New Rostock MAX

Just a quick update again guys. Just popped off the first few injection molded parts right now. We started molding the Platform pieces out of glass-filled nyolon, super strong, and heat-resistant compared to ABS. Check out the thingiverse page and you can see the new pics. They are 2-pieces that screw together to capture the axles for the universal joints, we are using 1/8" axles instead of bolts as the axles for the u-joints. The threads on the screws would wear eventually, and the axles and machined joints will not wear in glass filled nylon forever!

No, I didn't watch it. Don't really need to (no offense intended) as I've already watched a few similar videos on the same basic premise.

Didn't mean to sound contentious, and I apologize if I came off that way. I was just reminding you that the precision increase has a concomitant cost, in the form of multiplied step frequency to achieve the same velocity.

Personally, I'm all for increased resolution and think your improvement is great.

Me, I was advised to use either T5 or MXL for my first build (in process) Prusa Mendel variant, chose T2.5 to increase the X/Y resolution. And once this printer is running, I'll be designing a second machine that uses acme screws instead of belts, which for my longer term purposes, is just plain better, speed reductions at given stepping aside. (and where I was reminded of velocity cost vs resolution, since these arduino-y microcontrollers are afaik, for all their availability and popularity, only able to go "so far" with their motor control clocking frequencies, no?)

I'll admit I'm fascinated by the Rostock design, (so star trekian) and I might consider building one down the road, but for me as a newbie to 3D printing, I'll be happy just getting my simple version running and for a variety of reasons will likely go with more horizontal expansion rather than vertical, at least at first.

xiando, no worries, I didn't read into it at all. You're exactly righta bout speed vs torque/stepping frequency etc... The RAMBo board has 1/8 step max microstepping, not the 1/16 that most people are running. So that helps with the microstepping and clock frequency you were talking about. Because we machine our own pulleys and parts, we can make whatever the combination we need. I see your point on building a more traditional bot for your first build. We all had to start somewhere right? Just keep on poking around the forums here and you'll have great success. Too many smart people on here not too!

not sure if anyone has suggested this but you might use "makerslide" for the linear rails. It is pretty cheap and very accurate. Just look it up on google

@ galaxyman7: LOL. Read their wiki page linked in the first post of the thread. The whole point of their "cheapskate" bearing is to avoid the need for custom extrusions.

Exactly crispy. Makerslide and the open-rail stuff is awesome, but very hard to get, unless you know something everyone else doesn't? T-Slot from 80/20, mitsumi etc... are readily available anywhere in the world, and really isn't that expensive. Also, cheapskate uses 608's, which you prob. have laying around, and if not, the 90's are gone, tear apart them Sweet lookin rollerblades you're hiding in the closet and repurpose those bad boys!

John I see you've uploaded a close-up video of the cheapskate bearing to youtube, thanks! (http://www.youtube.com/watch?v=bSz6DfQ26pc)

I think I finally understand your comment about the 4th bearing providing pressure - it's mounted orthagonally to the plane of the first 3 bearings and provides some transverse pressure to take out up-and-down slop, not lateral slop. Right?

crispy1, correct. You got it. It's pretty smooth isn't it, for a few dollar linear axis!

Also, I'm uploading another video as i type. I got the MAX up and running all three arms on it's own "brains" today. The RAMBo boards we got in from UltiMachine have their new USB vendor ID associated with them, and I couldn't get the Arduino drivers to install. Finally figured out it was their ID#, hacked the driver, and BAM, RAMBo is born on the MAX. I posted the driver file up on the Rambo wiki page in case others run into it. It's still pretty new, they just released it on their site this week, so I doubt hardly anyone has one yet.

Ok, so apparently our site and forum is involved in the great go-daddy outage of 2012. Sucks. Just posted a quick clip on our Facebook page of the MAX running with the molded arms. Still trying to get the RAMBo's digipot control into the newest rostock firmware Johann has up. The RAMBo uses the firmware to set motor current and micro stepping, which isn't in any of the existing Rostock-based firmwares yet. There's a guy working on it in his repetier firmware port he's doing for the rostock projects, but not done yet.

We've had the machine running that same loop you see in the video on facebook all day today, and havn't missed a single step yet. Hope to get the extruder mount and bowden setup on it tomorrow, then we'll see how she prints!

Gonna be streaming to Ustream today, have some ppl we're going to chat with online about the machine, thought it'd be neat to leave the feed on all day for anyone who wants to take a look at it.

SeeMeStream

ustreams ads suck, but if you have ad blocker plugin on firefox, that works There's a chat widget, I'll try and check there occasionally too

Got the speeds up to ~400/sec., it looks so cool moving that fast! Still don't have the delta FW for a RAMBo board yet. Sousoux has working repetier firmware with delta moves, and repetier has RAMBo support but no delta, there are marlin FW's with Delta, and with RAMBo, but none with both rambo and delta I'm close to getting it delta'd out with RAMBo, not quite there yet.

Should be easy enough to combine the changes in the two git repos, I doubt there is much in the way of overlap for the changes on Rambo and Delta.
In fact it's likely the Rambo change is just pins.h unless they add M codes to control the driver settings, even if they do I can't imagine it being a difficult merge.

Neat!.

One question, how do you deal with the (apparent) non planer "rocking" of the extruder mount under certain linear translations in one or more of the drives? By that I mean when watching the vid, I noticed that under certain motions, the carriage *appeared to tilt slightly, instead of staying coplanar with the base? (not a criticism or anything, just an observation, although it may not have been anything of the sort, and instead been caused by my viewing perspective instead) Or is it an artifact of the fast initial testing?

Love the way it zooms down lickety-split and almost kisses the base without crashing into it.

