Hi
I am looking to build a Rostock 3d printer. It calls for 762mm long smooth rods, I was thinking about making it taller and using 1m rods so I don't have to cut them.
Other than making the belts and plywood supports longer would there be anything else I would have to do to make the Rostock taller? Will there be anything in the programing?
Here is a link to the printers page of the Reprap website.
Reprap.org Rostock 3d printer
Thanks
Tanner

Its the software configuration you have to tweak, other than what you mentioned it should work. I am going for both taller and wider. Why pay for 1m smoth rod and only use 2/3 of it. I am sure the increased build height could come in handy when you get your printer tweaked and want to build something much larger. As for making it wider, the rods need to be lengthened. My rods will be in the 330mm range if im not mistaken. This should give some decent width as well.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

You can buy rods of appropriate length, VXB is where I got mine, and they carry the good rods (hardened and chromed). Cheap rods and bearings work, but will slowly grind each other to death and are noisier than the already noisy good ones. Mcmaster also carries the good ones, but for a lot more. Keep in mind, the good rods can be quite costly.


Almost every delta on the market is a bit undersized or under-built, particularly the Rostock. It's capable of good prints, but like many early repraps, keeping it dialed in takes some effort because it lacks rigidity. To go taller, you REALLY need to use larger rods. Kolergy on Thingiverse has done a lot of work on putting larger rods into a Rostock. [www.thingiverse.com] I can't speak for the quality, just that he has upscaled it.

While the Rostock worked, it was an exercise to see if if would. Like I said, it prints good, but we have moved on from that design. The Rostock was temperamental, noisy and expensive. I would highly encourage you to look at the 3DR, Cerberus Pup, Wolfstock, Griffin, and to a lesser extent, the Kossel Mini (the sliders drive the price up). While each has it's advantages and disadvantages, each is a definite step up from the Rostock and cheaper to build (except maybe the Kossel Mini). Personally, i prefer the Wolfstock and Griffin as they are more rigid designs and therefore handle movement and speeds better.

Whichever you build, look for the ability to adapt a probe for auto-level and auto-calibration. While building a delta is easier than a cartesian, manually calibrating a delta is a VERY time consuming process. The probe systems takes a 6 hour+ labor intensive process and makes it a half hour automated process.

I am building a rostock and can say that if done properly the rostock is a very capable design even with 8mm rods, the next one I build may have 6mm rods in it to reduce friction, as for the rod quality, I have not done this however putting polishing compound on a rag and spinning it with a drill should smooth just about any rod out. I find most of the noise comes from the fact that the rails flex while the printer moves, creating a sturdy casing for this printer solves the issues. I experimented with just placing the rods and moving the sliders as fast as possible then I put them under stretch, they quieted down so much so that the only thing I can really hear is the servos moving. I have put my full body weight on my printer so far and it does not effect the rods at all, jokingly refering to this printer as if it fails I can always re-purpose it as a bar stool. The one thing that every rostock design lacks is the ability to stretch the rods when mounted. They all either place the rods in or have the rods as part of the structure. The rods should be treated as a carrier and nothing more, once stretche by a small amount they fall within specs. My design is up-scaled quite a bit expecting at least 450-550mm of build height and probably around 350mm cylinder in it. I am still in the early stages, taking a step back and ordering new longer rods and traxxas joints for my build just last night. Removing lash from the printer seems to be the biggest challenge and so far the only system I like is the ball joint route. The magnetic joints I am not particularly fond of. I have taken to calling my printer TUKI, for one reason or another.

Edited 1 time(s). Last edit at 07/01/2014 10:22AM by jaguarking11.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

While I don't disagree with your approach and reasons, it's not exactly a Rostock anymore.
The Rostock used the rails as the frame, you are using a frame to hold the rails, this is more along the lines of how the 3DR works and why you can use smaller diameter rails, even on a larger printer. However, you may want to reconsider the 6mm rails, at least one person is claiming that the rails can deflect a full mm while printing on the original design. Remember, the loads are not 100% vertical.

By the way, spinning a cheap rod in polishing compound is not the same as using ground, hardened, polished and plated. Will it be a dramatic effect, probably not at first, but give it a year and I bet your rods do not look half as good as the good ones would.


Again, I'm not saying it's a bad printer or can't be fixed (I liked my Rostock), just that we have newer, cheaper, easier to build designs now. I nearly built two Griffins for what I spent getting my Rostock working well, and the Griffins run faster and quieter.

The rods I'm using are tool steel rods. They are hardened, and I am using synthetic grease on them. As for deflection, it is true that they might deflect, however when put on a stretch the deflection is nearly non existent. Like I said my printer is rostock based, however it strays from the design, being that it is open source you can call it what you will. I am aware the 3dr uses a similar design, however 3dr does not stretch the rods, it merely places them in their perspective holders. While this reduces the quantity of deflection, it does not eliminate it. For steel to be used as a precise guide it must be exploited to its designed potential. When builders make such a design tension is desired in any structure where one hopes for torsional stability.

