Titanium Conveyor For Makerbot ABP and raftless printing
Description
Titanuim foil is an ideal platform material because:
1) Unlike aluminum foil it does not normally crinkle and maintains a smooth flat shiny surface unless deliberately creased. It it actually somewhat springy.
2) It is flexible like a sheet of paper and rolls smoothly through the mechanism
3) The coefficient of thermal expansion is much lower than plastic over the temperature ranges used, therefore although the plastic will adhere nicely when hot, when both are cooled the plastic shrinks more than the surface beneath it and pops right off.
4) It is physically extremely tough and will survive a lot of abuse
5) It is chemically inert and immune to solvents
The titanium build surface is immune to solvents, and this allows a unique sort of raft-less printing. Essentially what you do is use waste plastic dissolved in acetone to rapidly paint on a thin film that helps the first print layer stick. This has the key benefit of conserving useful material and saving the time needed to print a raft. I've had a great deal of success with this using ABS plastic, never tried it with any other plastic. So far using this technique my prints have shiny smooth bottom surfaces and no warping or pre-mature detachment or slumping type defects. I've never had to abort a print due to a failed first layer and it's pretty much my main method now. If you want to print normally with a raft, the film is not needed and the raft adheres very well to the metal.
EDIT: It seems like other people have used the scrap plastic-acetone method before on metal build platforms, didn't realize that when I came up with it. As far as I know this is still the only metal build surface that rotates and functions like the original conveyor belt though.
Instructions
1) Measure the conveyor you removed, add several centimeters for attachment purposes, and cut this length of foil from the roll or larger piece.
2) If your piece is too wide measure the appropriate width and trim it down to size. I was lucky that my piece was pretty much the right width, but the continuous piece needs to be at least as wide as the platform you want to use, trimming down is easy but attaching multiple strips side to side won't work well.
3) Remove the rear conveyor spindle and thread the foil through the mechanism. Overlap the foil slightly and use scotch or capton tape on the inside of the loop to eliminate an edge and again on the outer surface for strength.
4) Insert the spindle again and tension the belt. The tape adheres quite well to the titanium surface and the belt is fairly strong. It may take some troubleshooting to get the right belt size and tension, if you remove the tape do so slowly and carefully, sudden forceful movements can put tiny little crinkles in even this foil.
5) That's it! Happy printing. You will need to use the control panel to rotate the belt between prints to ensure that you're printing on a clean metal surface and not over the tape.
Printing Raftlessly:
1) Take scraps of ABS (eg. rafts from other prints, failed or broken prints, small scraps of filament, etc. Everyone probably has plenty of this lying around) and put them in a jar with a layer of acetone. The acetone fully dissolves the ABS into a pasty sort of goop. There's no exact mixture but in the end it should have a smooth creamy consistency somewhat like molasses. play around a bit, if you make it too thick or thin it is easy to evaporate off some acetone or add more. It may take several hours to overnight to fully dissolve uniformly depending on the size of your starting scraps.
2) Use a brush to apply a light coating of the ABS solution to the Ti Build surface, try to get it as uniform as possible and only cover the area where a raft would normally go. It is best to do this when the platform is cold, if the platform is hot the ABS solution may bubble into a foam and negatively affect print quality.
3) Print normally on this coated surface, disable the raft in the skeinforge settings before making your g-code.
4) When the print is done, trim away the excess film from the edges of the bottom. It should be transparently thin and papery, the layer actually below the print remains with it and adds negligibly to the height, the only noticeable or measurable difference is a smooth shine finish.
Note the second image is a view of the bottom of a large flat surface I printed using this method, notice the nice smooth flat finish.
You must be logged in to post a comment.
The belts produced by GalliumSource can be found here:
These belts are wider and longer than the one I originally used, allowing for better coverage of the platform and more flexibility to individual machines, but the critical thickness is the same. Having searched through many different suppliers of titanium foil I would say these are reasonable pric
ed too, with 2 pieces for $49 when a roll of this stuff can easily run over $1000. Given how long my first belt lasted, I would say these two belts are all you should need for a few years at least if you take good care of them.
I wonder if there's a way to offer these through the makerbot store.
