rostock max or something else

[philjherman]

Hello everyone!
I wanted to ask all of you what your thoughts are on the rostock max printer.

I am currently looking into a 3d printer and the rostock max is one of my choices. Unfortuneately in my research I have found that this machine has plenty of issues as well. I am not looking for a plug and play machine by any means.

Could some of you who own the rostock max please post your thoughts on the machine as well as photos of any of your successful prints?

Do you spend most of your time printing, calibrating or repairing?

Thanks in advance for any replies. I really do like the rostock design and I hope to hear some success stories... But if there are failures too I would love to hear about that too!!

[cambo3d]

my suggestion would be to read some build logs and see prints in the "what are you making" section.

[Jimustanguitar]

I didn't write much about the process, but I took some pictures of my build. http://imgur.com/a/PWPR3#0

I'm very pleased with the Rostock Max. I have a friend who bought a much more expensive 3D Touch from Cubify ($3500-$4000 range) and he's jealous of my $999 Rostock. He doesn't have a heated bed, has a smaller build volume, and there are a lot more moving parts to adjust and tweak between each print. He's also had a much less personal experience than I have, both with the manufacturer and the machine itself. The SeeMeCNC guys are on the spot with help and advice (they frequent this forum) and they want nothing more than to help and make sure that you're satisfied. I feel that I know my machine very well too, because I built it. My friend with the Cubify doesn't feel this way at all, it's a pre-assembled mess that he's always hoping doesn't break and need to be shipped out for service.

If you're not scared of some assembly, and know how to use a soldering iron, I would definitely invest in the Rostock.

[DaGroundZero]

I really enjoy my Rostock Max. I've had a few bumps along the way getting to where I am now, but the quality of my prints are getting better and better as I learn more about the software setting side. If you don't mind putting together kits or doing a bit of assembly, then I would suggest it. If you don't like or aren't good at putting together Ikea stuff, then I would look for more of an off the shelf printer. At some point, it will break down, so you will be doing some work on it anyhow.

I've had my printing for a bit over 3 weeks now, and I have friends that can't believe how well the prints are coming off the machine. They have had their printers for much longer, and aren't at the quality I have just a few weeks into it.

[JohnStack]

I have had my machine up for a week and it has been printing every day for 3 - 4 - 5 hours.

I haven't had to make adjustments - so cumulative time 24 - 30 hours, pretty damned good!

My challenges have been around getting prints to stick to the heated bed & software. Both of those challenges affect everyone who has a additive printer. I solved my problem last night with ABS Juice (small pieces of ABS diluted in acetone).

The support from SeeMeCNC and this forum are incredible so what are you waiting for?

[msimpson]

It all depends on what you want to get out of it. If you want a printer that is consitant, maintenance free and just plane makes good prints. The stay far away from the Rostock Max or any Delta printer for that matter.

On the other hand you want to tinker, have tons of time on your hands or want to learn about delta printer, then the Rostock is a good start.

I purchased mine for a series of articles I am writing of Servo Magazin, and a book that I will be writing. I own (as of this moment) five 3D printers. Until recently All gave me better prints than the Rostock. After replacing the extruder/bowden cable/hot-end with a direct driver extruder I now can get pretty close. I also had to add a second power supply to get the Onyx heated bed up to the 100c range.

Were these changes worth it? Yes. Now I have a 3D printer that I can create parts with and write about.

Is this my best 3D printer? No, but when someone comes to my lab, they walk past the two CNC machines, Epilog laser, and head directly toward the Rostock and say, "Can I see this one work".

[geneb]

I call shenanigans.

The Rostock MAX was not only my first 3D printer, but the first 3D printer I'd even been in the physical presence of.

I will concede that sanding some of the parts is a pain in the ass.

That being said, calibrating it was pretty straightforward and was accomplished in two or three hours time, spread across a few days. (I have little free time these days)

I've since put about 60 hours on the machine and the only adjustments I need to make is when I change the boro glass out for a boro glass disc with window tint on it (ABS sticks nicely to it). That is accomplished by tweaking the Z height to account for the tint thickness.

I print with both ABS and PLA with _zero_ issues.

I strongly suspect my ease of use has more to do with a lack of pre-conceived notions about how it "should" work than to do with how well my brain works.

I see people having problems with the bowden, extruder drive, etc, and I can't for the life of me figure out why. I'm no mechanical genius by any means and I've had _none_ of these issues once I learned what slicing parameter did what.

Putting the extruder drive on the effector end completely defeats the whole point of a delta - high speed printing.

