Thread Rating:
  • 1 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Help wanted for 3D printer selection (comprehensive details provided)
#1
Hello everyone,

I work as a laboratory assistant in motor psychology, and our main goal is to make various and numerous
plastic stimuli for testing visuo-motor cognitive processes. For this reason, we know exactly what we would require
from a 3D printer, and I am hoping that you guys will be able to offer valuable feedback, or at least point us
to the right direction.

Not looking for :

-Anything that produces objects larger than 20x20x20 (cm)
-Colours (indifferent)
-Huge material range (indifferent)


Looking for : (preferences numbered in order of importance)

1. Quality of build (identical replication consistency is a must)
2. Non-warping
3. We will be using force sensors inside the printer objects - a printer that can print hollow objects and close them sounds ideal
4. Possible ease of use (the printer itself, we don't mind about the design software, that can be learnt)

Other details :

Budget range : 2,000-3,000 (either £ or $) (approximately, will consider flexible deviations from that price range)
Printer size : Desktop, or even larger (if desktop does not meet the above requirements)

Apologies for the amount of information, but that may actually help anyone willing to aid us in our selection.
We would also appreciate some information on what type of printer would best suit our needs (SLS, FDM etc)

Thanks and looking forward to any replies!
Reply
#2
Hello Diterpenoid,
you may want to check out my answer to http://forum.3ders.org/showthread.php?tid=161 .

That sounds like a really interesting application, if you are willing to share some info, I'd be interested.

Most cheap printers have a smaller print area, what would be the minimum? You can also glue things or turn them 45 degree to fit them on smaller build areas, that is how I do it.

As for your "wish-list":
1- Both reprap based printers and the open source software have evolved into great products. With enough tinkering the quality can match commercial devices.
If you are looking for something tinker-free, I can recommend the UP from PP3DP, but it may not suit your requirements (works best with ABS and has a smaller print area then some printers). But the quality is great.

2- Print with PLA. Other then with ABS there is little to none warping. PLA does get soft at lower temperatures though, so it also depends where you are using the parts. If used in an environment where it gets over 70 degree Celsius such as motor holders it may result in problems.

3- Most printers are able to print hollow, even the UP now in the latest software version. Before that Reprap could do that without problems. There you also have the advantage of automatically letting the print pause when reaching a certain point, but you may have to create a script to do that (not aware of what programs can do that atm).

Other then pause the print and insert something it is often easier to create half-parts and glue them together. Sometimes a lot easier, especially if you make a lid you can remove for testing and then glue it shut when everything is working a 100%...

4- This is a bit of a point-of-view thing. Of course you can buy most printers assembled, but it is helpfull if you build one yourself so you are able to understand and repair it. Then it's easy to use :-) Also it has the benefit of repairing it yourself. For a commercial printer like the UP or bigger commercial printers you will have to pay for replacement parts and/or get a service contract...

Soft- and Hardware need to be set up perfectly, and sometimes you have to change things for every print. Sure, it can be learnt, but keep in mind that it takes up quite some amount of work-time to find the right settings for a new part, especially the first months you own your printer, and then again when buying new plastic from a different vendor or production line... Probably one reason the UP is so easy if you keep buying the OEM material, but it costs two to three times as cheap no-name plastic...

Here is a comparison of the settings you have to make,
Slic3r (open source, for reprap printers)
http://www.ringohr.de/tmp5/slic3rSettings.htm

PP3DP UP's software
http://www.ringohr.de/tmp5/UPSettings.htm
While it is easier, it limits your tuning possibilities... though in 98 out of 100 prints the defaults will do :-)


What kind of parts do you want to print?
Please share an example of what kind of quality you had in mind. It will help to recommend a suitable printer. As for the technology (SLA, SLS, FDM/FFF) read the topic in my first link, I tried to summarize some up- and downsides of the different methods.
In general FDM/FFF/Extrusion is a rather clean, hassle-free and cheap&fast solution, while powderprinting/SLS has advantages when printing a higher number of parts.
If you print a single part on a SLS printer it will take long and waste lots of material.


Even under 1000£/ €/$ there are great devices available.
Also check out the Make 3D printer buyer's guide
http://makezine.com/3d-printer-guide/
Reply
#3
Wow Marcus, thanks so much for such a detailed and comprehensive reply!!

