Need For A "Default Setup"

[Bill Havins]

The purchase of my Rostock MAX kit was my “introduction” to the world of 3D printing. I’m a newbie in this area. I want to share some of my impressions of attempting to assemble or “build” a Rostock MAX, and offer a suggestion that might be helpful to Rostock MAX builders of all levels of 3D printing experience.

Let me preface my comments by noting I am not a newbie to the world of open source efforts, on-line development, users groups, and the internet. If you “google” the term “Chinese scales” one of the first links that comes up is an article I wrote in 2004 as part of a yahoo! group (the article is entitled, Reliability and the Chinese Scales “Experience”). We were working to develop a three axis digital readout that hobby machinists could assemble themselves. The original units, called the “DRO-350,” are still in use around the world.

Before I ordered my Rostock MAX kit I did many, many hours of research on 3D printers. This included research on the internet and in print media. I called several manufacturers and was fortunate enough to get several technical support members to talk with me about their respective printers. I eventually chose to order a Rostock MAX.

While I was waiting for my Rostock MAX kit to arrive I read most of the threads on this forum. Then I read them again. I was struck by the fact that forum members were completing their builds in different ways. I was also struck by the fact that components in the Rostock MAX kit were “evolving”; parts were changing from one release of the kit to the next and little information was being released about these changes. I took comfort in the fact that there was an Assembly Manual available and that there was also an Assembly Manual available for the accessory LCD Controller I had ordered.

After my Rostock MAX kit arrived I quickly began “the build.” I began to encounter errors in the assembly manual/changes in the parts included in the kit since the manual was written. I posted at least two queries on this forum, and also added a thread offering a method for assembling the new LCD Controller/RAMBo interface board. I also continued to read threads that appeared on the forum, hoping to glean nuggets of wisdom that might speed development of skill using the Rostock MAX.

As I prepare to power up my Rostock MAX for the first time I am struck by the fact that there is no clearly-defined “default setup” for this machine – there is no “standard” that says “ your Rostock MAX should look like this, function like this, and produce prints like this.” I strongly believe that users who are new to 3D printing, and experienced printers who are new to delta printers, would benefit from four basic “defaults.” The first of these includes a “standard” method of assembling the Rostock MAX that is clearly and accurately defined in a unified Assembly Manual.

The second “default” that I would suggest is a .zip file that includes all software necessary to operate either version of the Rostock MAX (i.e., versions with or without the LCD controller). This software would be most useful if it already had things like baud rate, stepper motor counts, and other variables set to allow basic calibration.

The third “default” I would suggest is a “standardized” spool of high quality filament to be used for machine calibration. Controlling the media used in calibration would eliminate a number of variables known to affect machine operation, printing issues, and print quality.

The last of the defaults I would suggest is a standardized set of printing examples that would allow the user to calibrate and troubleshoot their Rostock MAX. These examples might start with those already detailed in the present Assembly Manual and progress through more-demanding designs.

Establishing these “defaults” would control the variables of machine build, software installation, filament quality, and printing calibration. I believe this would allow for easier troubleshooting when users had problems, and would allow users to turn their attention to what they bought their Rostock MAX for – creating. Once a user knew that her/his machine could "do the basics," he/she might then choose to customize their setup to meet their individual needs.

These “defaults” would need to be available from a single source – preferably a page on the SeeMeCNC web site, or a locked thread on this forum. The defaults would also need to be updated when changes in Rostock MAX parts are made, or when software updates are available. And, of course, there would need to be a clearly documented “change log” that would note when “defaults” were changed. But such is the beauty of the internet – updates can be easily done once the basic framework is established.

I am aware that much of what I have noted above has appeared in various forms in other threads on this forum. I chose to post this thread to share my impressions and reinforce the thoughts of others.

Bill

[cambo3d]

i wasn't too impressed either with how the build went but im gonna continue to keep going and get this thing working properly.

my final review will be when the rostock is completely ready for print.

[Eaglezsoar]

I am in the process of building and I agree that these defaults need to be created. Firmware is created but you need to know where to find it.
Instructions are in different places and different authors, I could go on and on. I await Cambo3d's review and I hope that those in charge take
heed of what he has to say and make some changes.

[jestah]

I really feel what your saying as well.... I have had my kit for far longer than I expected to get running but can see the light at the end of the tunnel...

I have spent more time searching for soloutions and work around than actual building. Finally printed my first full calibration cube today so starting to feel a little less cranky.

Right now im up to trying to get my software running well with the printer so starting to search for the best settings. Once I have a few decent prints under my belt Ill be happy to post up my settings!

[JohnStack]

I think all of this experience closely parallels the typical technology adoption model - where many of us are the early majority - driven to refine and democratize everyone's key experiences and learning.

Geneb worked on the instruction guide - and now, without a build weekend, I believe I could take what he's written and build one. Many of you were pioneers (or early adopters) before that happened. I thought an FAQ would be helpful - but in the end, I'm not sure it would work except as a troubleshooting database. This forum is ok but even I can't find the right results on a google search.

So, how to move forward? A standard print setup in Slic3r for a few prints, the excellent setup for Kiss by Geneb(?)...

To me, the challenge here is not just setup but also the different characteristics of objects - like me learning that I had to put mouse ears on flat prints - and then finding out that you need at least a 15mm diameter - and then finding out that I was better off rotating it 90 degrees... arghhhh...and then finding out that my bed - even with glass - doesn't quite heat evenly...

Maybe it's just me - but I like the experimentation - but only to a certain point - when I really want to get something done....

Maybe a hangout in Google to go through all of this - synchronously - and come up with a plan....

I'm open to any time this week if anyone wants to organize an effort to get over the hump....