The mechanics of the Rostock should guarantee the carriage is always parallel to the base, you's got 3 pairs of control arms which should prevent any roll pitch or yaw.

Yeah, I'm working thru the bugs in my firmware-kung-fu, repetier has been AWESOME at helping me make those changes, he added RAMBo support and others to his firmware last night, it's in the dev branch on his github, and I'm hacking into souxsoux's repetier firmware with delta to add the rambo support. I failed miserably my first few attempts. I just got an email from repetier back, he looked at my code and pointed out some obvious errors I did. I have to run and get kids shuttled around, then I'll be back at it

And nope, the platform stays level in all but extreme/binding situations. I have the arms doing things they wouldn't do in printing just to "stress-test" the mechanics. It's been running for 3 days at full speed and full travels. There's very minimal "wear" on the extrusion where the 608's ride, 'd imagine about as much as a few months printing equivalent as hard as I've been running it. We had 2lbs. on it for a few hrs yesterday, and then 5lbs. on it for 1/2hr. just to see how much torque it had. We can swing 5lbs. on the platform at nearly 400mm/sec with a bunch of accel dialed in. Not exactly what would be happening with printing, but fun to see it rostock-rockin' around the bench! LOL

"And nope, the platform stays level in all but extreme/binding situations. I have the arms doing things they wouldn't do in printing just to "stress-test" the mechanics."

right. gotcha. I kinda figured as much but wanted to make sure. thanks. Very impressive movements.

Hmm... but it also presents another question. I wonder how much deviation from coplanar you could get if one wanted/intended to do so?

With three arms, using the two parallel arms on each column, none. With 4 or more? Prob. could get some cool movements going for sure!

To get all 6dof you'd need 6 arms and considerably more complex positioning logic.
4 arms would give you some odd combination of RPY as the extra axis and I don't think it would be very useful.

Well, broke down and threw a RAMPS board on it, stuffed Johann's latest firmware on it, and BAM! Nive linear x,y motion, as well as z. Excited to get the extruder on it next. Still bummed i couldnt get the RAMBo digipot stuff into Johanns firmware, or sousoux's repetier firmware either. These deltas are still just new enough, and digipot control too, that everythings one iteration behind. Thanks def goes out to repetier thougj, he got me a dev version on his github for the rambo boards, so others can use it with cartesian bots. And the new marlin with rambo support is equally awesome too.

So much accomplished tonight, so much more to go.

Appreciate all the feedback, good/bad it all helps

johnoly99 Wrote:
-------------------------------------------------------
> ... These deltas are still just new enough, and
> digipot control too, that everythings one
> iteration behind ...

... not really ... I've posted some infos reagarding my Tripod-setup ( [builders.reprap.org] ) and ideas for string-driven variants ( [forums.reprap.org] ) and some other ideas ( [forums.reprap.org] - look at the attached images and links) back in 2007 ... and the basics were from maybe 1980 or so

The software were then the harder part, but this is solved now - so I'm awaiting much more progress in this direction

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Viktor, There have been others for sure, WAY before any of us were doing it. I havn't seen all your stuff, good stuff for sure.

Well, Sousoux portedin the digipots that repetier added to his firmware, so I now have Delta Firmware WITH RAMBo digipot control for the stepping/current! THANKS Sousoux and Repetier!

... this isn't "who was first", but more a hint, that many usefull stuff and ideas are 'buried' in old threads and in the wiki, where nobody will find it again - if not expecting/remembereing it or not searching with the correct phrases ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I've tagged it all for reading, you def. have something when you talk about how much knowledge gets buried in the past. Anyone say Nikola Tesla?

Streaming live again, testing more, hopefully adding hot-end/extruder

Ustream

... "Nikola Tesla" ... wait ... hmmm ... ah, yes! - AC-current, 'Tesla-coils', worldwide energy-transport through resonant EM-fields through the ground without wires, wireless remote of flying devices (imagine intercontinental rockets short past WW1!) ... some more such stuff ... discredited... retired ... paranoic ...

Yes, there are some more similar occasions, where ideas/concepts/inventions were brought to public, before they could be realized for common use and were delayed for years ... or even forgotten completelly, until someone else 'reinvented' them accidentally ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Ok, so here we go again, better than ever, I hope!

Rostock MAX on Indiegogo

Hi John - Really love the 'Rod' design - Maybe you should consider putting up just the mechanical parts as a mini-kit? (No Alu sections, just the Rods, joints and sliding sections). It's a really big jump for most people to go from $10 to $600+

I just spotted the belts run in line with the Alu channels, Great!

It's an nice design.

Good luck with the Campaign.

Rich.

---

[richrap.blogspot.com]

Thanks rich! Yeah, we're thinking of making the arms, platform and U-joints as a kit, so you can bolt it on your printed rostock. Those are the really critical parts, and these are CNC machined u joints, the platform and arms are glass-filled-nylon injection molded, and the holes in the arms are drilled after they have "normalized" from the injection process for a day to make sure the center-center is spot on. Right now, our rods are 269mm instead of 250, to accomodate the larger print area on the MAX.

Here's some photo updates on the MAX, we finished modifying the Steve's Extrtuder for a bowden setup, and Brian Laser'd the extruder/spool mount out of the white laserboard. It's looking like tomorrow night, we ran out of thermistors so I can't put a hot-end on it yet.

nice feed for the filament. That's pretty much exactly how I was planning on feeding my printer's extruder, in order to keep moisture intrusion to a minimum from my moisture controlled storage. I hadn't thought of using a quick-fit pneumatic tubing connector though. nice touch.

Push-to-connects are nice and easy. If you search around on ebay etc... you can find them fairly inexpensive. Thanks!