To provide more specific details :

The nature of our research is having patients with brain damage grasp objects. The objects will be small, possibly cuboid in nature, generally geometric. We will insert force sensors inside the objects so we can measure the force they use to grasp the object, but we will alter the objects' appearance (hopefully texture) to induce an illusion of weight. In other words, we are testing to see how participants use their perception to estimate the weight of an object. The reason we need a 3D printer to begin with, is that we are also monitoring hand motion using magnetic sensors attached to their hands, which unfortunately are prone to artefacts when in close proximity with ferrous metals. So we need plastic stimuli, that can be made with exact specifications, dimensions, and variations.

To sum up, we will not need to produce massive amounts of objects, possibly a dozen per experimental design. We do require high detail, because even the smallest variation between objects can be food for negative criticism on our results. The size will never exceed the average grasp of an adult human, so maximum 15cm (so the maximum cuboid shape we will require would be 15x15x15, to give the impression of a cube).

I would definitely be okay with difficult software, but I would really prefer something that comes somewhat ready out of the box. (although I am aware of the inevitable need for me to learn all the parts and how to repair them.)

My employer's suggestions were : Replicator 2x, and CubeX by Cubify. However, any other suggestion (or one of the two mentioned), are very welcome!

Thanks again for the amazing feedback!
Reply
#4
Interesting!
Reminds me a little of http://www.3sat.de/page/?source=/nano/me...index.html (german science tv magazine, Professor Joachim Hermsdörfer from Munich, germany, http://bewegungswissenschaft.sp.tum.de) where they used motion sensors inside a box to mesure pressure, motion and precision when testing how children, adults and seniors pick up a object.

If not producing a big ammount of parts I think plastic extrusion would be the way to go.
Too bad about the print size, the UP is limited to 14x14x13.5cm (minus raft border but this has been adressed in more recent versions).

The Replicator seems like a nice machine, I haven't seen it myself though. Have you read the make magazine?
Print result Replicator 2
http://makezineblog.files.wordpress.com/...cator2.jpg
What annoys me is the fact that Makerbot went closed-source. Of course they brought innovation to the Reprap project, but they also did benefit from it.

I'm not a fan of the cube, but perhaps that's just me. As far as I can see it's a reprap packed into a commercial case, and expensive plastic filament in cartridges. The smaller cube has only "OK quality" according to the Make-3D-Printing magazine. I do not know about their bigger printer, but I would advise to stay alert, if it does not meet the quality you need, you should consider a different device.
Print result Cube
http://makezineblog.files.wordpress.com/...e_cube.jpg

The best method to pick a printer is to go out there, visit a hackerspace, technology/plastic trade show or such. Perhaps pick up some print samples.
Keep in mind that every owner and company will sell their printer as the best deal...
...probably like myself with the PP3DP UP, but the reviews acknowledge, it's a great printer.

All plastic printers will have some variation when printing, but I think if you're not trying to get 0.01mm accuracy it'll be fine.

To throw another idea out there:
2,000-3,000£ is a lot of money, though limits you to the the devices mentioned or similar machines. It would be enough though to get two machines if speed and a backup device would be important.

No mater what printer you choose, I hope you'll find a device suitable for your project. Good luck, and let me know how it works out :-)
Reply
#5
Thanks again Marcus for the follow-up!

Indeed, what you have seen in the German experiment is virtually what we are doing in our lab. As you can tell, we are keen to make it top-notch, and for a neuropsychological lab to have and control its own stimulus production is something that will kick it over the top.

The bad thing with the replicator 2x, is the numerous amount of complaints on constant extruder failure and replacement, which has left many forum commenters in frustration.

I tend to agree on the UP!, quality looks fantastic, and the price is indeed very tempting (for a lab's budget anyway). How are you finding yours? Is it difficult to wrap your head around it, software and hardware-wise? I guess we will have to purchase a 3D modeler software separately though, right?
Reply
#6
Sounds very interesting.



I have no experience with the Replicator, what I've read is mixed.

I have the UP (bought directly at PP3DP.com over 2 years ago) and use the white OEM ABS material (also from PP3DP directly).
I only print small models, so the higher material cost is not much of an issue (plus keep in mind that even dense prints are not solid but filled with a structure, so a 1cm³ cube does not use 1cm³ of material).
Others have reported to get great results with cheaper plastic from other vendors, but it may cause trouble when you send in the extruder for repair.

I don't go through as much material as other people, but I use it very frequently (daily earring project, and other hobbies...).
So far I had neither an extruder jam nor any other hardware failure.
The one issue I had at the beginning was resolved from pp3dp rather quickly. Most UP owners seem to be happy, though I've read two reviews where customer service wasn't as good. Also the cheaper UP Mini had some issues with bed leveling, at least the first devices.
Both of PP3DPs printers have overall great reviews though.