We would like to make these titanium belts for people with the Thing-O-Matic but we would like to know the measurements including the desired thickness of the material. We have heard that the belt should measure 4" x 14" but we need the thickness. We can then offer these for sale on amazon.com and on our website which is www.galliumsource.com. Please drop us an email at kurt@galliumsource.com to let us know.
I just bought your Gallium Source TOM belt via Amazon; will install in a bout a week (working off last plastic belt). Just for reference my belts measure:
10.4cm (104mm) wide (from a brand new MBI belt mounted with no use)
28.5cm (285mm) long (from a spent and cut belt)
Hi Kurt,
That sounds like a good offer, I've had a hard time finding a good source with the proper dimensions and reasonable minimum quantities. The thickness that I used on mine was roughly .001in (~.025mm) as determined by calipers, and it seemed to work nearly perfectly. The material was definitely titanium, although I don't know what grade or heat treatment was used. I'll go back tonight and measure the platform and wrap-around on the machine build platform, generally too long is better than too short though since the foil is easy to cut with scissors and a slight overlap of the edges is needed to secure it. If your company offered these there seems to be a fairly strong demand and I could post the link up in the item description.
Hi, I'm also a frustrated ToM user that threw away the ABP because of the less-than-decent results obtained with the plastic belts. I was htinking of building one of these, but I have a question: if I use titanium I can print straight onto it, right? On the other hand, using steel foil like in Tunell's derivative y need to coat it with kapton tape. Did I get that right? Thanks!
Hi, sorry it took me so long to see this.
You have it pretty much right that you can print directly to the titanium, no capton needed.
I typically use the acetone-ABS solution method to save the time and material of printing a raft, but of course that kind of negates some of the usefulness of the conveyor since you have to re-apply it each time and can't just leave the thing to make copies all night.
I have found that if you use the raft you can print directly to the bare Ti surface with very good results, my parts never had a problem popping right off or adhering initially. A word of caution is to check a part or two before leaving it to crank allnight though because some larger or taller parts can start to detach and warp or tip over part way through the print simply by virtue of the titanium being so inert and smooth that most things don't really want to stick to it. Both problems can be eliminated by using the acetone-ABS solution method and you can get parts as tall or flat as desired, but again at the expense of intervention between prints. It mainly takes some playing around.
UPon more reflection regarding using a highly non-thermally conductive material as a heating medium. Heat will dissipate through a path of least resistance. Yes, your belt will warm up - reluctantly, because Titanium is VERY non-thermodynamically conductive. It will get warm/hot but, the heat generated by your heating circuit will stay mostly in the board and will migrate more through the platform. This in the long term probably would not matter for a home toy that will not be used for long and is well designed (MB is well designed for the board - that is my opinion). On a long build you can expect the entire platform to begin to be very warm. Yes, the platform is temperature controlled and fortunately or unfortunately shut off when at temp. but, then your belt won't be at the temperature you want/need it to be. So, what will happen? The operator will complain and crank on the heat. This then raises the heat generated and well the oven you have created beneath the Titanium belt gets nice and toasty.
For the long run I've fully changed my mind on this fun idea. I'll stick with a highly conductive BP of my copper and others use aluminum. I'll work to a less thermally isolating and thinner material for my belt or platform cover and assure that my heat dissipates up and through my part rather the
n back down and into my circuit board, thermistor, wiring, wood, bushings, screws, nuts and such.
P.S. Titanium is used for camp stove
&
amp; fire cooking heat shields.
Remember that the original conveyor belt supplied with the kit was made of kapton plastic, which is listed by DuPont as having a thermal conductivity coefficient of 0.37 W/mk. The thermal conductivity coefficient for titanium is 21.9W/mK, so the Titanium conveyor is at least 59 times as conductive as the original, possibly more as the foil's high strength allows it to be thinner than the plastic. I've printed hundreds of prints on this with the platform set to the recommended 120 deg, and there is no problem with heating. See the calculations below that I made in response to bstott, the build surface of titanium 0.025mm thick has a thermal conductivity of 8760W/K, which is about 17.5x more than the total max output of the makerbot's 500W power supply.
P.S. They also use Titanium for camp cookware, and the water in a titanium pot boils just fine on a stove, and if you touch it you will still be burned as it conducts that heat into your hand. It is low conductivity only in relation to other metals, compared to many other common materials it is st
ill a good conductor.
