Mods.. (The Good, The Bad, The Ugly)

[wepollock]

My original intent was to keep the Artemis in a solid state from the factory; instead I am using the machine as a "lab machine" (with my Rostock Max V2 being my main "production" machine). It's not that the Artemis needs to be modded in any way as its very much turnkey. If you make a mod it would be good to describe whether or not it is good, bad, or just plain ugly.

1) Capricorn Tubing - I think that is a very good value and an easy mod to make. Also the Razor Tube Cutter is a great tool. I would recommend that you read the release notes, I would not use this tubing within the hotend.

2) Geared Extruder Titan - Having installed a Titan extruder I don't see any advantage to it, in fact there are some major disadvantages to it @12v in a Bowden Configuration

3) Patch Board to make any brand of hot end or effector plug and play - This is a very ugly type of change that I think will be very valuable over time. I would rewire my production machine to an end configuration to make it look nice. I think home run wires should run from the effector to a patch board without connectors. At the patch board there should be spare wiring I think this is a very good change. Also it would be great if the machine was designed to make snaking wire a bit easier - there are plenty of access holes at the sides of the channels but the top is not very accessible once the machine is assembled.

4) PETG roll in clips - these are great.

5) Duet Board Rewiring - (will eventually change) - I want to use heating elements without polarity, I want every fan connection to have positive and negative.

6) 1.8 motor for bowden extruder (like to test)

7) Titan Aero with extruder on effector (testing ; moved the filament spool near the top of the machine; filament mount can be used to enclose the machine. The filament mount is designed as a hockey stick.)

8) Bondtech extruder on effector (Looking at the Bondtech Extruder - its the bomb)

9) 24V

10) Ir Probe (like to test)

11) Fan -added 4/12 - I am currently printing with a Titan Aero in PLA w/o a fan.. I have been printing on my Rostock Max w/o turning on the part cooling fan. My conclusion here is that, while useful in many circumstances, fan cooling is really overused.

12 pt100

[wepollock]

First Print with a Titan Aero
1) Speed is the only downside I see with this mod.
2) An 8 ring Gyro is a reference print I use https://www.thingiverse.com/thing:2067014

a) the centerball is usually extremely difficult to print (I sometimes don't print the center ball)
b) to print on the Artemis I had to run lean
c) it took a tremendous amount of tuning to get this printing in good quality in a Bowden configuration


With the Titan Areo my first print WOW! I used NO cooling, .5 nozzle, 25 Layer Height.. 2mm retract 1mm coast 205 temp .52 layer width (three layers are visible without seams or spaces). A wimpy 14NM stepper @.8 amps There was NO tuning! The only downside I see is 25% more print time (which is tunable higher, I just did not want mass flying around)I had my doubts that the introduction of mass would be an improvement.

As per the patch panel. - If I want faster speed I can just change effectors! SE300 or what have you.

I used Hackney's titan Aero Mount. http://www.sublimelayers.com/#
I had to position the filament spool holder to the top of the machine and away from moving parts.

[img]http://i.imgur.com/oAiFdXG.jpg[/img]
[img]http://i.imgur.com/qtIyFpM.jpg[/img]
[img]http://i.imgur.com/lxbsL1d.jpg[/img]
[img]http://i.imgur.com/2IdaCDN.jpg[/img]
[img]http://i.imgur.com/jU63VNB.png[/img]

[dmo]

Why wouldn't you use the Capricorn inside the hot end?
Are you having to print slower with the Titan because of the weight from the direct drive. Have you seen the Nimble yet. They're out of stock but I want one for my v3.
The ir probe is not nearly as accurate as the fsr. Plus they're a pain to use.
The 1.8 degree motor might work well with your geared extruder.
Thanks for sharing your experiences.

[wepollock]

Hello DMO

Good to know that the IR sensor is not as accurate as the FSR.

