Reprap Standard 2012

Ok well I'm tired of arguing so I decided to go ahead a make a page in the wiki. If you have an opinion about the direction of the standard, please post it here and we'll have a reasoned and objective discussion (hopefully).

What's on the wiki pages is just placeholders for now, but i expect it to be final by the end of April. That should give us enough time to decide, generally.

I'd like to hear thoughts about a time frame for the standard, too. I think 1 year is good since it takes a while for a new design to get enough users in order to get constructive feedback about it's ultimate pros/cons.

That all looks pretty good to me as a reasonable starting combination of cost/quality/tried and true/most common. (at least based upon forum comments I've seen over the last few months)

What would make this perfect is if once we agree on a 'basic" model we can include a complete set of files for it in a single place. All the stl's, marlin (configured for the specific ramps, hotend, etc) pronterface and all it's required python dependencies, etc.

Need to make an obvious note that while the decided upon model is a sort of "most common" model there are lots of variations around which can work equally well. Ideally we do this in a way that doesn't step on vendors toes too much.

---

www.Fablicator.com

Although I agree with all the items you choose, I don't think the term "Standard" is in anyway correct. It is not a standard but rather a recommended build. A standard is something that is to be cohered to and shapes the way the next generation is to head. A possible better title would be "Community recommended RepRap" or something similar.

I have to say whenever I hear the term "Standard" it makes me cringe. It is usually followed by "stop trying to reinvent the wheel" and I always think if we thought that way we would all be in our horse drawn carriages on our wood and metal wheels.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Andrew Diehl Wrote:
-------------------------------------------------------

> What would make this perfect is if once we agree
> on a 'basic" model we can include a complete set
> of files for it in a single place. All the stl's,
> marlin (configured for the specific ramps, hotend,
> etc) pronterface and all it's required python
> dependencies, etc.

I think we can start doing that after April, once we've got the list nailed down. It's a great idea they'll help newbies greatly.

> Need to make an obvious note that while the
> decided upon model is a sort of "most common"
> model there are lots of variations around which
> can work equally well. Ideally we do this in a way
> that doesn't step on vendors toes too much.

Yeah and we can attach the discussion here onto the wiki page, after April, so if anybody wants to see the reasons for the selections there will be an open record.

----

First and foremost, I think for anything to be included in a standard, it has to have a wiki entry already. for example, I would have put the greg's extruder in for the cold end, but it's not on the wiki. Wade's is tried and true, and we can build off of the instructions that are already there.

Second, many buyers of a "reprap" kit end up coming here for support. That being the case, maybe we should concentrate on components that are easiest for us to support, not cheapest or easiest to build (at least not for this baseline standard). That being said, almost everything is cheaper and easier to make today than it was two years ago.

To have any sort of impact, it needs to be linked on the front page of the wiki. Or no one will see it, especially noobs (not meant in a derogatory way) who don't know you have to search for every single page on the wiki, as it is not linked in any coherent way.

I don't think there's anything wrong with standard as the definition of what we're doing. Reprap.org is the community that develops reprap printers, so any standard would be a community recommended standard.

However, It's just a word, and we've spent a lot of time worrying about the connotations of the word instead of working towards the goal. Once we pick the parts, we can worry about what it's called.

I'll explain some of my choices:

Slic3r over skeinforge because of the ease of getting good settings quickly. Skeinforge allows more precise slicing, but for most new users this won't matter.

Printrun is widely accepted and cross platform, with lots of active development

wade's for the wiki page. maybe if we get a greg's page up by the end of April we can use that instead.

j-head because it's been generally reliable. it's also resistor based, groove-mount compatible, removes common failure mode of the gear extruder nozzle derived designs, and is available in 3 or 1.75 mm filament sizes and multiple orifice sizes

marlin because it's what I use. I tried sprinter, and teacup before that. I get the best results with marlin, but other opinions here would be welcome, of course.

Ramps- somebody else will have to make this argument. I use gen6 and love it, but ramps is also widely used.

Prusa v1- it's well established, and i think most kits are still prusa v1, not v2. I'd prefer Mendelmax, but it's still relatively young. I think mendelmax would be a bigger contender next year.

Ramps because it is the most upgrade friendly (fan and/or dual extruder and/or heated bed, etc) and it allows for mistakes to be made that only cost the price of a Pololu driver not an entire board.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Ramps also appears to be the most widely available, and is more modular (less likely to fry everything at once for a newb)

I think marlin should be the firmware. The look ahead makes all the difference in the world as far as print quality/speed, and it's been working reliably (for me anyway) for a couple months.