Delta printers ARE NOT HARD TO ADJUST. Hell, take a look at the BerryBot. That thing is a total from-scratch build and he's doing amazing things with it.

I understand that the delta configuration is the new kid on the block, but it's silly to blame the tool just because it's young.

g.

[msimpson]

What you say may be true , but I spent hundreds of hours trying to get my rostock to put out a quality print. Nothing I could do could generate even an acceptable print. I got little help other than, try this, and try that.

Later I found that some of the nozzles may not truly be .5 and could be as high as .9. This could definatly be the cuplrate, but this goes back to qaulity control. As for the eated bed. The Onyx is not compatible with the Rostocks power supply if you cant get past 75c.


Delta Printer Problem areas:
Acceleration control is on the motors not on the actual axis. This means it can smooth motor movement, but little else. It cannot help you reduce high speed vibrations.
Backlash control (Non Existant)
I dont care what you say bed calibration is much more compex then that of a traditional pinter.

Bottom Line if you are asking yourself "Rostock or something else" you better go look at the something else.

[truenorthtrader]

If you have TIME and PATIENCE and have a love for making things I would suggest it.

It's a complex machine that takes alot of tweaking. It's not something you can hurry.
Sometimes as in my case I got ratty prints even recently just because 1 number was not set properly, so unless you learn everything it can be frustrating.
Again time and patience. I had lots of frustration before thinking "this is way over my head" but was always helped by this forum and Directly from SeemeCNC and very quickly too!

At times I said this was a bad idea it's not working no matter what I do....but in the end I learned it was because I didn't KNOW what to do. Once I did, I solved every problem with the help of this forum and the tech support.

The forum here is amazing, the guys at SeemeCNC are the best people you can imagine, they always go out of their way and have for me in the past to make sure I get things right and replace parts that need replacing, and they've sent me updated pieces as well.
Few companies will do that!

I'm not familiar with other pre built machines but I would imagine since they are machines with basic similar parts, they need to be adjusted regularly also if not more often.

The only qualms I have is with the Sanding of some parts which will put tolerance and accuracy off if not done properly and the Top guide for the Belts which seem to go on one side or the other of the bearing I can't get it straight. I had an idea for a printed cap with the same gears as the belt to keep it centered I hope to have time to work on making one with the Rostock itself.

Here's the pieces I'm printing now and a recent iphone gear case, no sanding or anything done to the parts

There is a learning curve but I did it with patience and not giving up and it paid off.

[geneb]

That's a neat phone case.

If you have the belts tensioned even reasonably, there's no "backlash" in the system at all.

The only way I could possibly conceive of someone taking "hundreds of hours" to adjust a Rostock MAX is if you were using your nose, both elbows and the neighbors cat. There simply aren't that many things that can go wrong on this thing.

"Bed calibration" consists of tweaking ONE parameter. Assuming of course you've got a flat bed (all three towers perfectly perpendicular to the top of the base), the only thing you're tuning for is to get the head to make a flat traverse across the build volume.

Bottom line is this - if you like the look of the Rostock MAX, buy it. It's no more complex than most other 3D printer kits on the market. Ignore the doubters, they've obviously pushing the Cartesianist Agenda.

It's a poor craftsman indeed that blames his tools.

g.

[Polygonhell]

My 2c on this, is the Max is a good printer, as is.

This is all just my opinion, I own 4 printers and have access to at least 3 others including a Beta MendelMax 2.
There are IMO enough additional difficulties in configuring a Rostock style printer that I would not recommend it as a first build. Nothing is overtly difficult, but if you've never gone through the process of "debugging" a 3D printer additional variables increase the difficulty.

Steve will hate me for saying this, but the weak point on the MAX is the extruder and hotend combo. The new nozzle is a huge improvement, but it's still not in the ballpark of a JHead, and the although most of the extruder is over engineered it's difficult to assemble correctly, there's too much space between the output from the drive wheels and the Bowden tube entry and the drive wheels don't have enough bite on the filament.
The Bowden tube just makes these issues worse. There are plenty of printers with Bowden tubes putting out high quality prints including many MAX's, Ultimakers are still considered to produce some of the highest quality prints out there and they are all Bowden based.

Secondary issues include the care required to assemble the universal joints (though apparently since remedied) and the relative ease the belts can be damaged.

Having said all of that I still think the MAX is a great kit, and by far the most complete I've ever seen.
FWIW My MAX produces more consistent better aligned layers than any other printer I have access to, and that's a large portion of print quality right there. I thing this is largely just a function of the Z being belt driven, but it's worth noting.

