mcwire/mendel and aluminum t-channel?

I was taking a detailed look at the mendel and mcwire designs. Is there a reason aluminum t-channel isn't part of the designs or could not be? To me it could eliminate almost half the parts of mendel, while being cheaper and readily available ? It is available in metric pretty universally (20mm should easily be strong enough).

If you don't know what it is take a look at mcmaster page 1908-1910 . It is basically industrial erector set parts and vary widely used in industry here in the US at least?

Is there a reason aluminum t-channel isn't part of the designs or could not be?

No one got around to it. Yet. If you'd like to work on it, it would be good to create a wiki page:

(Instructions: just cut and paste from)
[objects.reprap.org]

---

I think what you're proposing is very cool - an t-channel-based CNCRouter/RepRap.
This sounds separate from mcwire/mendel so pick your own name for the instance of the machine, e.g. goininreverse-bot, or mycat'sname-bot.

Or AlEx-Bot, for Aluminum, Extruded-Bot.

Once things get established, we'll give the subject a dedicated forum.

---

This is a good place to keep notes, links, drawings and such.
[objects.reprap.org]

---

I think there must be some machines, however poorly sketched out or documented around, up on CNC zone. I know shopbots have some t-channel in them, but they're not going to give away the store by GPL-ing their crown jewels.

---

I think the ?nylon/hdpe? sliders that run in the T-Channels wear out, and so you'd have to reprap new ones.

But I think the RepRap community needs a machine like this.

Edited 1 time(s). Last edit at 01/20/2010 04:44AM by SebastienBailard.

Since you asked...

I've almost completed an aluminum t-slot RepStrap design using 80/20 parts.

I'll try to post some info later this week.

-Tim

Tim,

Your repstrap looks very nice. I'm looking forward to a more detailed description.

that is a beautiful machine! how much did it cost you?

Until we get metal extruders, or solve the issue below, all of what we make at the moment are really RepStraps to me.
The problem with making say an ABS RepRap is to design a machine that can make something that is the same size as itself. Once we work out that little physics problem we can then work on a true RepRap

Making untestable semantics based statements only encourage flame wars ("It IS a reprap", "It's NOT a reprap", "it's a RepStrap".. "Oh yeah? Prove it!")

Rather than semantics, try stating your issues with testable assertions and preferences, for example, in paraphrasing and attempting to decipher your intent, I believe you may have been trying to say something like:

The Mendel RepRap can produce something like 100 of it's 200 parts for $500 (vs. ~$1200 without an existing machine to turn plastic into parts). It requires about 40 hours to create a new copy with all non-RP parts available, consisting of 30 hours print time and about 10 hours for manual assembly. --- I made educated guesses on some of the numbers, others can modify my numbers and provide their own findings. Compare that to the untestable notions that a piece of hardware has some intrinsic property of "RepRapness" or "RepStrapness".

It would be desirable to produce a collection of machines which asymtotically approach 100% of parts to produce a copy of that collection of machines for under $200, with a start to end time of 10 hours.

My own personal take on the repstrap built by bothacker: I really like it. The design appears to allow someone with minimal tools and skills to recreate the machine, with time and cost only slightly higher than that of a second generation Mendel (<$100 extra expensive from my guess..) It also appears to match the print speed of a second generation Mendel, something that I've not seen in comparible RepStrap machines.

I had these design goals for the RepolaRap -- not much more expensive, about the same time to construct, and about as fast as a second generation Mendel.

Edited 1 time(s). Last edit at 01/21/2010 01:00PM by BeagleFury.

Been playing with this a bit more... Bothacker can you provide some of the dimensions of your device and the parts list from mcmaster. I am especially interested in the rod you used, the bearings, belt and the interface from the motors to the rods.

My theoretical mcmaster order list (note zero cutting, 4 acrylic drilled holes necessary at this point to reach mcwire stage)


2 4 each 5537T105 Aluminum Metric T-Slotted Framing System Four-Slot Single, 40 mm Solid Extrusion, 2' Length 2 weeks $14.33 each 57.32
3 4 each 5537T286 Aluminum Metric T-Slotted Framing System Corner Connector, 90 Degree, for 20 mm Extrusion 2 weeks $5.92 each 23.68
4 4 each 5537T177 Aluminum Metric T-Slotted Framing System Extended Plate, Double, 8-Hole, for 20 mm Extrusion 2 weeks $7.25 each 29.00
5 4 packs 5537T154 Aluminum Metric T-Slotted Framing System Std Zinc-Plated Steel End-Feed Fastener, for 20 mm, packs of 4 2 weeks $2.55 pack 10.20
6 1 each 4615T31 Impact Resistant Clear Extruded Acrylic Sheet .220" Thick, 12" X 12" in the morning $8.78 each 8.78
7 6 each 6112K14 Hardened Precision Metric Steel Shaft 6 mm OD, 600 mm Length in the morning $12.56 each 75.36

ryangeorge -

I'm not sure what my machine cost -- I've never added it up. The aluminum is fairly inexpensive from McMaster. The main cost for me was the time I spent designing and building it, which was significant. But I had fun doing it. I used a Taig CNC mill for stuff like the stepper mounts, and a lathe for some of the axle/drive parts. And I spent a lot of time in front of the drill press. Thus, it probably isn't a very good repstrap option for most folks.