Reference Print

One of the first prints I did with the stock extruder was a "hard to print octopus" I had no problems printing it
---http://www.sublimelayers.com/2017/12/pr ... print.html

Now I am using a "super gyro" (which I give away - see how you do with it on your machine?)

https://www.thingiverse.com/thing:2067014


Capricorn Tubing

I noticed with the Capricorn tubing that the outer diameter (outside of the tube) tends to be a wee bit smaller than the generic "white tube."
Capricorn states that they are 3.95 +/- 0.05 and the white stuff is 4.0 +/- .1.. I am measuring consistently .1 thicker on the white tube. With the bowden tubing inside of the hot end - I think you want smooth on the inside and grippy (firmly held) on the outside. I would definitely recommend that the Capricorn tubing be run from the extruder to the hot end, but I would not try to put Capricorn tubing into a hot end. {I also remember reading about this concern somewhere}

Titan Extruder - Print Five

I am getting the BEST prints I have ever gotten from a 3d printer. (printrbot (direct), folger tech kossel, flashforge (direct), custom build x2 , Rostock Max 2) I am getting the BEST prints I have ever gotten from a 3d printer, without tuning anything.

I started out by printing very slowly, and now I am up to Rostock Max speeds. (I am not going to go any faster).

Newton, High Skool (school) level physics, and plain old common sense indicated to me that putting mass at the long end of three sticks and moving it around by jiggering rubber belts was a bad idea. Having that mass "fly around" sounds even worse. I am now quite convinced that a long bowden tube is sub optimal. With the bowden I was able to print the "hard to print" octopus, I had to tune the hell out of the Bowden to get good prints. There is nothing wrong with the stock configuration. Bottom line, my perception has been proven wrong.

With Printrbot, Creality or a Prusa there is lots of mass moving around, and the workpiece is moving around! and that mass is not being held into place from six points.

I may address the speed constraint with a larger nozzle diameter. And I have modified in such a way that I can still use the stock configuration or any other effector platform. And I can swap them out in about five minutes.

Thermistor

I have a post on the Duet tuning page where I measure the thermistor vs a cheap thermocouple. The issue is getting a repeatable temperature (which I can easily do). It would be nice to know the actual working temperature of the hot end.

Cork

I have a single sheet of 1/2 inch cork on order $7. The big picture in the prior frame may support a partial enclosure. I want to see if Cork is a candidate for building an enclosure. My first thought was lexan. Cork might have some advantages. Its really cheap, comes in sheets, easily cut, somewhat flexible, and is an insulator. I am looking at building a partial enclosure.

Nimble. looks neat! https://www.youtube.com/watch?v=balfNZNx_58
(be interested in finding out what folks experience with it -- I think some weed whackers, dental tools, and dremels transfer energy the same way)

Based on your avatar you might like this print... https://www.thingiverse.com/thing:2375134

[adarcher]

Why wouldn't you use the Capricorn inside the hot end?
Are you having to print slower with the Titan because of the weight from the direct drive. Have you seen the Nimble yet. They're out of stock but I want one for my v3.
The ir probe is not nearly as accurate as the fsr. Plus they're a pain to use.
The 1.8 degree motor might work well with your geared extruder.
Thanks for sharing your experiences.

During my initial debugging of my build I found my original white PTFE tubing in the SE300 to be kinked. I had gotten some cap tubes in an attempt to lower the friction of the longer bowden so I snipped a piece and put it into the SE300. It barely fit. The outside was much wider than the white stock PTFE and therefor caused more binding the hotend. The filament was barely able to be pushed through the section of the cap tube that was pushed into the heat break when I finally took it out. The SE300 heat break has a smaller diameter PTFE insert seat than the e3dv6 heat break--cap tubes fit perfectly in there.

I might play with sanding down the outer edge a bit or ream the inner edge and test again some day. The cap tubes were much more compression resistant and the inner smoothness is amazing.

It's more likely that I'll give up on the SE300 and install a e3dv6 though.

[wepollock]

It's more likely that I'll give up on the SE300 and install a e3dv6 though.

I am still very surprised that an effector mounted extruder is far superior to the bowden configuration, even without cooling.
--I am even getting better prints by disabling zhop...

Z hop of .3 is still required for the super-gyro.. (as the inner ball tends to curl a bit) give it a shot!

https://www.thingiverse.com/thing:2067014

(I am guessing that 24v is not going to produce big results like the effector mounted extruder... )

[wepollock]

Here is an idea for a partial enclosure; not to keep heat in; but to stop cold air from rolling into the printer from the adjacent window.
A piece of Lexan would work too... but this is much cheaper.
Thoughts?


