Profile

Cover photo
Whosa whatsis
Works at Deezmaker
Attends University of Autodidacticism
1,620 followers|1,163,135 views
AboutPostsPhotosYouTube+1's

Stream

Whosa whatsis

Shared publicly  - 
 
Found this teardown of the MBI extruders via +Fabbaloo. I wasn't surprised by much based on what I had already seen/read/reasoned-out, but there were a couple of interesting things. First, a minor surprise was that the encoder wheel is optical rather than magnetic as I had assumed.

Second, the nozzle actually moves up and down during retraction/recovery. This is an idea that I have been working on for a while. The idea is to get the benefits of Z lift without moving the Z axis by pulling the nozzle back slightly in its mounting. There are several types of actuators that could be used for this, but they're not necessary if you can get the nozzle pressure to push/pull the mechanism the way you want it. The normal back-pressure of the nozzle will push the nozzle down toward the platform when you are extruding, so you really just need counteract the weight of the nozzle to lift it (normal retraction will produce a negative pressure, but it may not be enough, and the weight will slowly pull it back down as the nozzle pressure equalizes after the retraction.

The obvious solution to this is to use a spring, but a spring's force increases as it compresses so the spring will want to reach an equilibrium with the nozzle pressure that may not result in the spring being fully-compressed when the nozzle is under pressure, and the spring's force (at least if it's a weak enough one to ensure that the nozzle pressure will completely overcome it even at low speeds) is likely to drop off completely when the nozzle is fully raised. You want it to hold strongly in the raised position and only pull weakly (just enough to lift with no nozzle pressure). You need a bistable mechanism that will hold strongly in the down position when the nozzle is under pressure, and hold strongly in the up position when it is not, but will transition easily between the two with the rapid movement of retraction/recovery.

The solution turns out to be a magnet, which pulls strongly when the assembly moves up and gets close to it, but the magnetic force drops off quickly as the metal moves away so that only a little nozzle pressure must be maintained to keep the assembly separated from the magnet. Apparently, $403 million in stock from Stratasys is enough for MBI to hire someone capable of figuring this out, rather than just staying two years behind everything the Reprap community has been doing.
23
7
Matej Rozman's profile photoBenjamin Rockhold's profile photoMike Miller's profile photoCarlos García Saura's profile photo
6 comments
 
Well, you can't say they're not moving the state of the art forward. That optoencoder would have saved a BUNCH of my prints. 

ETA: Still, a little janky that every 5th Gen is going to have a vestigal appendix like organ where the filament was supposed to go. 
Add a comment...

Whosa whatsis

Shared publicly  - 
 
Build later share.
 
So ya... This is amazing!  I think this is going in my list of CNC projects to do!  +Jason Smith  thanks for sharing!

Pyro Board: 2D Rubens' Tube!
17
9
jl roberts's profile photoSteve Graber's profile photo
Add a comment...

Whosa whatsis

Shared publicly  - 
 
I still want to see patents abolished, but this is the next best thing.
5
Billy Zelsnack's profile photoWhosa whatsis's profile photoCamerin hahn's profile photoStephen Baird's profile photo
19 comments
 
+Billy Zelsnack that doesn't work... they attack both other lawyers and innocent people and bring nothing to the market
Add a comment...
 
Oh, tiny microcontroller dev boards that I already have too many of and will probably never get around to using, I wish I knew how to quit you.
19
2
Justin Walterich's profile photoCoco Jx's profile photoHaroon Rasheed's profile photo
 
I got "onboard" this one also.
Add a comment...

Whosa whatsis

Shared publicly  - 
 
The skanbot fits in my tiny car without disassembly, and it will fit even more easily once I get the new, more compact turntable mechanism working. This one is from an old project and is bigger than it needs to be.
3
Add a comment...
Have him in circles
1,620 people
Johnny Russell's profile photo

Whosa whatsis

Shared publicly  - 
 
I just thought of a way to unify the two major theories about bridging.

Some people say that they have the best results with bridging when they move across gaps as fast as possible, while others (like me) find the opposite.

It seems that the key to this difference is die swell. Pushing filament out quickly will increase die swell, resulting in a filament that will pull itself tight while becoming thicker as it leaves the nozzle. Pushing filament more slowly by moving slowly with a decreased flow rate will decrease die swell and make it easier to draw out a long, thin strand, that gets thinner as you increase its length.