I've seen a lot of friends new to printing struggle to get decent prints out of variety of machines, I can usually talk them through it in a day, but since pretty much every single problem can be either mechanical, electrical or software debugging as a newbie can be daunting on any machine, the additional potential problems on rostock like machines are just enough to push some people over the edge.

I'm s software engineer by day and IMO the Software for Delta printers still IMO needs a lot of work, and I think real solutions require a faster processor in the controller, which is one of the reasons I'm experimenting with embedded ARM boards as a motor controller.

[geneb]

Would a Raspberry Pi be fast enough do to the job?

g.

[Polygonhell]

Would a Raspberry Pi be fast enough do to the job?

g.

Yes it's relatively speaking very fast, but there aren't enough GPIO pins exposed on the board.
There are also issues with all of those micro PROCESSORS (I'll explain the capitalization later), linux is a less that ideal environment to write a motorcontroller in, it's difficult to meet the latency requirements to get a clean step signal, and some of the hardware is undocumented and supported by vendor supplied binary libraries so no OS isn't really an option.

Things that consider themselves "Microcontrollers" as opposed to Micro processors are generally a lot easier to work with from a software standpoint for hard real time systems.

My current work is with a 100MHz LPC1768 similar to the chip used on the smoothie boards, it's about 8x faster than the ATMega chip on the RAMBO board, the board I'm using has Ethernet, USB, SDCard, plenty of GPIO and a 2.8 inch 320x240 touchscreen for a total cost of $40, the motor drivers are currently sitting on a breadboard.

But nothing I'm doing is really tied to the hardware in anything but superficial ways, and very little of it is short term.

[easybeans]

My input... I had ZERO experience with 3D printers, machining, or the likes... I didn't even know 3D printers existed up until a week before I placed an order for my Rostock. So I went into this fairly blind - with no expectations, hell, I wasn't even sure this thing would work when I finished it. But after ~10 hours of building following Geneb's manual and the forums whenever the manual didn't apply (I have a newer kit) I turned it on, sliced my first calibration cube and started the job. Guess what... It worked.

Would I recommend the Rostock Max? You betcha. Do I have any experience with other printers to base it on? Nope. I know my prints look good and I'm getting better the more I learn... But it was totally functional right outta the gate.

I still have no idea what most of you people are talking about on here..........

[cambo3d]

My input... I had ZERO experience with 3D printers, machining, or the likes... I didn't even know 3D printers existed up until a week before I placed an order for my Rostock. So I went into this fairly blind - with no expectations, hell, I wasn't even sure this thing would work when I finished it. But after ~10 hours of building following Geneb's manual and the forums whenever the manual didn't apply (I have a newer kit) I turned it on, sliced my first calibration cube and started the job. Guess what... It worked.

Would I recommend the Rostock Max? You betcha. Do I have any experience with other printers to base it on? Nope. I know my prints look good and I'm getting better the more I learn... But it was totally functional right outta the gate.

I still have no idea what most of you people are talking about on here..........

funny you say this, "totally functional right outta the gate"
but on another post you state you have to use your "hotend to heat up the print surface faster", your definition of functional is probably different from mine.

I do agree with polygon, this was my first printer also, but im no novice to cnc machines. it took a lot of work to get it to print right.
a lot of it could have been alleviated by better documentation written by seemecnc and quality parts / quality control.

[easybeans]

You're very confrontational, Cambo. I didn't say I NEEDED to use the hot end. I said I use it to heat up the bed faster and to get higher temperatures (higher than some of the commonly recommended temperatures)... See the difference?

So yes, it is "totally functional right outta the gate"

[geneb]

Polygonhell, have you considered using the recent release of RiscOS on the Raspberry Pi? That's about as real-time as you're going to get on that device unless you write your own OS.

While I agree that there's not enough GPIO to brute-force stepper drives, but shouldn't someone be able to design a controller board that the Pi can talk to over SPI or I2C?

Easybeans, cambo3d isn't happy unless he's making snotty commentary from the sidelines. That's just how he is. (Frankly, I'd LOVE it if he'd produce a better manual - that means we all win! Until then, we get to bask in his sneering superiority.)

g.

[Polygonhell]

Polygonhell, have you considered using the recent release of RiscOS on the Raspberry Pi? That's about as real-time as you're going to get on that device unless you write your own OS.

While I agree that there's not enough GPIO to brute-force stepper drives, but shouldn't someone be able to design a controller board that the Pi can talk to over SPI or I2C?

Easybeans, cambo3d isn't happy unless he's making snotty commentary from the sidelines. That's just how he is. (Frankly, I'd LOVE it if he'd produce a better manual - that means we all win! Until then, we get to bask in his sneering superiority.)

g.