[img]http://i.imgur.com/Oeato5D.jpg[/img]

[wepollock]

I just put a Bondtech direct driver extruder with an E3dV6 hotend on my Rostock Max V2 Franken-Printer. Instead of using a stepper from E3d that is .9 degrees I used a RobotDigg 1.8 stepper which weighs in at 200 grams; both these motors are equally as wimpy @14nm. E3d suggested that if I go to town on tuning the .9 motor in software it would work with Rambo.. I was not so confident because with .9 you are doubling the move instructions on the stepper from a beefy 400 to ludicrous speed 800-900. The Duet could handle the added instructions no problem.

The 1.8 motor required no tuning or tweeking (spell check thinks that it required twerking, no twerking was required - if you are going from direct drive you will have to reverse the direction of the stepper.).

https://www.robotdigg.com/product/1034/ ... pper-Motor
https://e3d-online.com/nema17-slimline-stepper-motor


I used Hackney's Bondtech mount and I made a few cooling tusks but I decided instead to build a cooling ring. The tusks are a great idea but for me they were a bit limited in airflow and tight in tolerance around the heat block. However, It was lots of fun to drill itsy bitsy holes in 4mm hobby pipe with my rotary tool. I was going to use the tusks a rationalization to buy an itsy bitsy drill press, but drilling the tubes were easy-peasy. The tusks are not hard to make. Bottom line with cooling, on balance you need much less of it than everyone seems to perceive. The tusks put the cooling in the exact right place; my cooling solution does not put the cooling in the exact right place, but that is fine.

I also suspended the effector plate a full 1.5 centimeters in distance from the mounting plate. The Bondtech has a swingarm that would have been blocked otherwise. Turns out I really don't need to access the Bondtech swing arm because loading is totally better experience than with the Titan or EZRStruder.

I am currently printing PETG on the Rostock Max V2 with a Bondtech Direct Mount Extruder. Here are my findings;

The Bondtech is head and shoulders better than the Titan Extruder.
Gripping the filament from both sides without using aggressive gears is also good.
It is by far the easiest extruder to load.
The print quality is off the hook, even better than the Titan Aero.
There is no stringing on PETG, none; I have a 1.2mm extract set and a 1mm coast.. That is it that is all.
Only a single fan is needed to effect cooling.
On the downside I lost about 120mm of print height, and in the past I have printed within 1cm of the max build height.
On the plus side, this is no big deal because I changed the wiring around so my effectors are plug and play.

Bottom line - this is the best extruder I have used. However, it might not be advantageous to use this extruder in a Bowden configuration.
Putting the extruder on the effector is such a large improvement, I don't think other changes (such as 24v) would be justified. Stiffer arms, better belts might be the way to go.

Where the R&D effort needs to be; it needs to increase the payload that the effector can carry, or it needs to reduce the weight of direct extrusion. The Bowden configuration is easy to implement but it will not yield the best result compared to direct extrusion. PERIOD, I am still surprised..

[img]http://i.imgur.com/nx88Ez8.jpg[/img]

[wepollock]

When I mounted the titan aero on the effector I did not consider part cooling. I printed some PETG "tusks" that put the cooling exactly where it needs to be. I had to restrict bed.g a bit and I lost some printing height and radius. I think that the bed.g works a bit better without going to the full extent of the print bed, near the towers.

The next version of this "tusk" cooling will reclaim another 10mm in the Y direction. V1 and 2 are in the second picture.

I am currently printing a double-sized Marvin with a .6 nozzle and have a smaller tusk cooling solution planned. I don't want cooling to be in the way of the titan gears. My conclusion is that the having extruder on the effector is a major improvement even without part cooling. The bondtech extruder is far superior to the titan aero, the design of the bondtech is better for a delta as well. I just happened to have an inventory of Titan parts. I permanently removed the factory extruder, the stepper motor, and the factory hot end. I might want to rewire the printer. However, I am not going to 24v. I am looking forward to carbon fiber arms.

[img]http://i.imgur.com/u4gxaCX.jpg[/img]

[img]http://i.imgur.com/orN3QsB.png[/img]