How much $$$$ on a printer that can produce customer facing quality?

[Recypher]

Hi,

I am currently working on a project that is basically a torpedo shaped object made out of ABS, it will measure roughly 25cm x 9cm…printed in two halves.

It’s a simple shape, so I don’t need really intricate details however it will have to be accurate as there are control surfaces that will affect the way the object will move through water.

I am pretty convinced that purchasing a 3d printer is the way to go for my initial prototyping and refinement of design however if possible I would also like to have the option of 3d printing the initial batch of 100 (kickstarter style).

I have time on my hands and am pretty savvy when it comes to figuring stuff with technology out so am not afraid of spending some time sorting/fixing issues (i.e. I don’t have to have a push button & walk away solution).

Would something like an Umtimaker 2 be up to the job of producing my initial prototypes and the run of 100?

If not I am considering purchasing a much cheaper 3d printer such as the Solidoodle 4 for my prototypes and then look to a more commercial solution (3d printing or not) to produce the initial batch of customer product.

I have never even seen a 3d printer so really looking for some guidance here…there is no point spening x3 the money on a better printer if it is still not going to be good enough for a customer facing product.

Any suggestions/help much appreciated.

[Feign]

For "Customer Facing Quality", do you mean that it looks fairly good, or do you mean indistinguishable from an injection molded product? Almost any printer with 0.2mm layer thickness or less will look good, very good, but will still have the telltale grain of a 3D printed part (looks reminiscent of the grooved texture of a vinyl album, but finer) Depending on the orientation of the part on the table it might be a positive feature, but it will always have the texture from a FFF printer.
If you want glossy, smooth results, however, you will either need to do postprocessing (vapor smoothing or sanding and polishing) or use a polyjet (with a six figure price) or SLA printer (which usually have too small a build volume for what you've described) both of which have much higher material costs than filament-based printers.

[soofle616]

Even a polyjet or SLA printer is going to have visible print lines if we're comparing to an injection molded part. No 3d printer is going to produce a finish equivalent to a polished mold. Post processing can get you there if you're willing to put in the time and effort but that's really the only option.

The question remains though, what is meant by "customer facing"? That phrase has to be defined before a decision can be properly made.

[Recypher]

Thanks for the rapid responses!

I guess by customer facing quality I mean if someone backed by project (early adopters) on a crowdfunding website and they received the product they would not look at is and think it looks poor quality...bare in mind the product is functional rather than something to look at.

The solidoodle 4 ($600) layer thickness is rated down to .1mm where as the Ultimaker 2 ($2500) gets down to 0.02mm which I guess is where the extra $'s goes.

Sounds like the Ultimaker 2 might be the right choice as if I am not happy with the finish there would be less to correct in post due to an already finer level of resolution.

[GOC]

- Not sure if the build area is going to be big enough, but I'm very happy with the resolution and quality I get with my Orion Delta from SeeMeCNC. Its also a printer that doesn't completely break the bank at $1300. I have some pictures of some prints on my blog if your curious.

[richardphat]

I'll stop you right there, if you are going to based on the 0.02mm.

• First, it's complete garbage marketing deceiving you.
• Second, any printer using standard Nema 17 will be able to print at 0.02. Most company doesn't want to take the risk to advertise the 0.02mm as they fear to have a backfired from customer, such as ticket entitled "Why is my print taking so long? "
• Third, the smallest resolution you'll use is probably 75 microns or 50microns to an extent, for fliuid dynamic related application. Remember, you get what you paid for. It's a 4 figures machine, expect a lot of calibration if you want to go 0.02mm res..
• Fourth, printing at higher resolution will be useless if you're going to do post treatment.

I work with the Ultimaker 1 since 2012, it's stll working unlike those 5th gen makerbot.

The ultimaker 1 and 2 themselves are really nice machine and worth the money, just don't get fooled by the horse crap marketing.

[Bassna]

It really depends how perfect that part need's to be, but even a cheaper printer (My printer was $1350, Flashforge Creator Pro, and the print's look great like this link's) using ABS with acetone vapor smoothing makes a WORLD of difference. And acetone is very cheap, so is setting up a little spot to smooth part's at.

ABS is pretty cheap, I pay around $20 per 1KG. Acetone is also cheap. Printing out 100+ of those part's will also take quite some time with just 1 printer. Keep that in mind.

http://sinkhacks.com/building-acetone-vapor-bath-smoothing-3d-printed-parts/

[soofle616]

If your concern is a potential customer thinking the product is cheap then you need to spend more time informing your potential customers and less time worrying about surface finish. Set proper expectations and the people that back you will be happy with the products they get. If you go into it telling everyone that it's a super awesome amazing shiny wunderbobble then provide them with a 3d print that hasn't been post processed, you're going to have lots of angry customers. If you tell them the part is meant to be functional first, pretty second, that it's produced on a 3d printer, that the surface finish will have a stepped appearance common to all 3d printed products, and show them lots of pictures of what they should expect, then they will determine if the product at that level of finish is worth the asking price. If they decide yes, you're good to go. If they decide no, they won't buy it.

[Feign]

Yeah, the .02mm layers will be something that you can feel more than see. The thing is that actually printing that fine on an object as big as you describe will take forever.

Heck, I'll run the math real quick: If your torpedo is split exactly in half that's 125mm or 6,250 layers at 0.02mm per layer, depending on your infill (how solid you make it) you might be looking at between 15-30 seconds average per layer (this is just an educated guess on my part, really), which puts you at somewhere in the range of 26 to 52 hours per torpedo to print at that resolution. Printing a hundred of them on one printer would take months.

For an example, a piece of standard post-it note paper averages at about 0.1mm thick. If you've ever felt the texture on a side of a well-compressed post-it note block, you know what that layer thickness feels like.

[awerby]

If you really want 100 of these things, using a service bureau makes more sense than trying to print them all yourself on a single machine. Even at a more reasonable resolution, that's going to take a very long time. And these consumer-level printers are not 100% reliable; there are always going to be glitches in the printing process like filament tangles, clogged extruders, axis jams, etc. A good service bureau will be able to print more at a time on faster, smoother and more reliable machines, and will give you the ability to deliver a better product much more quickly. If you're still working the bugs out of your design, then sure, being able to iterate different versions quickly is valuable, and might justify the cost of a printer. But for production, you'll want help.

Andrew Werby
www.computersculpture.com