Next build...

[TobyCWood]

I'm thinking maybe a 350x350x350 Voron. Sourced as a kit from Aliexpress. Swap out he SKR and add a Duetwifi3 and perhaps the new Bondtech...

Anybody here do one yet?

[TobyCWood]

Having some real doubts after studying the manual. For example, the 4 separate steppers which drive belt loops which then drive gears that drive long belt loops that controls the Z axis. All four must be in perfect synch on each and every change in Z.

and...

I think Id rather build a large Delta, but is there an open source design like the Railcore or Voron???

[Gary Tolley - Grogyan]

The Voron, is a well engineered design from those that have one.
I'd like to build one, but my house is too small to have a 3rd printer

--
You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/db7ed935-a463-47dd-a16a-b1d94696dcfan%40googlegroups.com.

[James Fackert]

I would adapt the artist d instant swap hotels setup into anything I built.

Clean fast and easy.

If it had a better heated bed design, it would be a great printer.

On Mon, Mar 15, 2021, 5:14 PM TobyCWood <andyc...@gmail.com> wrote:

I'm thinking maybe a 350x350x350 Voron. Sourced as a kit from Aliexpress. Swap out he SKR and add a Duetwifi3 and perhaps the new Bondtech...

Anybody here do one yet?

--

You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/62e4e056-4254-4db4-b688-1e9dc1684cb1n%40googlegroups.com.

[LukeH]

Having gone through this research phase recently, there is no “Voron”-like design for Delta, except the Kossel design, which has been updated slightly, but is more or less the the same as it was in 2012. Individuals who build them do a lot of design themselves to update the design for their own purposes. 

There isn’t a single, actively developed, open source Delta design like Voron, but that said, the Kossel design is so straightforward, modifying it isn’t a huge challenge (says the engineer who hasn’t got his up and running yet, but that is mostly because of shipping delays and external issues, not technical issues). Obviously the most popular Kossel design changes are to turn it upside down (so that the steppers and electronics are at the top to help keep them cool, not trapped in a box underneath the heated bed), and to use linear rails, rather than wheels in v-slot extrusion. The rest is just careful selection of bits. 

If you do plan a delta, and are using a Duet board, I would start with the Duet Smart Effector, which uses a 55mm rod spacing, and design backwards from that. It will drive many of your part selections and sizings, plus some basic math. There are also a few Kossel calculators on the web to help with the design. I started with a different effector solution and now regret the decision (and am trying to figure out how to stick a Smart Effector on using a printed adaptor plate).

I looked long and hard at the Voron design, and if I wanted another CoreXY, then that is definitely what I would have built. I even borrowed/adapted some design aspects for the large Delta.

[Tim Keller]

I wonder how adaptable the VORON would be to be a tool changer? I have an E5P that works so I don't really want to screw with it, but every time I listen to a 3d Printing today pod cast and Andy crows about his toolchanger I get the itch to have one... just not a 4k itch mind you, but an itch none the less..

Tim.

--

You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/afb2b3a9-4778-401c-9266-622f00bc128dn%40googlegroups.com.

--

I am leery of the allegiances of any politician who refers to their constituents as "consumers".

[TobyCWood]

Seems to me the approach E3D did with the tool changer is best done with a machine that keeps the X/Y gantry at the top. I have seen some TC approaches with a Delta, but these were experimental. As such the Voron would need some compromise such as going to the top to make each tool change.

Also... there's no way I would want a TC without Duet controllers. You have to have as much flexibility in the setup as possible as well as decoupled scripts for wipes and purges. I.e., not in the slicer nor the resulting gcode, instead an external script.  I'm finding that purges are a function of the material. PETG needs a longer purge than PLA. Sometimes these need to be tuned on the fly.

