Potential User...but I have some questions.

I am a SolidWorks CAD guy for the day job, and recently have been moonlighting and taking on some minor work on the side. I can really see this as being a huge step forward for my side work...not only providing drawings and virtual models to clients, but providing a physical prototype or model of a part that the client can hold in their hands or use for marketing purposes.

As such, I am trying to get a handle on the machines long term costs...as pretty much the only way the wife will let me buy one, is if it can pay for itself over time.

So, 1KG of ABS from MakerBot is about $0.776 per cubic inch, or $.047 per CC. So material cost isn't bad at all. Not in the slightest. (Assuming this uses 1.7mm material)

My first question is what would be the cost of running the machine, per minute or per hour, so I can figure out my cost there. This would include things like electrical draw, maintenance/spare parts, etc. Not worried about operator cost, since once you start it, it doesn't necessarily need to be monitored.

My second question is where to buy a full kit (everything I need to assemble and start printing...frame, extrude, electronics, etc.) I am located in SE AZ, in the United States.

My third question is what is the accuracy of the machine? I'd like to be able to print mechanical assemblies, such as mating pins and holes...or even molds for resin casting.

My fourth question is how hard is it to optimize parts for printing? The Makerbot software will allow one to thin the model and fill it with a waffle pattern to save material.

Any help that anyone else can provide would be greatly appreciated.

I make and sell Prusa iteration 2 in New Zealand and from a 100m roll of 3mm PLA I can print one and a third sets roughly.

The slicing software does all the infill stuff I use mostly 25% fill which gives hex shapes inside with 3 solid layers on the outside, some parts I make 50% fill and some 80% fill which is near enough to being silid.

I design in Solidworks here and just run my .stl files through Netfabb basic afterwards to check the mesh is correct.

My machines run from a standard ATX computer PSU so draw is the same or less than a compputer.

Accuracy i pretty good although holes do shrink slightly often so I run a drill bit through them on my drill press

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

I appreciate your response NelsonRap. As far as accuracy goes, I was sort of hoping for a hardish number, such as 12 microns or .12 mm on Z, and .4 on X and Y axis...it really makes a difference in my evaluation for the sort of models my clients want.

Glad to hear it is easily exportable from SolidWorks. Makes my life much much easier. I am in the process of building a 3d scanning station from a couple spare XBox Kinects I had laying around. Getting some pretty promising results in point clouds, though you need the SW Premium package to gain that functionality with converting point clouds.

How about maintaiance? What sort of spare parts should I be keeping on hand? Bearings? Spare extruder head, etc?

Since I can get most parts overnight I don't actually keep any spares, the only thing I have done is print some spare parts for the ones that might break.

X carriage
extruder
extruder idler
gears

that's about it

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

I've been able to get the Z-axis resolution can get down to 0.1mm reliably, but others like richrap have gotten well below that. As far as x/y resolution it depends on the variables of your printer, including stepper resolution, amount of micro-stepping used by the electronics, belt tooth resolution and pulley size. For example, if you have a 1.8 degree stepper using 1/16th micro-stepping and a 32 tooth GT2 pulley you can achieve a resolution of .02mm (32*2)/(200*16). Or you could up the resolution by selecting a smaller pulley and/or more accurate stepper. For instance, a 0.9 degree stepper paired with a 16 tooth GT2 pulley would give you a resolution of 0.005mm. Note that what you gain in resolution you sacrifice in speed. And since these machines don't have feedback via an encoder errors can be introduced by overdriving the steppers.

Also, don't buy a makerbot. Seems like everyone who has one on these forums regrets it.

You seem like quality is of the utmost importance to you. In that case there really is only one kit suitable to your needs - mendelmax.

Good luck!

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

akhlut Wrote:
-------------------------------------------------------
> For example, if you
> have a 1.8 degree stepper using 1/16th
> micro-stepping and a 32 tooth GT2 pulley you can
> achieve a resolution of .02mm (32*2)/(200*16). Or
> you could up the resolution by selecting a smaller
> pulley and/or more accurate stepper. For
> instance, a 0.9 degree stepper paired with a 16
> tooth GT2 pulley would give you a resolution of
> 0.005mm.

This theoretical resolution has little to do with the accuracy of repeatable positioning of the print head (it only gives the lower limit). In an actual operating printer laying down plastic, a "resolution" of 0.005mm would be almost completely meaningless, because of all the other positioning error sources and the accuracy of the FDM method of object construction.

In a typical printer, an X-Y accuracy of 0.1mm is fairly easily achievable, with the minimum size of printed features depending on the nozzle size.

Yeah, I was thinking about that after I posted it - that the resolution is really a combination of the x-y plane/extruder/hotend quality/software.

Basically, it comes down to the ability to properly tune the machine, which limits to the resolution achievable by the measurement tool of choice. I've seen calipers that can achieve a resolution of 0.005, but the price for that added 0.005mm is about $80. Totally not worth it. $125 for Mitutoyo's that can get down to 0.001mm? Getting closer to what I'd be willing to pay for a highly tuned machine. But I don't have a machine that can get there. My theoretical best resolution is .01mm, so standard calipers work well enough. Is there another more accurate way of measuring these distances besides calipers without breaking the bank?

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

I have the multitoyos....so I can't really help in that regard. Thanks so much for the wonderful discussion. I'm getting more and more sold on this, and definately leaning towards the mendel max.