100 vs 200 microns layer resolution

[Jeremy Rosser]

Is there a huge difference between 100 and 200 micron layers?  I have the replicator 2, but the new mini looks amazing.

Please let me know if I would be taking a huge step backwards.

Thank you,

Jeremy

[PrettySmallThings]

It depends on what you're printing - and what matters for you.  I print most of my designs at 200 or 300 - I'm in a hurry, and generally these resolutions are more reliable.  I wouldn't miss the high resolution options, but I'm sure there are folks who would.  

[Jetguy]

This makes no sense, and I'm not talking about the original poster's question, I mean the specs that they limit the mini to less vertical (AKA Z) resolution. That is totally bogus.

I know for a fact from some inside information that BOTH the mini and the other new 5 series should be using the same mainboard controller. Also knowing how parts are sourced, they both would use the same basic motor and leadscrew in Z axis. The Z axis is higher resolution than even 50 micron so it's NOT in theory a mechanical limitation, but rather some lame marketing to lower expectations and steer users who don't know any better towards the more expensive machine.

Hands down, if they were selling the mini today I would buy one just to hack the snot out of the new controller. This is the one thing not yet getting much attention and it's because they don't even list it in the features or documentation.

even at CES they totally missed this point. A  new processor with more horsepower = even smoother motion. Not only that, but that is how they were able to integrate the filament detection stuff into the main firmware because the processor has much better IO and more of everything. This could be a major revolution in the printing world.

Basically the parts you really, really, don't want to have to upgrade or mod EVER on a bot are the extruder and the controller. the motion mechanics are the easy and cheap part.

So if the mini uses the same extruder and the same controller minus the over the top stereo knob and LCD controller (how soon do you think we'll have that hacked out on a mini?) you'd be a bit silly to buy the 5 series and spend on that money.

It's a steal for the price. Again, it has the new extruder, the same basic setup as the new 5 series but is missing the LCD and is a little smaller build area.

I vote for a why not, at that price, it's going to be a reasonable printer and very hackable in capable hands.

I'm right in the middle of hacking a CubeX, what makes them think we won't have this figured out in 3 days after release?

[Rob Griesbeck]

It depends what you're doing, but I didn't notice much difference. I've been printing almost exclusively at 200 and it looks great! Here's a prop kit I've been making and selling and everyone is very happy with the results - http://www.therpf.com/f13/fully-3d-printed-hellboy-samaritan-kit-202333/

I know when I got my 2X the 1st thing I did was try 100, and it looked good, but I found myself, mostly to save time, going to 150 and now I've pretty much stayed at 200 for almost everything.

I agree with Jetguy that the mini seems a steal. And while I have no proof, I'm feel the 200 microns might be a software limitation, not a hardware limitation. It doesn't make a lit of business sense to make 4 other printer that are capable of 100 and manufacture a totally different set of components just for the mini.

[Bryon Miller]

The quality looks a lot smoother, but takes longer to print.  I guess I don't understand what's so great about the new printers, but they are very expensive IMHO and have useless features like the webcam that I would never use.  They appear to print the objects just the same as the existing printers.  I personally would never use the Wi-Fi features, I'm sure there's a use, but not for me, I have to be at the printer to remove the parts anyway.  Do the new printers print any faster?  The new extruder head looks interesting, maybe they'll have one to swap out with a food safe extruder?

I don't think you'd be taking a huge step backwards, I think you'd be forking out a lot of money for a machine that does exactly the same thing as the one you already have, I don't think they'd go through the hassle of using a seperate z rod just for this printer, so the limits are probably in software for the .2 micron.

[Rob Griesbeck]

Agreed. As far as I know, and I've talked to Makerbot sales a few times the new printers are no better quality or speed. Just lots of new "features" or "bell & whistles". You can buy a standalone IP webcam for <$100 and set it up by your printer, which is something I may do to keep an eye on my growing "farm".

Everything I've seen led me to order a Rep 2 from Amazon today. I like the new printer, but at +$700, I just can't justify it.

[Justin Keenan]

I believe that the new Mini might be limited to 200 microns because of the build plate. Since the build plate doesn't require any leveling, it might have issues printing at 100 microns reliably (as in several layers up you may experience printing lines or problems of it not adhering to the build plate). So all in all I agree with the software restricting the layer height but not due to marketing or timing reasons.

[Jeremy Rosser]

This might sound like a silly question but how do you tell the printer to print at 100 or 200 microns? I use makerware to create the X3G file and the only options I use are the Low , Medium and High quality. Does that equate to the micron quality or do you need to go into advanced settings?