[Mark Napier]

I know there is no way you are going to watch a 3.5 hour livestream.  I didn't either just skipped around and got the high points.  The Bondtech LGX is a perfect fit to replace the standard Voron clockwork direct drive (had to be designed that way).  This guy put it on his big Voron with a high flow mosquito clone.  At the end it seems to be working well.

https://www.youtube.com/watch?v=v3B2TDN1moY&list=LL&index=2

On Mon, Mar 15, 2021 at 4:14 PM TobyCWood <andyc...@gmail.com> wrote:

I'm thinking maybe a 350x350x350 Voron. Sourced as a kit from Aliexpress. Swap out he SKR and add a Duetwifi3 and perhaps the new Bondtech...

Anybody here do one yet?

--

You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/62e4e056-4254-4db4-b688-1e9dc1684cb1n%40googlegroups.com.

[Luke Hartfiel]

The LGX looks the business, but once you buy the extruder ($100), short distance heat sink ($70), and Mosquito hot end ($150), you are talking about US$320.

Obviously not something you would strap on to a $180 Ender 3 Pro… :)

On 18 Mar 2021, at 2:45 pm, Mark Napier <napie...@gmail.com> wrote:



You received this message because you are subscribed to a topic in the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/3d-printing-tips--tricks/p640c32ih9g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to 3d-printing-tips--...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/CAFS2mBeR901jYCv%3DDn-Xf5i-m7d45pC75eR1wxxjJfYnVbNbUA%40mail.gmail.com.

[Ryan Baird]

I have been doing a whole lot of research on this for my next printer and first full build. The printers i have been looking at are;

Home - HevORT Advanced DIY 3D Printer

RailCore II 3D Printer · RailCore

VORON Design

ToolChanger & Motion System Bundle inc Tools & Hemera Extruders – E3D Online (e3d-online.com)

After spending a silly amount of time staring at these printers I think I have narrowed it down to the HevORT. it checks all the boxes its big draw back is smaller community. but going for it is the ability to use the tool changer approach to tool changing, the 3 z stepper true carriage alignment. ( i have an issue with calling it level because its not what your doing) I see the focus on speed as equal to rigidly. I am playing with the idea of using Stratasys recently expired patent to enclose the build area. 

[TobyCWood]

The HevORT looks killer. I really like the 3 ball screws. I wish there were more pics.

I have found with my CloneR1Duet that when you go over 100mm/s the effects come from the momentum of the material itself. The stuff is almost a liquid and it's flying around.

[TobyCWood]

Its kinda like they saw the issue of the 3 lead screws on the Railcore with the Smart Effector and found a solution that Railcore could shoulda done.

I can't help but wonder if one could use it to do printing on off angles with gravity favoring the use case object. oooooo!

[Jonathan Jones]

I'm pretty sure the HevORT developer has talked in some of his videos that he would love to do some off angle printing but the software support is lacking at the moment. 

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/1a0f09b8-7e12-41f7-8dd2-7f06c5f7c4dan%40googlegroups.com.

[TobyCWood]

Just did a search on Ali to see if there's parts sets. None, but I also notice from the search field pop down that others have done the same search. No doubt the parts suppliers overseas will see the opportunity real soon.

[Luke Hartfiel]

The fact that you can’t buy a parts kit straight off Ali shouldn’t be a deal breaker though - if you have a full itemised BOM, I’m sure you could buy all of the components online (even cut to size, in the case of extrusions), and most of it from Ali.

It might be a pain to search and add each component to your cart, but it wouldn’t take more than an hour or so, and you get the advantage of having a choice in the quality of the components (choosing where you might save your money, and where you might want decent quality), plus you can pick sellers with a reliable track record for individual components. You can also choose where you will be using genuine parts from Duet, BMG, E3D, etc, or where you will use clones (there is no point spending piles of money on a new printer, then sticking in a low quality clone V6, although you might decide to use a good quality clone, and spend the difference towards a genuine Duet 3, rather than a clone Duet 2, for example - it would be entirely at the user’s discretion).

It also spreads your risk, since you aren’t relying on any one seller for everything, and allows you to stage your purchases, so that items arrive when you need them, which may give you the option of putting off selection of some components for later stages of the build, until after you have seen and commenced work on earlier stages of the build, since your approach might change as the build progresses.