On the other hand, the t-channel repstrap I'm working on is very similar, but designed from the start to be easy to build and require no machine tools. So stay tuned

goinreverse -

Are you the Chris who emailed me about part numbers yesterday? If so, the info should be in your inbox. If not, let me know and I'll repost it here.

-Tim

Nope not chris... Please post them here so everyone can benefit if you are willing to share.

I got a crazy idea that I can leapfrog the whole bootstrapping thing and just build a mendel with all the printed parts replaced with standardized parts, aluminum t-channel, nylon racks and pinions, linear bearings, same shafts. Looking at your pictures it seemed like your machine was particularly large which is why I was interested in the dimensions.

Guys, I highly suggest you create a wiki [[page]] for this.

Maker Beam anyone?
[www.thingiverse.com]

I think something like this that is reinforced with metal to ensure there is no warp would do what you are proposing.

that thingiverse item is modeled exactly after extruded aluminum slotted t-channel, see the mcmaster part I listed starting this forum.

I am pretty close to a final design, involves 0 printer parts, only a drill press and screw drivers, is extremely similar to the mendel. I am computing a final cost for the parts excluding the extruder and motors but even from mcmaster right now it is coming in at $550 off which I think I can shave another hundred.

Still need to build of the prototype to test things out which will take a couple of weeks but so far this concept seems like a real winner to me.

I am still building a mcwire to build a "real" mendel but if I look at costs and time this boostrap process just seems idiotic and simple math shows that as long as you have a rapidly expanding user base it is IMPOSSIBLE to produce enough printed parts...

Cool! Going to wiki it?

I am still building a mcwire to build a "real" mendel but if I look at costs and time this boostrap process just seems idiotic and simple math shows that as long as you have a rapidly expanding user base it is IMPOSSIBLE to produce enough printed parts...

Hmmm ... I'm hopeful about the people making molds for mendel parts. They can pour a set or two of mendel parts in a day.

Now if we print molds for mendel parts, things change even more.

In complete agreement on the mold and pour, see my "mold and pour" post. However no one was willing to mail me a part to borrow to mold and pour as an example...

Directly printing the molds is a neat idea.

That led me to the idea of a mendel-strap which is a mendel without using any printed parts. I guess you could use it to print parts for a printed-mendel but at that point I don't see the point if I can get the cost low enough that is is competitive with the printed mendel which I am pretty sure I can.

I will wiki it as soon as I actually build one, I don't want to mislead anyone until I have functioning proof.

I am in the middle of building a mcwire to print mendel parts right now but the whole process has left a really sour taste in my mouth. Time and expense wise it just seems idiotic, the primary design (mendel, etc) should include a way to be built without printed parts or with no more printed parts than a single other device can print in a single afternoon. Right now IMHO the engineering on the primary mendel is goofy, a lot of parts are un-necessarily complex just assuming they would be printed.

Maybe a change to the license would be appropriate like a build one -- print one (even if it was honor system) with the printed copy going to some kind of repository for purchase/distribution to benefit the non-profit.

Mendel-strap:

Some things are probably way over speced like using 4 needle bearings for the z axis threaded rods, you could almost certainly use 2 and use PTFE bushings/flanges as bearings for the top if not all 4 points.

There is also a cheaper generation of roller guides ($35 a piece) than what is below but I wanted to see what the pricier ones were like first.

Basically you cut (bandsaw, hacksaw or chopsaw) the extruded aluminum to 15" X7 and make a square, the parts lock together using the corner connectors. Then you build a 3 piece vertical gantry and lay it on top of the square, the corners connect with screw in brackets and tie plates secure the gantry to the square base. Set 2 more lengths of extruded aluminum to span the base (moves on the x axis and the gantry (moves on the Z vertical axis and the Y axis).

The crucial part of the design are the roller guides that slide along the existing slots in the extruded aluminum. This takes away 40% of the steps of build a mendel. You just pop the guide onto the t-slot aluminum, that's it. The guides are pre-drilled with holes to mount the build surface and simple acrylic motor carriages onto them. I think the whole thing could be assembled in 3 hours including cuts, etc.

X and Y are driven by rack and pinion gearing that spans their traversal direction, z axis is driven by the belt/threaded rod arrangement same as in the printed-mendel. I haven't work out another way to do it yet, if the span is rigid enough and the guides don't bind it could also be driven with a rack and pinion which would cut out a lot of cost and complexity.

I think one pretty cool part of this design is that if you wanted to make it 2' rather than 15" for a little more money you could get all the parts in pre-cut lengths from mcmaster and have a huge build area for CNC or whatever.