Thank you all for your replies.

[Bryon Miller]

It is "layer height" in the slicer.

I think High selects 100 micron.  But You change that setting to .1

[PrettySmallThings]

You'll see more options if you hit "advanced settings"

100 = high, 200 = medium, 300 = low

[KM Design]

Ironically my favourite resolution to print at is 150 microns (0.15mm), right between the two. Like Pretty Small Things I'm a bit impatient so 100 microns is too slow for me. I'm mostly printing small objects. I try and orient them to take advantage of the vertical resolution. Unfortunately its seems like top layer horizontal surface finish is one of the tweaks Makerbot hasn't focussed on yet.

Kevin

[c f]

I usually design in inches, so I typically print with 254 micron layer
height (0.01").

> --
> You received this message because you are subscribed to the Google Groups
> "MakerBot Operators" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to makerbot+u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

[Rob Griesbeck]

+1 to 150

[Benji EOD]

I contacted Makerbot about the Mini. They said that you can adjust it to print at 100 microns. Out of the box with their one touch easy self leveling setup it is ready for 200 micron levels. They are marketing the Mini at consumers who want a simple process.

[Jeremy Rosser]

That is great, same quality as the big guys.

[Greg Fisk]

I used to print everything at 0.1, because it comes out so perfect.
I am starting to prefer 200 a lot because the lines add a little character.
I'm also getting lazier and just using standard makerware settings, rather than my custom RepG stuff.
I'll have to try 150...
I like the 'almost' injection molded feel of the prints I used to get with my white plastic. (The new stuff I have is nowhere near as good. )

Actually, that's probably the main reason for my change - the plastic seems to make a big difference to how much the resolution matters.
That roll of white was totally amazing. My black is pretty, but certainly doesn't look injection moulded even at .1
My blue might as well print at .3, it always looks crap.

I print a lot of smaller threads when I am using it for work purposes, so I usually use the higher res to get better threads. (Not certain if that makes much difference on a 1mm pitch thread or not).

[Greg Fisk]

I know nothing about the rep-raps, but since they have such a big community, what is stopping them from printing at 0.1 or lower? I would have thought they would be ahead of a commercial machine?
I personally think my rep2 is a very expensive pile-o-crap (which does amazing things when it's working right, but build quality and quality control of the machine is terrible, and it seems to need a lot of upgrades to make it nice)

[Billy Zelsnack]

On Friday, February 14, 2014 6:03:47 AM UTC-5, Greg Fisk wrote:

I know nothing about the rep-raps, but since they have such a big community, what is stopping them from printing at 0.1 or lower? I would have thought they would be ahead of a commercial machine?

Printing at 0.1mm is not really that challenging. It's the Makerbot marketing department that makes a big deal out of it. 

[wjsteele]

One other thing to note about the Mini is that the Z axis drive is entirely different from their previous designs.  It is not based on a lead screw mechanism and guide rods, instead, it has the lead screw and two inverse gears that ride in racks which all appear to be injection molded.

Bill

[Bryon Miller]

All the FDM machines I've seen in the last few years can do .1, they can usually do down to .035.  Until they have a way to idiot proof the leveling process, the tram and level is what keeps you from getting decent prints at those res.

Have you seen the other machines up close and personal?  I only know one other person personally that has a 3d printer.  He bought the solidoodle 4 that just came out.  I'd much rather have my replicator 2 than that machine.  Once it's upgraded this machine is awesome.  What I don't like is MBI seems to think the year is around 1980 and they're a small successful startup named after a fruit that can overcharge for the products they produce, or at least they seem to follow the same pricing structure.

I'd buy another rep 2, but I'd get the clone that looks identical for $800 from the flash forge guy.

[Jamesarm97]

Here is my example ;) sailfish on mighty board with FrankenPrusa

.25 vs .025mm

http://www.flickr.com/photos/jamesarmstrong3/9619266748/

[Darrell jan]

Nice! I've only tried as low as 0.05mm. And could hardly see the difference between that and 0.1mm, with the naked eye. Since it takes way longer to print, I almost never go to that resolution.

[Joseph Chiu]

When I did some 0.1mm ABS prints last year and ran it really hot, the layers actually melted into each other so that the result was approaching injection-quality.  

--

[delsydsoftware]

The only time I could ever see a difference below 0.1mm was when I had the filament diameter exactly dialed in. The lowest I ever got with the Rep 1 consistently was about 0.02mm per layer in ABS. That was at sub 20mm/sec movement speeds on very tiny guardian squirrel prints.