On 19 Mar 2021, at 4:16 am, TobyCWood <andyc...@gmail.com> wrote:

Just did a search on Ali to see if there's parts sets. None, but I also notice from the search field pop down that others have done the same search. No doubt the parts suppliers overseas will see the opportunity real soon.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/74009a2c-7d68-45a9-9a49-4aac3a13a36an%40googlegroups.com.

[TobyCWood]

You wrote:" i have an issue with calling it level because its not what your doing"

Actually that is exactly what the HevORT is doing. It's getting the four spots from the bed's corner and then leveling the bed using the 3 ball screws... and theres no chance of binding at the linear guides since the beds have hinged attachments. Unlike the Railcore which I'm told can bind. It's not "Mesh Compensation" like so many other machines do today. 

On Thursday, March 18, 2021 at 8:20:03 AM UTC-7 rlb...@gmail.com wrote:

[TobyCWood]

Actually... this assumes a Duet board. An SKR board can't do this, but the Duet can.

[TobyCWood]

Check out the underside of the bed! Oooooo!

On Friday, March 19, 2021 at 9:35:57 AM UTC-7 TobyCWood wrote:

[Luke Hartfiel]

I think the point being made is that technically it is “tramming”, rather than “levelling”, since it is making the bed perpendicular to the plane of the nozzle, rather than perpendicular to the radius from the Earth’s centre of gravity.

If the entire printer frame was on an angle, then it would tram on an angle so, as you say, it would still run cleanly up and down the rails. 

On 20 Mar 2021, at 4:05 am, TobyCWood <andyc...@gmail.com> wrote:

Check out the underside of the bed! Oooooo!

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/de10a224-1440-43d7-9adb-517c889633b6n%40googlegroups.com.

<HevORTunderside.jpg>

[Luke Hartfiel]

You can also do this on an SKR board.

For that matter you can run Reprap Firmware on an SKR board, and do anything a Duet can do (there is a version for the LCP1769 processor), which technically makes the SKR the most versatile board, since it can run RRF, Marlin, Klipper, etc.

That said I’m still using a Duet in my current build because of build quality - I once went through 3 SKR 1.4 boards in a week due to terrible quality control.

On 20 Mar 2021, at 3:52 am, TobyCWood <andyc...@gmail.com> wrote:

Actually... this assumes a Duet board. An SKR board can't do this, but the Duet can.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/04faee6b-a36f-4a56-ac28-2cfc0e383d35n%40googlegroups.com.

[bkaufm...@gmail.com]

Railcore binding? during 3 point leveling? maybe if you build it wrong. Especially with the kinematic levels points, binding would be highly unusual. 

[TobyCWood]

The Duet with expansion can control more than the usual number of steppers. If one uses a Duet2 with the Duex board or a Duet3 with expansion modules the three steppers in Z can each use their own driver. Most multiple Z steppers are all on one and work (hopefully, It usually depends on the quality on the steppers) in unison. The Duet can do it with the 3 as individuals. The Railcore with the Smart Effector (Although I don't know how it uses it) can do this as well as was shown to me at MRRF2019. However, the Railcore attachment of the bed to the linear rails does not allow for very much off axis movement. IMO this more than others aspects of the HevORT is the biggie. Not that the rest of the design is not awesome. I think it looks amazing and I am looking into building one.

[Luke Hartfiel]

Agree, but that is the same for the SKR Pro, which has 6 stepper drivers slots on board and another 5 on the expansion daughter board. And costs $35 for the basic board. 