I posted the screen-shots in my other reply, as you can see the UP software is very straight forward. When I first got it I had no clue about FFF/FDM printing, but almost every print turned out great. This encouraged me to learn and experiment more, as there is little frustration involved... The UP just works.

The hardware is sturdy, assembly is limited to tighten a few screws...
THIS printer is the closest to home printing as it's currently possible.
99% of the work-flow is 3D modeling, as the printer does what it is supposed to do.

After one year the warranty expired, I could have bought the warranty extension from PP3DP for another year ($200), but I chose to invest it in a Printrbot instead. The UP is still running, and I now have a Reprap based machine to tinker, experiment and work with. But right now time-usage is more like 30% tinkering with the machine instead of running it ;-) But now it's calibrated and set up so prints turn out quite nicely without much trouble. I enjoy the flexibility open source hardware and great open source software such as Slic3r gives me.

That's why I recomend the UP for everyone that just wants to print, and can live with the downsides (no self-repair, no tinkering with the electronics, and no 1000 settings to fine-tune).
For all other people I recommend a reprap based printer, it's a fun hobby, and if spending enough time, it can be used professionally too.


I love my UP, but please also consider other reviews. There are a lot of nice printers, all have their up and downsides. All I can share is my experience with a entry-level commercial device and a cheap DIY Reprap kit (the Printrbot).
At the end, it's your choice :-)


Software:
Yes, the UP has a interface software that can open STL (and it's own UP files).
Almost every 3D software can save/export STL models natively or you can use plug-ins (or third party tools such as Meshlabb).

You will need a software to create your 3D models, lots of free and commercial software available out there...

-Marcus
Reply
#7
Thanks again Marcus, you definitely seem to know a lot about FDM!

As I read on various methods, I am kind of leaning towards stereolithography, for some reason it seems much more suitable for scientific demands. Most importantly, the detail looks unparalleled. Resins are extremely expnsive, but my eye caught "The Form" that comes out in June, and the company is stating that their price per liter of resin will be around $150, and not in the $300-$800 as the market is right now.

http://formlabs.com/pages/our-printer

Should you have any input on it, would truly appreciate it!
Reply
#8
Yes, stereo lithography is really nice, I would like one of these printers. My hobby budget is low right now though :-)

I have not used any machine like that before, but I would guess the cleaning and handling the resin has some downsides (cleaning the printer, storing the resin, perhaps even health issues depending on the type of resin...).

Form 1, B9Creator and OpenSL are promising devices to keep an eye on. I'm currious on what these printers can do out in the wild instead of on the promotion photographs.
I don't recall if it was Form1 or B9Creator, but I remember reading someone writing that the prints do not look as good as advertised (yet), if I find where I read it, I'll post a link. But I am sure the production machines will produce amazing prints.
Though something like the Kevox SP (rumors say, AUD$17,000) or photonic-professional-gt would be awsome for the earring models I am making (though for that I won't buy one :-))


Do you have any sample shape you could post or something from thingiverse with a similar ammount of details? Just so it's easier to tell what kind of printer is capable of doing so.

I printed a 3D model of mars on the UP, it turned out pretty well, as well as 3D Scans. Pretty consistant quality, too. The printrbot isn't doing half as bad, either.
Reply
#9
... in a German CNC-forum we're testing with UV-modified DLP-beamers, accuracies/layer-heights down to 10 microns (see next 3 images in the linked post) and resins for around 100€ per kg from spota materials.

You have to register there, but I've linked some of the images here too: http://forums.reprap.org/read.php?279,17...msg-179881

I'm testing with other methodes and differerent UV-curing devices (BluRay-UV-laserdiodes @200mW, 405nm-LED's @9Watts, 337nm-TAE-laser with 200ps short pulses @400kW), but this is more testing for max. limitations of the DLP-chips and sometimes with devices not common in the DIY-scene - so can need some time until mainstream ...

Viktor
Reply
#10
That sounds very interesting, unfortunately the forum is down at the moment :-(

@ Diterpenoid
I forgot to mention the patent issues with Form1,
http://www.wired.com/design/2012/11/3d-s...s-lawsuit/
I am not aware of any news on these issues beeing resolved other then
http://formlabs.com/blogs/blog/6974594-o...e-priority
Let's see what will happen Undecided
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)