[Johnny Corvus]

Plenty of repraps and even many mass-production printers are capable of going below 0.1 mm, but there's a huge difference between coming up with settings that work just right some of the time for a single semi-custom machine whose operator knows its every quirk because he made it, and a reproducible solution that always or at least mostly works for many machines coming off an assembly line whose operators expect to be able to push a button and stuff happens.

[David Kessner]

In my opinion, there are huge reasons to print at 0.1mm instead of 0.2mm.  Not everyone needs these, so your mileage may vary.

Imagine a sloping wall, where the slope is almost causing an overhang.  Consider the case where the wall is sloping away at 45 degrees.  If the layer height is 0.2 mm then for every 0.2 mm you go up then you also go out 0.2 mm.  When the extruder puts down the perimeter bead, 0.2mm of that is hanging in mid-air off the edge of the wall.  The bead that the extruder is putting down is about 0.5mm wide, so about 40% is hanging off.

Now, redo the calculations for a 60 degree slope (more of an overhang than 45 deg).  Now instead of 0.2mm hanging off, you have .35mm hanging off, or about 70% of the bead is unsupported.  With this much unsupported, it tends to droop and causes an unsightly finish at best and at worst fail completely.  A simple fix for this is to switch to 0.1mm layers.  With a 60 deg slope, the 0.35mm/70% that is unsupported goes down to 0.175/35% resulting in a much better finish.

So, with 0.1mm layers you can print objects that can't be printed at lower resolutions and those objects will look better.

On my Rep2, I find that at 0.2mm a 45 degree sloped wall is the limit but at 0.1mm a 60 deg slope is easy.

-David K

On Monday, January 20, 2014 6:45:50 AM UTC-7, Jeremy Rosser wrote:

[Jay]

Exactly Johnny....I can get my R2 to that resolution but then It's crap for anything else. Since most of the stuff I print is .2-.3 I leave it set up for that. The first time I did it I was excited...then I thought about it....5-6 hours for a print that I can do in a lower resolution in about a third of the time. What would you show off? Any cool print at that resolution is not a spectator sport....most of my friends don't want to spend almost an hour watching a stretchlet print...but don't mind 15-20 minutes (that's about what a beer takes to drink). The ones who don't mind watching are the ones that tend to drink all my Yoohoo's anyway...Grrrrr.....

THAT is the biggest problem I have with a lot of bot makers.....they tend to gloss over kind of set ups and times it takes to make some of the sample prints they display. Oh well...I'm still good with .2-.3 prints so this isn't a big issue with me.

Jay

[Joshua Fairbairn]

when is the mini going to be released for shipment?

[Ryan Carlyle]

Hopefully AFTER they fix the extruder design shared with the 5g and Z18.

[Enginwiz]

Stacked infill is the turbocharger for high resolution prints.

100 micron perimeter with 300 micron infill prints nearly as fast
as conventional 200 micron resolution.

Especially on rounded parts the difference between 100 and 200 microns looks stunning.

[Scott K]

I have considered trying this for several weeks but just haven't found the time.  I think this is going to be my project for the weekend.  Glad to hear it has worked so well for you!

[Ryan Brodowicz]

That is a great idea varying the resolution for internal and external. What settings would you change to do this on a 5th Gen? I do not know the custom profiles too well yet.. I think I will experiment with this also this weekend! 

On Wednesday, May 7, 2014 4:28:54 PM UTC-5, Enginwiz wrote:

[Ryan Carlyle]

Can't be done on a 5g, Makerbot's slicer doesn't support it.

[Ryan Brodowicz]

wow. That's some awesome new tech right there. Latest and greatest cant even do what bots have been doing for a long time... 

I'm going to call MBI support this weekend and see if I cant get a way to configure this out of them, if not... damn. 

[Dan Newman]

On 08/05/2014, 7:19 AM, Scott K wrote:
> I have considered trying this for several weeks but just haven't found the
> time. I think this is going to be my project for the weekend. Glad to
> hear it has worked so well for you!

The other fun one to try with a dual extruder bot is to have a 0.5 or larger
nozzle. Then in S3D assign it to doing the stacked infill. That makes it
faster on two counts: higher output and thicker layer height. Of course,
there's the question of the extruder heater being able to handle a higher
flow volume per unit time.