There are just much easier options.
If you wanted to run a Linux based arm board for motion control I'd use a Beaglebone, the PRU's let you work around many of the none realtime OS limitations.
I think right now the sweet spot in 100-400MHz Cortex M3/M4 microcontrollers, they are more than fast enough to put out Step train topping out any practical motor/driver combo, you can put together something with a touchscreen for not a lot more than the current arduino based solutions.
Small runs of boards aren't too expensive.

[harley573]

I'm an engineer and have plenty of experience with traditional subtractive cnc machines. I have worked with some that, upon arrival at our shop, took a factory technician longer to calibrate and set up properly than it took me to get decent prints out of my Rostock Max and this was my first 3D printer. I think a lot of people fail to realize that this platform is very young within the community and the community is very young compared to the rest of the CNC world. It is unrealistic to think that anyone is going to open a box, plug the thing in and get a perfect print without spending some major cheddar on the machine and even then you aren't guaranteed success. The Rostock Max kit is designed very well. It could use some upgrades to the design, however. I am seeing a lot of positive things from folks who have switched to the magnetic ball joint setup. I also think the extruder needs to be re-done. Someone likened it to trying to get precision by running a tool at the end of a bicycle brake cable and I thought that was a fairly accurate analogy. Pushing and pulling filament through 18" of friction inducing tube is not the best situation. Despite this, Werner Berry has managed to put together an excellent setup that seems to be reliable and that guy gets incredible print quality. I'm planning to follow his lead, although I like the idea of building a lightweight direct drive for the motion platform. If kept light, I don't think it will negate the advantages of the delta arrangement. One of the biggest problems that I have had has not been related to the printer at all. It has been software. I use Slicer and it does an okay job most of the time. It's not great, though, and it makes a lot of mistakes and does weird things from a G-Code perpective. What the heck, though, it's FREE! And, it keeps getting better with every new release. If you are new to 3D printing, get the Rostock Max, correct the issues mentioned above and have a blast watching a truly unique machine do incredible things. I've seen and researched everything available and I can say without a doubt, the Max is worth the time you will spend getting to know it. Nothing else looks anywhere near as slick, which was an important point for me. I wanted a 3D printer I didn't have to spend $40k to get but I didn't want it to look like a lump of crap I cobbled together with some spare parts in my garage. The Max doesn't. It is a machine you are proud to show off and something that you can demo for clients if, like me, you plan to use yours for commercial ventures as well as home use. It bothers me to think that someone is writing a book that will influence others when they have an obviously biased opinion against delta printers. Hopefully, more objectivity will be shown in the book and Servo articles than has been shown on these boards. Especially when the bias and opinions are polar opposite to 95% of the posters on this forum. I'm not trying to pick a fight, but I am suggesting that when most people seem to be able to achieve good prints fairly easily and one person can't, maybe the printer isn't to blame. Just my 2 pennies worth.

[cambo3d]

Polygonhell, have you considered using the recent release of RiscOS on the Raspberry Pi? That's about as real-time as you're going to get on that device unless you write your own OS.

While I agree that there's not enough GPIO to brute-force stepper drives, but shouldn't someone be able to design a controller board that the Pi can talk to over SPI or I2C?

Easybeans, cambo3d isn't happy unless he's making snotty commentary from the sidelines. That's just how he is. (Frankly, I'd LOVE it if he'd produce a better manual - that means we all win! Until then, we get to bask in his sneering superiority.)

g.

doesn't seem like mr. geneb is likes any honest comments about his manual either. so guess what gene, in order to make you happy, i stopped giving opinions on your manual. but it seems like you insist on making these comments about me.. grow up.

frankly, ID LOVE IT IF THE GUYS THAT BUILD AND DESIGN THE ROSTOCK MAX CREATE THE MANUAL.! and kept it up to date.

[geneb]

I love honest commentary. I've received a lot of great advice from a lot of great people.

cambo3d, you lack a fundamental understanding of how open source, open collaboration and the maker movement function. It's sad to see that when given the "put up or shut up" option, you pick "shut up", thus missing a great opportunity to make something better.

g.

[cambo3d]

I understand completely what open source means. Sure I could write it, If they paid me to do it and seeing that I paid money to buy this machine it should come with a manual.
Open source or not, it still should have a manual written and updated by the manufacturer.

[MSURunner]

FWIW, I am an experienced user of bowden style 3D printers. I use my Ultimaker in the classroom daily, and I love it. I also love the idea behind the Rostock, hence why we bought one as opposed to another Ultimaker. I also convinced two of the other schools in town to do so as well. SO, while I'm new on here, I'm definitely not new to 3D printing.