Ramps/Marlin also are the only combo to have preliminary support for dual extruders (ie: expandability)

How about just calling it the community recommended printer?

---

www.Fablicator.com

Sublime Wrote:
-------------------------------------------------------
> Ramps because it is the most upgrade friendly (fan
> and/or dual extruder and/or heated bed, etc) and
> it allows for mistakes to be made that only cost
> the price of a Pololu driver not an entire board.

Great point. Since electronics are usually the most expensive component, you're likely to use the same electronics with a number of printer iterations over the years. Upgradeability and longevity are important.

Ramps, because it's the only one that has PTC fuses to protect from overcurrent faults, ie short circuits, in the heated bed and hot end. Which is why I consider Sanguinololu, Gen6 and Gen7 to be a bit dangerous.

droftarts Wrote:
-------------------------------------------------------
> Ramps, because it's the only one that has PTC
> fuses to protect from overcurrent faults, ie short
> circuits, in the heated bed and hot end. Which is
> why I consider Sanguinololu, Gen6 and Gen7 to be a
> bit dangerous.

ok I'm all for not burning down peoples houses.

So one of the points somone made on the 'Saving Reprap' thread, which this seems to be a continuation of, is that a standard would help someone new get right into printing without all the issues of sorting through hardware/software. Help the newbie get going right away. The topic links you added to this page are mostly the favorites of late, they also are at the root of all the newbie's questions. Mainly because newbies read these threads and chosen these same platforms. This simply puts them all on one page.

The problem is that with all this quick advise folks arn't really researching what it takes to get one of these going. By the way I'm not talking about engineering types who have a nack for technology. I'm talking about the folks who list these same components, show a photo of their latest blob, and ask what do I do to fix it?

There is a need for something, however this isn't enough.

Maybe it time for 'Reprap for Dummies'

Small steps.

Some people won't research what it takes to get one going no matter what. At least this way they might get something up to date and workable with as little hassle (to us and them) as possible.

---

www.Fablicator.com

Well done. I had some internal struggle about Prusa V1 vs V2 but I think your choices are reasonable considering the documentation is much better for V1 right now, especially the visual build manual etc.

I've only been involved with reprap for a couple of years, but I think I have bought and used most of the commercial hot-ends. I agree that the J-head design is probably the best to recommend to newcomers to have the greatest likelihood of success. I have two, and several Makergear hot-ends that I had a surprising lot of fun assembling (really!). I especially like the groovemount approach for attaching to the extruder that both share, and it seems that the Printrbot hot-end also shares.

I want to put in another vote for a Greg's variant of extruder over Wade's, primarily because if you choose the right Greg's you don't need an adapter plate for the groovemount. I have poor wiki-fu, but if someone can start a wiki page with the right initial formatting I will assist in populating it with assembly instructions, BOM and photos. One challenge will be getting agreement on which variant to recommend. I've been pleased with most, and suggest we stick with Greg Frost's, the supports under the bearing cut outs and under the idler holder have resulted in less post printing cleanup for me.

Wow. What a lot of misconception about electronics. RAMPS is currently the most complex piece of electronics, so it's also the most expensive one. The fuses thing on a Gen7 is simply not neccessary, because PC power supplies have such a fuse already built into the PSU it's self. Regarding repair after a fault: Sanguinololu, Gen7 and RAMPS all use the exactly same Pololu/StepStick drivers, which all can be replaced one by one. No advantage of RAMPS here. Last point for now: rebuilding a Gen7 or a Sanguinololu is cheaper than just the plug-in board making RAMPS, so this argument of keeping the Arduino (instead of just the ATmega) is more than questionable.

There are use cases where a RAMPS is advised, for example if one really wants a second extruder, but pretty much none of the reasons in this thread are of this type.

---

---

Regarding "standards". Defining a standard is an excellent idea! I fully agree with Sublime a standard doesn't describe a prefered choice, but a set of properties a piece of electronics or mechanics has to feature. Manifesting preferences too much supports monopolies/reduces choices, which hobbles development and is generally considered to be unwanted in any economics.