Steel has a few states which must be observed.

rest (holds its own shape however it is easily disturbed and does flex) - not desired
spring state (It is mildly stretched holding its shape to a higher degree as well as maintaining a higher accuracy when under load) - not desired
plastic state (overstretch and warping) - not desired

In this kind of build, especially on oversize printers where those extruded aluminum rails commonly used can get quite expensive in lengths of 36inches or more where torsional stability becomes an issue even for those extrusions. A design where one uses simple black pipe and solid steel rails or as in my case recycled aluminum tubes reinforced by all thread rod removes any torsional flexing in the chassis and further improves the stability and minimizes axial loading on the rails when stretching the guide rails probably less than .5mm....

I also believe there will not be much if any appreciable wear on the rods or bearings after packing them with proper grease and simply performing routine maintenance on the system to ensure proper lubrication, I personaly think the wear comes from the rods flexing and the bearings being forced to dig into them vs gliding on a thin film of lubricant. I also have an idea to automate the lubrication of the rails when the printer homes before a print, I may or may not implement this portion.

My goal is quite different from reprap mentality here. Frankly speaking; my goal is to build a solid 3d printer that once setup it will work as designed. My real goal is to remove all plastic parts from the interior of the printer, and keep those plastic parts out of it while adding fire safety and a heated build chamber allowing for large ABS prints with no warping. Time will tell.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Finally after getting a new DDNS my webpage is up again. If you want to check out my blog for some of my ideas mentioned above check out my signature.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Your printer looks good, but some of what you did is major overkill, other places you kind of short changed yourself. I'm not trying to pick it apart, just trying to help you avoid some of the problems others and myself have had and where you are headed.

I like your server PSU idea, much better than a normal ATX.
It's hard to tell from the pictures, but it looks like you used 16ga wire for your bed, you want 14ga. or you will handicap your bed. That may be a good thing considering your plans for the glass.

Ditch the universal joints from the Rostock kit. Buy some Traxxas ball joints and some sort of rods (I used Oak doweling and set screws, later garden stakes, and carbon). If you see slop, use rubber bands to pre-tension them. You may also be temped by magnetic joints, personally, I hate them. They chattered at speed and threatened to fly off. Low slop, but they are expensive and their weight limits speeds. Some swear by them. Most however agree, the universal joint arms are a pain in the neck, so much so that there is even a nickname for it, the Delta Arm Blues.

Don't build the head as pictured with Kapton tape wrapped half way up it, the black needs to be exposed for cooling purposes. I know one person with one of those heads, he says it works okay. That manufacturer can be iffy in the best of times, they also made your Ramps and lcd and you can probably see the quality just oozing from those. If it all works, great! Oh... The heated bed mosfet on the ramps REQUIRES a heat sink, even with one, and a cooling fan, it may overheat and shut off, and after a while of doing this, it will probably blow out. The heated bed mosfet is the one that sits parallel to the terminal block. I've killed two.

The idler bearings are problematic for some, there was a double sheer upper bearing mount on Thingiverse, but it's been removed and I don't have a copy of it.

Grind flat spots on your motor shafts for the sets screws, and use Loctite.

Overhanging your glass from the heated bed, doesn't matter how well you insulate it, if you use something other than Borosilicate, it will crack in a relatively short time. Normal glass doesn't like being unevenly heated or cooled. Your plan only slightly delays the inevitable. Even when fully heated, normal glass is prone to breaking, mine lasted a few months. I could be wrong, but I wouldn't expect it to last. $30 or so will get you a custom one from China.

Drill rod is not the same as ground, hardened, polished and plated, sorry, it's not. The Rostock Wiki page mentions them specifically because so many have had problems. Yes, bending can contribute to the wear, however, it depends on the type of bearing used. The ones you are using, are much more prone to digging into a rod, partly by design, but made far worse by being cheap Chinese bearings. You are making all the classic mistakes here and repeating the same phrases said by those who later fell victim. These, these, or these can help, but may not be a perfect solution.



Lastly, I think you both over and under estimate the amount of effort it takes to maintain some of these machines and the costs. You go through all this effort to avoid extrusions and then rely on wood and manual labor to keep it all aligned, together and running smooth. One reason we all tried to get away from wood was because it can warp, or in your case, crack, especially with the way you are loading it. This may not be a huge concern unless you live in a high humidity region like I do, but you already have cracks. You may get lucky, or it may split in half in 3 months.