In either case, the  bridge strands will be round rather than being flattened as a non-bridge layer would, but the first strategy results in thick strands (possibly thicker than the nozzle bore), while the latter results in thinner strands. I've always moved at a slow speed while decreasing bridge flow ratios for bridging with thick layers, and increasing it for bridging very thin layers, but this suggests that the optimum strategy might be to leave the bridge flow ratio alone, and only change the bridging speed. Simply increase the speed to increase die swell if your bridges sag, and decrease bridging speed to decrease die swell and gently draw out the strand of plastic if your bridges are breaking. If the slicer is already using the volumetric flow logic to put out the right amount of material for the bridge layer, this should put you pretty close to the optimum layer volume. Always adhering to one of the above bridging strategies, on the other hand, will result in bridges that span well, but are either over-compressed (swelled plastic on thin layers) or under-compressed (drawn-out plastic on thick layers). This combined strategy should avoid both of these problems while reducing the calibration process to a single variable: speed.
25
4
Ben M's profile photoNils Hitze's profile photoTyler Harney's profile photoAshley Webster's profile photo
Ben M
 
Thanks, I've always thought there was some magic to the bridge flow rate setting but this theory makes a lot of sense. I'm going to give it a try
Add a comment...

Whosa whatsis
owner

Development of the standard  - 
 
I've found another value that we need to include. I've noticed, specifically when working with smaller-diameter drive gears, that the compressibility of the filament is an important factor. A very compressible filament will need an increased flow ratio because, as the teeth of the drive gear are able to sink deeper into the filament, the filament will be closer to the center of the drive gear. This effectively reduces the diameter of the drive gear, thus increasing the steps/mm. Figuring out how to calculate this increase, and even what measurement to use for the calculation, will take some more research, but this is a value that will be needed for the type of "plug and print" functionality we're trying to make possible, at least with current forms of extruder hardware (though it may be possible to bypass this need by using encoder-equipped extruders).
5
1
Whosa whatsis's profile photoCamerin hahn's profile photoJo Bigger's profile photoMarcin Lukasik's profile photo
9 comments
 
But the radius of the hobbled bolt changes with the depth of the bite. Assuming you are using a ~10mm diameter pulley for your extruder and the teeth byte .5mm deeper on ABS than PLA you will need to adjust your flow rate 5% more for ABS (if i am doing the math correctly)

I am hesitant to add this until we find a nice way to correlate a measurable number to a print flow rate. +Whosa whatsis i know you are doing some experimentation on this. we would need to try and decouple this from other information as well to try and make it work without asking the user to preform computations before every print (IE, my material stiffness is X, my harness is Y, my print speed is A, the ambient air temp is..... today happens to be vernal equinox, so my flow-rate is 1.05)
Add a comment...

Whosa whatsis

Shared publicly  - 
 
Maslows_hierarchy[] normally looks like this:

{Physiological needs, Safety needs, Love and belonging, Esteem, Self-actualization}

But some days, it looks like this:

{Coding, Coding, Coding, Coding, Coding}

My guess is that there's a buffer overflow vulnerability that allows it to be overwritten this way.
8
Add a comment...
 
Why hasn't anybody made a small motor controller to that uses an encoder and takes step/dir signals in a format compatible with a pololu driver so that we can use servos as drop-in replacements for steppers on printers yet? I would think there should be chips designed to do this, but even if it requires a microcontroller and a separate H-bridge, it shouldn't be difficult...
3
1
Brad Hopper's profile photoKevin LaSanders's profile photoMike Creuzer's profile photoOla Sikström's profile photo
38 comments
 
+Whosa whatsis Sorry for going on about shaft encoders but I really think they may be useful. A geared dc motor with encoder will cost roughly the same as a stepper, but will be stronger and way lighter (probably closer to 20-50 g, while a stepper often is 200+ g). The most obvious place for this is of course the extruder.

Linear encoders for the axes sounds like a good idea anyway.

Where you actually put the encoder shouldn't really matter though (implementation-wise), so wether you use a linear- or a rotary en coder, you could still use the same driver hardware.

As for the hardware, I'm confident this can be implemented with a cheap ATtiny and an H-bridge driver chip, on the same footprint as an A4988.
Add a comment...
 
A couple of interesting ideas here. Looks more like a CNC mill's toolpath than a printer's. Shallow-angled nozzles need not apply.
39
7
Camerin hahn's profile photoScott Leighton's profile photoWes Brown's profile photoNick Parker's profile photo
15 comments
 
+paul mplr that's where I've been looking with all this. Arbitrary layer shape should allow you to mitigate layer interface weaknesses in an appropriate way for your application.