Build: One of the other teachers I convinced to buy a Rostock and I built our Rostock's together. We both teach drafting/intro-engineering. I work on my own cars, he works on his own race car. We're pretty mechanically inclined. We spent what seemed like forever building this thing, and had some expletives to share. We also realize that's part of the price of being on the edge of what's newest. SO, while the documentation and build was easier for the Ultimaker than the Rostock, I won't complain too loudly about that.

Product: Everything I've seen about the machining quality of things is spot on. The team does a good job with the product.

Design: Well, I'm with Polygonhell on the major flaw in the product. The extruder I can work with, but the thermodynamics on the hotend are not the most conducive to printing in PLA, which is an indicator, to me at least, that it's not the best design as PLA should print easily in a good hotend. Here's my thoughts on it. One: the new nozzle does place a lot of heat very close to the part. When I'm printing PLA I want that thing to solidify as quickly as possible. The proximity of the heated block to the part allows for greater radiant heat. This is exasperated by the inability to get a fan shroud effectively down to the part without serious rerouting of pieces. That's where many are just blowing a fan across the bed (inefficient and can cause problems in high speed printing). Second: Why is the heated block so large? I don't think the aluminum piece should continue up towards the PEEK as much as it does, and if it does need to be there structurally, it should have some fins on the radiant part to dissipate heat. Third: Why not make the insulator larger? This should help prevent heat conduction, but that still won't completely solve the problem. Fourth: The teflon tubing connecting directly to the nozzle and up into the cold zone is basically a heat highway. The more heralded hotends (E3D and Ultimaker) have no such thing. I understand that it allows for a modularity, but I would rather have a rock solid hotend in one filament size than a Swiss-Army Knife hotend that I'm trying to work around, which is further complicated by SeeMe no longer offering the 3 mm setup. Fifth: HEATER CARTRIDGE and ideally, thermocouple (not nearly as important as the cartridge). Again, looking at the more heralded hotends, cartridges are used. They allow for greater heating capabilities and consistencies and they don't have exposed wires that can effectively turn the hotend into a 12v power source.

Having said all of that, I still am holding out hope for my Rostock. I haven't gotten it to produce a solid PLA part yet, but I'm hoping in another hour or so the UM will be done printing the bracket/shroud setup for fixing this (small scale seemed to show a lot more promise). I think anytime someone is looking at a 3D printer <$15000, there needs to be an understanding of with the additional savings in purchasing a kit verses a Dimension or Stratsys, you can buy a hell of a lot of aspirin, beer, and plastic.

[Gr8Scott]

Polygonhell - Regarding the Raspberry Pi. If you put the OS issues aside for a bit... On the lack of pins front, there is a shield/board that is out there called the Guzunty Pi that essentially creates a bunch of programmable GPIO pins via CPLD technology. That would solve any potential issues with digital pins and you can run stepper drivers directly from the Guzunty Pi once the programmer puts together the needed core for that purpose. He has driven a single stepper motor directly from the Pi and Guzunty combo without stepper drivers as well at 5v already, so I know that he would likely be pretty close to driving multiple steppers from the same single board. One only needs a way to check temps and run switches to power the heaters and you have everything you need to build a 3d printer from that combo. Using I2C hardware should help with temperature checking etc given the lack of analog pins.

[img]http://raspi.tv/wp-content/uploads/2013 ... ropped.jpg[/img]

[img]https://github.com/Guzunty/Pi/wiki/Guzunty-anatomy[/img]

https://github.com/Guzunty/Pi/wiki

http://raspi.tv/2013/guzunty-pi-open-so ... spberry-pi

[lazerusrm]

Biggest problem with Rostock MAX after owning it for.... well since october 2012.

1.75mm filament. -- Never should have been used.

I cant tell you how many times i've had the extruder fold filament inside itself... CONSTANTLY. Maybe if i only printed 20mm/sec it would be reliable.

3MM is WAY better. NEVER a problem with folding.

Bowden tube fittings are weak, i kept breaking them... Industrial Grade all metal fittings fixed this issue so far.

Hotend... Sucks ass to build, very delicate. Heaters are not sufficient and burn out easily. Internal diameter not correct for 3mm filament.. tolerance needs to be taken up to prevent fat melt zone... my opinion

A Wiring loom kit with connectors would have been much appreciated for the hot end side of things.

Extruder.... It flexes during printing increasing the distance for retraction required. Though works pretty decent.