[geneb]

I think it was Mhackney that published his KISS settings.

g.

[Jimustanguitar]

Please document everything that you're learning and post it for the good of the group! The whole idea of open source is community participation and contribution toward a common goal. We are the alpha and beta testers as well as the end users and consumers. We take the pictures and make the manuals and find out what works and what doesn't. The forum will be your best friend, and has a ton more information than a manual ever could.



I built my Rostock using the v1 manual back in March. There were some instruction gaps (Onyx bed, LCD kit, LCD to RAMBo adapter, etc) but I was able to figure them all out without feeling shorted or anything. The things that I didn't have instructions for were all upgrades that I was excited to have and would have spent extra money on anyway. I think the actual build process is pretty cut and dry, especially with the new v2 manual nearing completion.

As far as the software goes, this has changed a lot lately. Originally, the manual just described which programs to go and get. Then there was a "Rosoft Bundle" on the seeme page. Then there was a download section of the website. Whichever you choose, I think the lesson here is that it's all open source and you can use whatever you want. There isn't a single right answer, and it's all constantly changing, so it's hard for the manufacturer to say "use this one". You should try what you want and use what you like. Forum posts with other people's settings are helpful here, and they're out there to be found.

For calibration, the forum has better information than a manual can provide. There are more than a dozen things to tweak, and everybody's setting will be a bit different. That's why a few of us have compiled and posted threads with the steps we took. Again, I don't know that there's a single right answer here.

[626Pilot]

If I had it to do over again, I would have just bought it pre-assembled, and pre-CALIBRATED. There are one or two people who do that. The build is difficult and extremely time-consuming, and when you're all done with that, you say to yourself, "Hooray! Now I can actually use this thing."

Nope! Welcome to 30 hours of calibration. A few hundred bucks extra to get a machine that's fully built and that doesn't put me through the root canal of calibration is MORE than a fair trade.

Also, I'd ship this thing with a better hot-end. The one it comes with is not well suited to PLA.

[geneb]

30 hours of calibration?

Horsepucky.

g.

[Eaglezsoar]

30 hours of calibration?

Horsepucky.

g.

I had to look up that word and here is what the dictionary says:

1. Horsepucky Cleaned up version of "horseshit," meaning bullshit, garbage, lies or just in-general kinds of crap.

Never too old to learn!

[Batteau62]

My dad called it...

[Jimustanguitar]

30 hours of calibration?
Horsepucky.
g.

I think there's a hugely varied view of what good calibration looks like. When should you call it good enough? There isn't one definitive test object, or one standard set of Slic3r settings, etc... Some people also misunderstand what a quality print looks like. It's not going to look like a perfectly molded plastic part, and that can't be blamed on calibration. It's the nature of the manufacturing process.

You can do a standard setup with a slight twist towards bridging, or tweaks for overhangs, small parts that get too hot, fast prints, high definition, or whatever you want. Unfortunately there isn't one universal "right" setting that will work for for different prints.

I've kind of settled on trying it out first, and tweaking something later when you know whether it worked or not. I spent a lot of time fighting calibration and not making things, and I wish I decided to just settle and print for the fun of it sooner.

[626Pilot]

30 hours of calibration?

Horsepucky.

g.

Did you know there's a musician who thinks that anyone can sit down in front of a piano for the first time, and just start playing, and that the necessity of learning to play it is entirely an illusion?

I've never done anything remotely like building and calibrating a 3D printer before. 30 hours is charitable. If you think this stuff is so simple, you must have forgot all the head-scratching, false starts, and the sheer volume of mistakes a person has to make just to learn enough to be halfway decent at this.

[geneb]

Neither have I. The Rostock MAX was the first printer I'd ever been in the physical presence of, let alone built.

It's not that hard to calibrate.

It's a lot trickier to tease the right settings out of the various slicing programs.

g.

[Jimustanguitar]

Neither have I. The Rostock MAX was the first printer I'd ever been in the physical presence of, let alone built.

It's not that hard to calibrate.

It's a lot trickier to tease the right settings out of the various slicing programs.

g.

The difference between the two is kind of blurry. Which calibration items would you list as "printer" items and which would you list as "slicer"? Several settings can be tweaked in more than one place, you know?

[geneb]

To me "calibration" is a mechanical and firmware (because it controls the mechanics) issue.

Once the machine is "trued up", it's calibrated. Calibration cannot take being fed crap data into account.

g.

[626Pilot]

It's not that hard to calibrate.

Says the guy who knows so much about Rostock printers that he literally wrote the book on them. You have something like eight months more experience than me. Maybe more.

Once the machine is "trued up", it's calibrated. Calibration cannot take being fed crap data into account.

I don't think it's calibrated until it prints something presentable. I can get the machine mechanically in shape and set the EEPROM values correctly, and it can still drool all over the place and rip up the piece before it's even done printing the first layer. Why call it the "calibration cube" otherwise?

[geneb]

There are people out there that make me look dumber than a sack of hammers when it comes to 3D printers. I'm far from the expert everyone seems to think I am.

The assembly manual I wrote is just that - an assembly manual. It will get you to a properly functioning printer. I got my first 3D printer from the SeeMeCNC Indiegogo campaign in December of '12, so at the MOST, I've got 7 months on you.

A 3D printer is just like any other CNC machine - it's going to do _exactly_ what you tell it to do. Unfortunately, there's a LOT of slicing parameters that have to be at least in the right ball park in order to get a good print. That's the kind of thing that that exists outside what I consider to be "calibration".

It's called a "calibration cube" because that's what the guy that designed the part called it. I chose it because it's a very simple first object to print.

g.