Dan

[Joe Larson (aka Cymon)]

Don't be so excited for this thing you've never heard of before. In the previous implementation it actually just halved the outer layers by interpolating the geometry of the current layers and the one that came next. So you weren't actually gaining fine details like you would if you decreased layer height proper. You just make your existing chunky layers smoother. Makerbot probably didn't implement it because a proper implementation would be an entire rewrite and the gain wouldn't be that big. Fill layers go pretty fast. Still, send an e-mail to thou...@makerbot.com and who knows.

That's a pretty good idea, except by specialzing your nozzles you loose the ability to use them both together for dual-colored prints. Still, I've sometimes thought it would be fun to have a nozzle for the outside and another for the infill and use junk plastic for infill, hiding it away. Would be a good use for that HTPE filament you're making with your filastruder and recycled milk cartons. Of course then you can't use disolvable support material.

What I really need are 4 nozzles. 2 for color, 1 for infill and 1 for support.

[Enginwiz]

Locking the Gen 5 users in deep down in the Makerware dungeons with the new Json print file format
was a stupid move.

Even the CUBEX users eventually came up with a jailbreak and can now use Kisslicer.

Maybe somebody has to do the same to Makerbots new Json print file format and develop something
like GPX for Json. This would allow the Gen 5 owners to use advanced slicers like Kisslicer and
Simplify3D.

[Dan Newman]

On 08/05/2014, 12:36 PM, Joe Larson (aka Cymon) wrote:
>
>
> On Thursday, May 8, 2014 11:03:29 AM UTC-6, Ryan Brodowicz wrote:
>>
>> wow. That's some awesome new tech right there. Latest and greatest cant
>> even do what bots have been doing for a long time...
>>
>> I'm going to call MBI support this weekend and see if I cant get a way to
>> configure this out of them, if not... damn.
>>
>> On Thursday, May 8, 2014 11:39:59 AM UTC-5, Ryan Carlyle wrote:
>>>
>>> Can't be done on a 5g, Makerbot's slicer doesn't support it.
>>
>>
> Don't be so excited for this thing you've never heard of before. In the
> previous implementation it actually just halved the outer layers by
> interpolating the geometry of the current layers and the one that came
> next.

That's certainly what Skeinforge did. I've not looked at this in
MakerWare -- I'm not much of a MW user to be honest. Note, however, that
slic3r and Simplify3D actually do use real slicings at the finer layer
height: they do not do the linear interpolation with coarse layer heights that
Skeinforge was doing. Cura and KISSlicer may also use finer slicings --
I'm not sure.

Dan

[Ryan Carlyle]

The new JSON format is super easy to understand. It's literally just a list of coordinates and speeds. Completely human-readable. I don't think it'll be hard for 3rd party slicers to export to the 5g at all, but I'm far from an expert on this. 

[Scott Goldthwaite]

What needs to be changed in the profile config to do this (using ReplicatorG)?

[Qays Poonawala]

Who did you talk to at MakerBot? Their support agent told me this was incorrect and was curious to know who said this, so the agent could double check with whomever you talked to.

On Thursday, February 13, 2014 3:24:50 AM UTC-8, Benji EOD wrote:

I contacted Makerbot about the Mini. They said that you can adjust it to print at 100 microns. Out of the box with their one touch easy self leveling setup it is ready for 200 micron levels. They are marketing the Mini at consumers who want a simple process.

[lassi kinnunen]

most of this thread is very old. I guess jetguy would have different opinion now, and it wouldn't surprise me if the changes to specs came later.

basically, going from rep2 to mini is no upgrade..

-lassi

[Steve Johnstone]

Hi Joseph,

How hot is really hot ?

[Musk]

How many models actually have unique details at .1mm resolution?  I can't think of a model where you'd notice unique .1mm details.  Maybe jewelry is an exception, but if you were doing small jewelery you probably aren't using FDM and you definitely aren't worrying about stacking the infill.

That said, I do want the slicer to give the actual detail contained in the source part. :)  Interpolation is like upscaling a dvd instead of having a real hdtv signal, which is lame.  I just don't think you see the extra crispness at .1mm fwiw.

[Musk]

Just to be clear, I'm talking about interpolated .1mm vs true .1mm just so you guys don't think I'm nuts.  I think the .2 to .1 difference is certainly noticeable.

[DonaldJ]

To really appreciate the detail at .1mm layer height you'll need to use a .25mm (or smaller) nozzle.

And that creates a whole new set of issues with direct-drive extruders...

[Joseph Chiu]

250C with the bot fully enclosed all around.  The printed piece was small, too, so the layers had very little chance of cooling before the next layer went on top.

[Steve Johnstone]

Cheers Joseph, I will definitely give this a go.