I haven't tried the titanium foil by the author, but I tried steel foil, and it seems to work great. The adhesion is terrible, though. I haven't tried using acetone, though as I doubt steel will react well.
I did try masking tape, though and it seems to work well (although it gets bubbly).
I've ordered wide kapton from makerbot and we'll see how that affects things
I'd pay a reasonable amount if someone were to sell reliable belts that hugged the surface. I went back to the HBP cause the plastic belts are nearly worthless.
so I got the titaium foil from the amazon.com link I posted. As you may have noticed, it's for hair dying.
It looks pretty good. Probably a bit too thin, but it's definitely titanium and should be an excellent alternative if you can't get your hands on the industrial stuff. You may notice that on the shiny side, there is a divot texture that may leave tiny depressions in the prints. The dull side is much less noticable.
so $7 for a huge roll at 150mm wide is not bad. I haven't been able to try it on the TOM yet, because my stepper driver blew out and I'm waiting for a new one.
Here's a picture:
For anybody in Tucson, I found that Elliot's electronics has a huge roll of steel foil .002" for $1/ft.
I picked up a few feet and will report on that as well
Hey everyone,
Thanks so much for your interest in the part, I'm really excited that it's generated this much enthusiasm. I'm currently working on acquiring a large roll of the same type of foil I used since there seems to be a lot of variation among how much success people are having with foil from different suppliers and it seems to be hard to order small quantities. I don't know how long shipping and such may take, but once I receive the foil I'll calculate out the total length of the roll and the strips and offer strips for sale roughly at cost, which should be inexpensive for a large roll. Each strip weighs about five grams, so if a kilo size roll costs $200, a strip would in theory only be a dollar. I don't know how much the roll will cost or weigh yet, and shipping+processing expenses may add to the price, but they should be very affordable. This won't be happening immediately since shipping and such may still take awhile, but it is on the horizon for everyone still interested but put off by the prospect of shelling out $20+ for foil that may not work.
Has anyone been able to find the foil? I tried contacting a distributer in china and they only take orders by the kilo, thats very long spool of foil.
I'm wondering if lightly sandblasting the belt would give better grip.
Actually i was just thinking about it. Why not use aluminum foil? Will it work? I'm gonna try tomorrow...
Unfortunately aluminum foil will not work as it is too soft and wrinkles easily. The wrinkling will imprint itself on the bottom of your prints and mar them. It is not suitable for a longterm conveyor belt. If you have a way of making it perfectly flat it works fine, I used it as a disposable platform a few times.
I went ahead and purchased a sheet from the link ferret7 posted. I just cut it down to size and seem to have good tension. However Its not flush to the surface. Wondering what I could do to remedy this situation.
I turned the HBP on and heat transfer is not a problem at all this thing gets plenty hot. Much better then the provided belt.
How thick is your foil? Mine measure .005"
I also ordered from titanium goat. It's way too thick! The motor can't drive it if you get it tight.
Could I just pay you to cut 2 belts worth and mail them to me in a tube? That way I know that i have the right stuff. If your ok with that just shoot me an email. My email is {myusernameronthingiverse}@gmail.com
My foil was .025mm, converting from .005" I get .127mm, so your foil is quite a bit thicker than mine and that may have something to do with it. Where is it not flushing up with the surface? is it staying flat and remaining a small distance above the platform, or is it curling up at the edges or something?
Tried for several hours again last night and no luck. I'm starting to wonder if the foil from Titanium Goat isn't to thick.
Hello,
I just gave science a look and from a table of thermal conductivity for metals Titanium looks pretty dismal. So, why did we get excited about a titanium belt? Because it is stronger, light and more cool (literally) than a plastic belt? It is durable but would waste our power trying to get heat thr
ough it?
Got the following thermal conductivity values below from: http://www.engineeringtoolbox....
A higher Btu/hr oF ft value is better.
Aluminum, pure 68 degrees F 118 (Btu/(hr oF ft))
Copper 68 degrees F 223 (Btu/(hr oF f
t))
Silver, pure 68 degrees F 235 (Btu/(hr oF ft)
Beryllium 68 degrees F 126 (Btu/(hr oF ft)
Cartridge brass 68 degrees F 69.4(Btu/(hr oF ft)
Titanium 68 degrees F 11 - 13 (Btu/(hr oF ft))
After looking at the above
I'm not too sure if I'm going to try and make a belt or just rack it up to another lesson.