For example, a standard could be like:

- support for at least 4 bipolar stepper motors, 1.5 Ampére

- at least 1/8 microstepping

- support for one extruder and one heated bed ( = 2x thermistor, 2x heater)

- at least one firmware available which accepts G-code (firmware feature set TBD)

- usable/configurable over USB

Most RepRap electronics comply to the above set already and such electronics can easily be exchanged between all RepRap mechanis, being it Sells Mendel, Prusa Mendel, MendelMax, Huxley or whatever the next piece in fashion will be.

The later is important to newbies: "Take an electronics complying with RepRap electronics standard 1.0 and be sure it'll work for your mechanics." That's the message we should get across.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

A PC PSU may give 16-20A on the 12V rail. Its overload trip point will be higher than that. Are all the connectors, PCB tracks and downstream wires rated for that? I think not, so a fuse is necessary to be considered safe.

Edited 1 time(s). Last edit at 03/21/2012 11:11AM by nophead.

---

[www.hydraraptor.blogspot.com]

Traumflug Wrote:
-------------------------------------------------------
> The later is important to newbies: "Take an
> electronics complying with RepRap electronics
> standard 1.0 and be sure it'll work for your
> mechanics." That's the message we should get
> across.

I'm no electronics guru so the nuts and bolts discussion is beyond me, but just the bare capabilities are only one aspect of the electronics.

I'll comment that with gen 6 I had to wire up my steppers to the molex connectors, which was finicky work without a crimper. However, I've plugged and unplugged the connectors 50 times at least, so if i had to unscrew all those connections it would have wasted a lot of my time, in the long run.

on the other hand, which other electronics use the same connector? If I wanted to switch electronics I'd have to cut them off, which would be hard for sentimental reasons :-) Interchangeability might be a better path to follow?

I wasn't sure if I should even include the electronics category on the wiki page, since it will probably be one of the most contentious. Standard electronics would be fine if they were the same shape and had the same connectors like PC components. If that were the case one set of wiki instructions would apply to every type of electronics. That is my desire, with this standard: to get as many people as possible concentrating their wiki-editing efforts on a few key pages, so we can have a clear set of assembly instructions. I imagine, for instance, that the "installing firmware" page would just use the '2012 standard' electronics as just an example, just as the old Mendel instructions assume you are using gen3.

All looks like a good starting point to me.

Mendel90 could also potentially be an easier starting point, but I have not built one yet.
Prusa V1 is good as it's reasonably easy to reuse and upgrade most parts into being a MendelMax - I still think that's a desirable upgrade path.

Marlin can be a little tricky for starting out, setting up and tuning, we need to make that easier for first Prusa machines.(running at reasonable speed)

Do keep the electronics in there, when I first started I was shocked and amazed by the Gen3 electronics - A) - Being surface-mount and b)- almost everyone blows up the extruder board and C) - only 2x stepping - very noisy D) - not having the FTDI onboard E) - the list goes on.

We are going to see lots more 'low-cost' electronics this year, that will be 'compatible' with Marlin/Sprinter but will be very easy to blow up and not easy to repair yourself.
Stepsticks/Pololu's are not perfect but give so much more room for error and repair than an all-in-one surface-mount design in my view, we are a little way off needing an ultra low-cost integrated design just yet. Keep it modular, easy, and if a little more expensive so be it.

RAMPS 1.3/1.4 and GEN7 at least give the new user a fighting chance when they first get onboard with RepRap

I wish the Sanguninloulo PCB was ~20mm bigger in both directions, just to make it a little easier to build and fit connectors into etc. and the point about on-board fuses or polyfuses is very valid, you need them, if they don't come on your electronics, fit them off-board.

We should also not be using ATX power supplies, many are the source of machines locking-up, freezing and generally acting odd, yes you can argue this point, but I'll argue back.

---

[richrap.blogspot.com]

Quote
richrap
yes you can argue this [no ATX PSUs] point

Indeed, as they usually work just fine and come with all the required connectors already. If you use an electronics which is designed to be driven by an ATX PSU, of course.

Quote
Buback
which other electronics use the same connector?

Motor connectors agree on the (4 pin) Molex KK100 already. Heater connectors are a bit more tricky, as pretty much all of them are intentionally undersized. 12A going through a pair of pins specified for 1A to 2A each, depending on the data sheet you find. Gen7 uses bigger ones, but still not within specified range.

Quote

Are all the connectors, PCB tracks and downstream wires rated for that [20A]?

No, they aren't, mostly for cost reasons. They aren't even rated for what's intended to go through there and just yesterday there was a case where a heated bed connector started to melt and fume without receiving overcurrent. So, any fuse can pretty much protect against shorts only. At least that's my understanding so far.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Quote

Heater connectors are a bit more tricky, as pretty much all of them are intentionally undersized.