Of course you might need to buy a few of them to get one that works… :)

On 20 Mar 2021, at 8:47 am, TobyCWood <andyc...@gmail.com> wrote:

The Duet with expansion can control more than the usual number of steppers. If one uses a Duet2 with the Duex board or a Duet3 with expansion modules the three steppers in Z can each use their own driver. Most multiple Z steppers are all on one and work (hopefully, It usually depends on the quality on the steppers) in unison. The Duet can do it with the 3 as individuals. The Railcore with the Smart Effector (Although I don't know how it uses it) can do this as well as was shown to me at MRRF2019. However, the Railcore attachment of the bed to the linear rails does not allow for very much off axis movement. IMO this more than others aspects of the HevORT is the biggie. Not that the rest of the design is not awesome. I think it looks amazing and I am looking into building one.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/2868674e-0eef-4878-ba27-6719dc94c122n%40googlegroups.com.

[SeaJaye]

I really wish you would build a VORON2! I'm building a v2.4 300mm myself and I'm amazed by this machine every time I work on it. The development team has grown to over 50 people now (I believe) and the design has gone through EXTENSIVE internal and community testing. It really is an incredible machine. 

That being said, I know how skeptical you are about... well, a lot of things in 3D printing that you haven’t tried yet. I’ve been a listener for a few years now, and I remember how dismissive you were of any Mendel-style bed slinger (and still are to a degree, lol), especially Chinese knockoffs. I wanted to buy you an Ender 3 just to prove you wrong! So, you can imagine how happy I was to see you take the leap on your own, and be relatively open-minded about the whole process. </digression>

Well, I personally believe the VORON2 will be an even bigger leap for you. There are several design decisions that you’ll have a hard time accepting. I know you’ll want to make changes and customizations before you even try the base configuration, such as using a Duet controller. This would change your entire experience, IMHO. You should really try using the SKR / Pi / Klipper combo for yourself before deciding it’s not for you. The stock config is capable of independently driving all 7 steppers (1 extruder, 2 for x/y movement, and 4 independent motors for z-axis), and it’s expandable with inexpensive off-the-shelf controllers. The community configs and print profiles reportedly work amazingly well out-of-the-box. There are some folks who’ve gone with the Duet instead, but I’ve seen many switch over to the cheaper boards and use Klipper.

In addition to the controller, you’ll probably question the belted z-axis, 4-point gantry leveling, a 3D printed extruder, etc. You’ll end up spending $1500+ for a 350mm VORON and you won’t be satisfied with the results. 

That’s why I think the VORON v2.4 is not the right choice for you right now. 

In my opinion, you should actually go for the VORON v0. It has a lead screw-driven z-axis, MANUAL bed leveling, and you can build it for $500 or less. The design is simpler and you only get a 120mm^3 build area, but you still get the VORON experience. If you hate it, then at least you’re not out $1500!

I recommend checking out Stefan Hermann’s v0 video. 

One of the VORON developers has his own YouTube channel (Nero 3dp), which Stefan mentions in his video. His videos are an indispensable resource; I wouldn’t have even considered building a VORON without these videos for reference. He has build-sourcing advice, hot end comparisons, Klipper setup and tuning, and even live streams where he does full builds of v0, v2.4, Switchwire, etc.

His most recent video about the new Bondtech LGX extruder is excellent. (You’ll also see how similar the LGX is to the VORON Clockwork extruder design.) 

You should also see his hot end comparison video.

Anyway, just my 2 cents.

[TobyCWood]

First of all... your comment about Mendels. ALL Mendels are a slower print speed. The slower print speed is no big deal if speed is not relevant to your use case. But for many it is. 

We've always liked and recommend the Ender3 for many hobby level USE cases. We've never said anything different and so I really take issue with your comment "how dismissive you were of any Mendel-style bed slinger (and still are to a degree, lol)". I went into our show dB and counted all the episodes we have done on Mendel designed machines. One or two would be dismissive! I counted twenty nine. How is that dismissive?! 

However, when we got the Ender 3 in 2018 it was after years of 3D Printing at 80-120mm/sec. The Ender 3 (as well as the i3) prints optimally at 40mm/sec no matter what upgrade you spend on for it. To switch directions of the entire beds mass requires it to go slow. It is what it is. As such to our respective experience with machines that use industrial sliders and a CoreXY kinematic it was a major performance downgrade. But of course one can still get a great print from it and it is super cheap.