You are also forgetting tho other advantage of titanium. Very low thermal expansion. In theory this would allow you to run the belt while the platform is hot and not stretch the belt. In practice this means less time between prints when printing using the abp.
The heating isn't a problem at all. While titanium is a poor thermal conductor as far as metals go, even this metal has a MUCH higher thermal conductivity than plastic or paper that is typically used. The thermal conductivity of titanium (in SI units) is 21.9W/mK vs 0.17M/K for ABS plastic. Also consider that the actual rate of heat transfer is dependent on material thickness and the foil is extremely thin. Thermal conductivity p=k*A/L where k=thermal conductivity coefficient, A=area, L=thickness. We have k=21.9 W/mK for Titanium. The foil thickness is 0.025mm=2.5E-5m. The platform has an area of 10cmx10cm=0.01m^2. Plugging these numbers into the above formula we find a total thermal conduction rate of 8760W/K for the platform, which is absurdly more than sufficient.
Great innovation. I got quite a surprise when I read about this. I love the way this community works. Thanks for sharing.
This is a great leap forward - materials science is always my favorite lever
1st hit on google -
&
gt; 'titanium foil'
http://www.titaniumgoat.com/wi...
~$16.00USD
I ordered some from their website. I'll post if they aren't on the up and up.
That's a good find, my hope is that if more people get good results with it makerbot will start giving the option of shipping it with their kits or as an extra part.. I've seen really large industrial rolls for a couple hundred dollars, which is expensive until you chop it up into conveyor belt size pieces, in which case each belt would be a few dollars at most. The roll at my school would supply well upwards of a hundred belts and they probably got it for less than a thousand. With enough demand economies of scale are always the way to go.
Have you tried covering it with Kapton? That might remove the need for priming with dissolved ABS.
I've been using super thin Kapton that I found on eBay in 4 inch wide strips, with excellent results.
I haven't tried kapton, but honestly the plastic priming isn't even necessary most of the time. The plastic sticks very very well when hot and you can print raftlessly directly to the Ti surface about 90% of the time, the reason I usually prime it is so I can just hit run on a long print and walk away without any worry of missing that first layer. It still ejects right off the front too because the plastic pops off (sometimes audibly) as it cools. Do the kapton strips leave any little creases on the bottom surface or interfere with ejection at all? If not I may have to try it.
What thickness foil did you use and where did you buy it from? I am seeing prices
&
gt;
£1000 for a piece 100mmx100mm!
You can get titanium shim stock for what I suppose are reasonable prices. For example, mcmaster-carr 5747T11 is .005" (.125mm, so thicker than what's proposed here) in a 6" x 50" roll for US$76.20 plus shipping. Personally, I doubt mcmaster is the cheapest source, but even so that's several belts' worth.
I'm not sure exactly what the thickness is because it didn't say on the roll, but my calipers give a reading between 0.02 and 0.03mm, so I'd guess around 0.025mm since that's a pretty standard foil thickness. A pretty wide range should work though, it just needs to be flexible enough to roll and tough enough to last. Honestly I'd probably prefer thicker if I could find it, the thinner you go the more easily it crinkles, anything below 0.1mm would probably still roll just fine.
I'm a materials science student so I've used this stuff before in experiments where we needed a metal foil that wouldn't burn through or loose strength when at red-orange heat, I have pretty much free access to small scraps like the one I used. It's really expensive from some scientific and indust
rial suppliers, but small sellers disposing of perfectly usable old stock can be really cheap, you just have to shop around a bit. Here's a spot I found with a pretty wide and reasonable priced selection:
http://shop.ebay.com/shineline...
&
amp;_armrs=1
&
amp;_from=
&
amp;_ipg=
License
More from tmartin
Does this solve the issue of belt lifting of the build platform? No matter how tight I get it, the belt seems to lift slightly as the nozzle applies torque to the opposite side. After two prints, I removed it and simple put kapton on the aluminum foil (which also means I can lower the temp significantly)...I honestly think the solution is to use no belt at all. I like a sweeper arm much better, and I would bet the only reliable method of using a belt of any material would be to apply suction to the belt from below. I don't think that's practical though.
I guess so.
It does not expand like plastic does.