Here is the current parts list if anyone is interested:

2 each 60585K36 UHMW Plain-Bearing Guide Block for 1" Width Rail, Double Flange, 4" Length Your reference   in the morning $57.94 each 115.88
2 each 60585K35 UHMW Plain-Bearing Guide Block for 1" Width Rail, Double Flange, 1-7/8" Length Your reference   in the morning 45.43 each 90.86
2 each 47065T174 Aluminum Inch T-Slotted Framing System Extended Plate, Double, 8-Hole, for 1" Extrusion Your reference   in the morning 6.93 each 13.86
8 each 47065T242 Aluminum Inch T-Slotted Framing System Corner Connector, 90 Degree, for 1" Extrusion Your reference   2 weeks 6.83 each 54.64
1 each 47065T126 Aluminum Inch T-Slotted Framing System Six-Slot Double, 1" Solid Extrusion, 8' Length Your reference   in the morning 44.64 each 44.64
1 each 47065T211 Aluminum Inch T-Slotted Framing System Plate, Double, 4-Hole, for 1" Extrusion Your reference   2 weeks 6.20 each 6.2
3 each 5174T1 Steel 20 Deg Pressure Angle Spur Gear Rack 20 Pitch, 1/2" Face Width, 1/2" Height, 2' Length Your reference   in the morning 17.60 each 52.8
3 each 5172T11 Steel 20 Deg Pressure Angle Spur Gear 20 Pitch, 15 Teeth, .75" Pitch Dia, 3/8" Bore Your reference   in the morning 15.35 each 46.05
1 each 98942A115 ASTM A193 Grade B7 Alloy Steel Threaded Rod Znc Yellow Pltd, M8 Sz, 1-Meter Length, 1.25mm Pitch Your reference   in the morning 11.00 each 11
1 each 6484K509 Trapezoidal Tooth Neoprene Timing Belt .200" Pitch, Trade Sz 370XL, 37" Outer Circle, 1/4" W Your reference   1 week 5.09 each 5.09
1 each 2958A113 Metric Black Oxide HSS Jobbers' Drill Bit 6.7mm, 101mm Oal, 49.6mm Drill Depth, 118 Deg Point Your reference   in the morning 3.26 each 3.26
3 each 57105K21 Acetal Pulley for XL-Series Timing-Belt for 1/4" & 3/8" Belt Width, 1.63" OD, 22 Teeth Your reference   in the morning 8.07 each 24.21
1 each 47065T124 Aluminum Inch T-Slotted Framing System Six-Slot Double, 1" Solid Extrusion, 4' Length Your reference   in the morning 22.98 each 22.98
1 each 60585K36 UHMW Plain-Bearing Guide Block for 1" Width Rail, Double Flange, 4" Length Your reference   in the morning 57.94 each 57.94
4 each 1434K15 Self-Align Stl Flange-Mnt Needle Roller Brng for 12mm Shaft Diameter, 60mm Overall Length Your reference   in the morning 10.26 each 41.04
4 each 2640T11 Metric Acetal Sleeve Bearing for 6mm Shaft Dia, 12mm OD, 10mm Length Your reference   in the morning 2.78 each 11.12
1 pack 93835A360 PTFE Spacer with Flange Short Barrel, 1/4" Screw, .045" Length, packs of 5 Your reference   in the morning 7.96 pack 7.96
1 pack 95630A480 Low-Friction PTFE Flat Washer 5/16" Screw Size, .88" OD, .074"-.086" Thick, packs of 5 Your reference   in the morning 4.90 pack 4.9

We've got a moldmaking forum now:
[dev.forums.reprap.org]

---

[[Metal/Mendel]]
[objects.reprap.org]

---

Maybe a change to the license would be appropriate like a build one -- print one (even if it was honor system) with the printed copy going to some kind of repository for purchase/distribution to benefit the non-profit.

I proposed it on a mailing list. I don't have the time to spare on.
[objects.reprap.org]

It might be better to have a traveling set of molds. Or metamolds. But if you're doing a pourstrap, might as well do a simple block geometry:

[[Epoxy-Granite]]
[objects.reprap.org]

Basically you cut (bandsaw, hacksaw or chopsaw) the extruded aluminum to 15" X7 and make a square, the parts lock together using the corner connectors. Then you build a 3 piece vertical gantry and lay it on top of the square, the corners connect with screw in brackets and tie plates secure the gantry to the square base. Set 2 more lengths of extruded aluminum to span the base (moves on the x axis and the gantry (moves on the Z vertical axis and the Y axis).

I think I'm going to make a Repstrap folder, with dedicated forums for extruded aluminum, wood, etc.

I don't mean to be obstinate but why do we need all the different forums, seems like given the volume "general" is adequate for everything and I have hardly enough time to keep track of posting to it?

Just my opinion but I would rather see more activity concentrated in general than spread out where things seem to die, I guess it would be different if there were 1,500 posts a day to general but there seems to be barely a dozen or so...

I have made several posts about this but it still doesn't seem to be having any penetration

take a look at [www.cubespawn.com] please, RepRap, CNC mill, got a guy designing a slide CNC turning center, 'nuther guy:UV potting/curing station for a packaging line...
a vacuum forming trimout line is in preliminary stages
AND when completed the designs all bolt together to form an assembly line, in any sequence...

starting to get activity on the control architecture to manage multiple cubes and coordinate thier activities - but even with all that going on, this project still needs your support

Thanks
James