Also, for ninjaflex and similar, you could manipulate flexibility with layer shape.
Add a comment...
People
Have him in circles
1,620 people
Johnny Russell's profile photo
Work
Occupation
Professional Maker
Employment
  • Deezmaker
    2012 - present
Basic Information
Gender
Male
Story
Tagline
Polymathic autodidact
Introduction
On a good day, I'm an iconoclastic, autodidactic polymath. The rest of the time I'm just a cynical, dilettantish tinkerer.

Also, I make stuff.
Bragging rights
Designer of Reprap Wallace
Education
  • University of Autodidacticism
    present
  • Internet School of All Human Knowledge
    present
Whosa whatsis's +1's are the things they like, agree with, or want to recommend.
Deezmaker Introduces Bukito Portable Mini Printer Kit #3dthursday
www.adafruit.com

One of the new models launching at Maker Faire Bay Area this year that caught my eye was the Deezmaker Bukito — a capable, affordable, porta

Canny Machines
www.cannymachines.com

OpenSCAD Dimensioned Drawings. Tags: openscad · dimensioning in OpenSCAD. Dimensioned Drawings with OpenSCAD. OpenSCAD is an open-source pro

Ultra-high-molecular-weight polyethylene - Wikipedia, the free encyclopedia
en.wikipedia.org

Ultra-high-molecular-weight polyethylene (UHMWPE or sometimes shortened to UHMW), also known as high-modulus polyethylene (HMPE) or high-per

Stepper motor - RepRapWiki
reprap.org

What is a Stepper Motor ? Stepper motors are one kind of electric motor used in the robotics industry. Stepper motors move a known interval

Project Glass
plus.google.com

Thoughts, designs, and stories.

Re: [MakerBot] Hot end mod - Google Groups
groups.google.com

Re: [MakerBot] Hot end mod. It's kinda hard to see, here's an illustration and a better photo. The cuts start around 12mm away from the thre

Google Drive
drive.google.com

Access everywhere. Google Drive is everywhere you are—on the web, in your home, at the office and on the go. So wherever you are, your stuff

Pleasant Software - Pleasant3D
www.pleasantsoftware.com

Pleasant Software for the Macintosh - Podcasting solution, iChat enhancement and more

Polyimide - Wikipedia, the free encyclopedia
en.wikipedia.org

[edit] Classification. According to the composition of their main chain, polyimides can be: Aliphatic (linear polyimides),; Semi-aromatic,;

RepRap Wallace by whosawhatsis - Thingiverse
www.thingiverse.com

Thingiverse is a universe of things. Download our files and build them with your lasercutter, 3D printer, or CNC.

Acrylonitrile butadiene styrene - Wikipedia, the free encyclopedia
en.wikipedia.org

Acrylonitrile butadiene styrene (ABS) (chemical formula (C8H8)x· (C4H6)y·(C3H3N)z) is a common thermoplastic. Its glass transition temperatu

Printable solar cells that can be folded up when not in use - Hack a Day
hackaday.com

Here's a photovoltaic cell that can be printed onto paper. The manufacturing technique is almost as simple as using an inkjet printer. The s

Kickstarted: Finding Space (and Making a Makerspace)
blog.makezine.com

For many maker pros, finding the right amount of space can be difficult. One makerspace, Artisan's Asylum, is offering a different model.

Is One of Our Open Source Heroes Going Closed Source?
blog.makezine.com

A new player on the field by the name of Tangibot recently arrived, not with a new printer or even an innovative new derivative, but with a

Filament Tolerances and Print Quality | ProtoParadigm
www.protoparadigm.com

Filament Tolerances is a subject that doesn't get much discussion, yet is arguably the most important quality to look for when choosing a pl

Android Terminal Emulator
market.android.com

Access your Android's built-in Linux command line shell. Unleash your inner geek! Want to know more? Check out the wiki: http://github.com

StepStuck
hydraraptor.blogspot.com

When I built my Mendel I used A3977 stepper drivers. Before that I did some maths to show that the component values need to be carefully sel

How to view open Android Chrome Beta tabs in Chrome desktop
howto.cnet.com

Trying to view open tabs from an Android device in the desktop client didn't seem possible, but a savvy Twitter follower saves the day. Read

Pololu - Stepper Motor: Bipolar, 200 Steps/Rev, 35x36mm, 2.7V, 1000mA
www.pololu.com

This hybrid bipolar stepping motor has a 1.8° step angle (200 steps/revolution). Each phase draws 1000 mA at 2.7 V, allowing for a holding t