Intentionally? I though it was just oversight. Why would anybody intentionally design something to fail? Connectors should be overrated (like most other components) for reliability. Especially on a Reprap where connectors fail well below their specification due to vibration. They need to be screw terminals, or screwed together, rather than simple friction wipe contacts.

Quote

They aren't even rated for what's intended to go through there and just yesterday there was a case where a heated bed connector started to melt and fume without receiving overcurrent. So, any fuse can pretty much protect against shorts only. At least that's my understanding so far.

Fuses give protection for over current. So if it was say a connector rated for 5A then it should have been protected with a 5A fuse. Then when you pull 12A thorough it the fuse would have blown instead of melting the connector and potentially causing a fire.

Most of the electronics would not pass any safety regulations because of this. That means anybody selling complete machines breaks the law in the EU and US to my knowledge and probably most other places in the world as well.

When people build kits it is their own risk but Reprap will get bad press when somebody inevitably burns their house down.

Edited 1 time(s). Last edit at 03/22/2012 08:22AM by nophead.

---

[www.hydraraptor.blogspot.com]

Quote

Intentionally? I though it was just oversight.

Well, I have to assume intention, as the connectors were changed a number of times and designers capable to use correctly rated capacitors should also be able to use correctly rated connectors, right?

That's how lots of RepRap stuff is designed. Don't care about specifications, but keep the cheapest parts which work for most people. Using bigger, more costly connectors is pointless, as 90% of all RepRappers will solely decide by price and feature set. Being within specification isn't recognized as a feature.

Quote

Most of the electronics would not pass any safety regulations because of this. That means anybody selling complete machines breaks the law in the EU and US to my knowledge and probably most other places in the world as well.

Entirely true and that's the reason why I don't sell soldered Gen7s. Again, however, people buy what looks convenient and I currently see no chance for a solid design to compete against underspec'd, but cheaper and reasonably working designs.

Another thing is EMF protection. Very likely all RepRap electronics radiate more electromagnetic noise than allowed by law, but did you ever see a RepRap electronics mounted inside a metal housing to countermeasure this? I didn't.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Quote

. Being within specification isn't recognized as a feature.

Well this thread is about a standard reference design. I think that should at least be guaranteed to work in all cases and be safe, not something with known design flaws. I also think 90% of people would pay a few dollars more for something reliable. The mass exodus from PLA bushings to LM8UU bearings suggests that is the case.

When you consider the running cost of the machine in terms of the plastic it uses then it doesn't make sense to save a few dollars on the machine. One failed build can easily cost more than a connector, fuse or a logic drive protected MOSFET.

I agree it is unlikely a machine will pass EMC regs, but that is unlikely to cause a problem in a domestic environment unless you live next door to a radio ham, it is certainly less important than safety and it isn't bad enough to stop the machine working.

---

[www.hydraraptor.blogspot.com]

How about alternatives to the jhead that are easier to construct by hand?

edit: although, the bulk of the instructions for a hot end, in terms of assembling the printer, is how to properly wire up the resistor and thermistor to the heater block, and how to mount the hot end to the cold end. Should it matter if it is easy to make or not?

Edited 1 time(s). Last edit at 03/23/2012 12:16PM by Buback.

I wouldn't recommend any threaded-rod based design to someone building their first machine. Mendelmax all the way. Set it and forget it, no tweaking. Sure it's light on documentation.... because its like an erector set. It only fits one way and if it fits it's straight and correct. The rest is just regular reprap x/y/z that doesn't require explanation beyond what's already out there.

I'd say recommend the model that requires the least docs, not the one that has the most (to a beginner).

How available worldwide are the extrusions? Reprap is supposed to use common vitamins found everywhere.

---

[www.hydraraptor.blogspot.com]

I also don't see how MendelMax requires fewer documentation. If there is less, this is not exactly a bonus.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Because its just simple. It only fits together one way. No adjustment necessary, no jigs, no complex part assembly. It goes together the way it looks like it goes together, was my point.

Why make a standard for a bleeding edge project based on an old v1 design?

I submit for your consideration the next generation RepRap, combining the best of Prusa and the MendelMax in one: the MiniMax.

MiniMax Photos

MiniMax Bill of Materials

Aleph Objects, Inc. thingiverse parts

Thanks,

-Jeff Moe
LulzBot