot 2024-11-08 at 14.34.45.png