V-slot is CHEAP, even at long lengths and larger sizes, and I don't have any exposed grease, nor has the system needed any maintenance. Okay, I had to pull bits of filament out of a belt pulley once. It's also quiet. My Griffin are the quietest printers I've come across, and get comments on their lack of noise wherever they go. Far quieter than My Rostock which could be heard in other rooms under hard use (the dampeners you got will help). My real maintenance has been trying to get the extruder to keep up with the printer, otherwise, I don't really do any maintenance. V-slot gets even more cheap when you add in the cost of good linear bearings, which are ridiculously expensive.


Oh, if you shrink the pictures on your site, it will load a TON faster. You used full size pictures, then used HTML to reduce them. You are probably using 80% more bandwidth than you need.

My pictures load slowly because my internet connection is not exactly commercial. This is self hosted blog, although some of the pictures from the earliest posts are way too large, I will compress and shrink them at some point.

As for the wood, well as you pointed it out its not exactly the best material, however this is a work in progress. I am planing on a upgrade to steel or aluminum ends for the printer at some point. However at this point the wood only has to live for a few prints. The wood deceives you however, it is not cracked aywhere, I splintered some of the edges when machining it with my limited hand tools. If it does crack then as mentioned I will be ordering some precision parts to replace it.

The universal joints, its funny you mention those. I have already scraped them when I disassembled my spider. I have traxxas joints on order and aluminum 1/4 solid rods sitting on my desk. I will in time remove the remaining 4 parts of printed plastic from the spider and replace them with aluminum only using the plastic parts as a means to setup a jig. However for the time being I am going to have to deal with those parts as I tweak the firmware and the hardware to a point of satisfaction. I am planning on wrapping 2/3's of the printer in aluminum and insulating the interior with cork, I will be using some sort of clear plastic window for the front of the printer as well.

The idler bearings I am planning on addressing as soon as I get the printer to function properly.

Good point on the motor shafts and set screws. I have been thinking along the same lines.

The bearings, its funny my parts were delivered with nylon bearings which I did not use, I would rather use the metal bearings right now. At some point if wear becomes an issue, I will upgrade the system.

As for the heated bed, it is a good point on borosilicate glass. I may eventually go that route, or may use stainless steel. For now this was a decent solution that did not involve me waiting 3 or 4 weeks to get.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Using bar stool tops for your frame will be your undoing. They are not meant for this type of application. I am not going to recommend that you replace them, I am going to TELL you to replace them with a piece of 3/4" plywood. As an engineer I would hope that you understand that pieces of wood that are glued next to one another with no jointing will fail much sooner than sheets of wood stacked on top of one another and glued together with grain patterns switching directions every layer will.

When you start getting movement and heat inside that machine, the glue is going to begin to break down and the wood will begin to shrink and swell. I would give it all 2 months, tops, of working before your printer falls apart.

I would also suggest 10mm rods over the 8mm given the length you are working with. I learned the same thing the hard way when I build a large scale printer. Thicker rods are better.

Wrong box! Disregard!

Edited 1 time(s). Last edit at 07/04/2014 07:41PM by Iceman086.

Quote
jaguarking11
As for the wood, well as you pointed it out its not exactly the best material, however this is a work in progress. I am planing on a upgrade to steel or aluminum ends for the printer at some point. However at this point the wood only has to live for a few prints. The wood deceives you however, it is not cracked aywhere, I splintered some of the edges when machining it with my limited hand tools. If it does crack then as mentioned I will be ordering some precision parts to replace it.

The bearings, its funny my parts were delivered with nylon bearings which I did not use, I would rather use the metal bearings right now. At some point if wear becomes an issue, I will upgrade the system.

As for the heated bed, it is a good point on borosilicate glass. I may eventually go that route, or may use stainless steel. For now this was a decent solution that did not involve me waiting 3 or 4 weeks to get.

I've been working with wood since I was a child, I'm not deceived by the wood cracking, I know it's from you working with it, but as an engineer you should know about stress concentration or stress risers. You just created a bunch, and right in a bad spot...

Ever see people breaking boards with their hands, they break it along the grain, and it's actually pathetically easy to do. As kids, in woodshop we used to do it with our heads, once in a while, someone would convince one of the new kids to try it, but turn the grain sideways and watch them give themselves a nice headache. Take a look at this picture, you are straddling the grain right where you apply your force. Not just once, in the front, but also in back, through the same section of grain, right where your stress risers are. Short of hitting it with something, you are doing all you can to break it.

Once wear becomes an issue, changing the bearings isn't going to help. The damage is already done. That was why they shipped the rods with those bearings, because they know normal ones will utterly destroy them in short order. Again, this is a classic mistake, and you're saying the same things everyone else did before they had problems.

Any heated metal bed, will warp, quickly. I've heard of Makerbot beds bending several mm.

Go the way you want, but you will be redoing a lot, and while you think it's not a big deal, you might once you realize it can take 6 -10 hours to calibrate a delta, every time you make a significant change. Which pretty much means everything on a delta short of a belt or pulley change.

I'll PM you regarding your pictures as it goes way off topic.