Having lots of people involved does not automatically make something an ideal design. (I learned that the hard way working on Space Station!)  On paper I would not have gone with GT2 belts and four steppers for the Z axis. Way too many places for backlash to happen especially for use cases that require Z hopping... Which BTW is kinda needed if one ever does any form of multi materials printing. Ah! A different USE case!  As we've said, its all about the breadth of USE cases a machine can do optimally. The Voron overall is an impressive design and it does look like a much easier and way more supported build. But the HevORT addresses aspects of 3D Printer design that we have seen issues with over the years. Come on! Ball screws Vs GT2 belts? Really? If I'm investing any of my time and money it will be that way and not the Voron. Sorry if that upsets you.

[TobyCWood]

Wait... I found two more segments we did about Mendels! Ah... but ... no... those two were dismissive!

[SeaJaye]

I really didn’t mean to come across as insulting, Andy, and I apologize.

When I said you were “dismissive” and I “wanted to prove you wrong” I was referring to your PRE-Creality podcasts (2016-2018). I also meant it in a joking way, like a gentle ribbing you would give your buddy. But I reread my post and I now see how that didn’t quite translate well, especially from an internet stranger. That was my bad.

[Luke Hartfiel]

If it helps, I don’t have a podcast, but I am totally dismissive of “bedslingers”.    :)

I think they are obsolete technology, and that the only reason they still exist is because it is relatively easy to ship them, with r to he end user only having to connect the uprights to the base. Shipping an assembled CoreXY, by comparison is basically shipping a large box of air. The technology in a bedslinger also doesn’t scale well, which is why the gold standard for these devices, the Prusa units, are limited to a 250x210x210mm build volume. Just because it is more convenient for the seller, that doesn’t make it better for the buyer.

In every other respect - speed, print quality, potential build volume, breadth of materials that can be printed, ect., the bedslingers fall short of other designs, especially the fully enclosed ones.

Is there a place for them? Sure - for a home hobbiest, who has plenty of time on their hands and doesn’t mind waiting days for their small print to finish, then they may be a cost effective option, but we all know a well tuned CoreXY (including a HevORT, Railcore, Voron, etc.) or Delta (less the flavour of the month, with only modified versions of the Rostock and Kossel kicking around) could churn out the same print, at the same quality anywhere from 3 to 6 times faster.

That said, it has been a long time since I’ve had and bedslingers in my garage.

On 28 Mar 2021, at 9:01 am, SeaJaye <c...@alcorta.org> wrote:

I really didn’t mean to come across as insulting, Andy, and I apologize.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/e3bc53d9-0aec-49f9-b057-1f1c775a8906n%40googlegroups.com.

[TobyCWood]

It's OK... No offense taken. Please do not feel inhibited to post again.

The Voron does look awesome as does the Railcore. Great designs overall especially compared to the now obsolete Replicator designs and as Luke said, bedslingers. Mendles have their place and they have earned my respect not because they are optimal 3D Printers (which they are not IMO) but because they are optimal for the price and simplicity which makes it super easy for anyone to get a 3D Printer. As we've talked about it before, if you can get it into a smaller flat box you'll be able to compete in this crowded market space.

My opinion on the Voron was based not just on the GT2 belts for the Z axis but really due to the GT2 belt loops that transfer the torque from the steppers to the GT2 belt line for each Z axis corner. We've seen over the years across lots of designs that toothed belts and toothed pulleys are hard to stop from having backlash. CoreXY approaches cancel out these backlash potentials or slop in the system. Now if one's use cases only require downward stepping per layer the Voron approach will work fine. However, I for one am no longer only going down in Z on my cartesian machines. I Z hop more than not these days. IME the ball screws in my R3D N2 are granite rock solid and allow lots of fast Z hopping with zero backlash as can be seen in the uniformity in the resulting layer lines. As such if I take the plunge into a next build it's the HevORT.

BTW... there's another reason I really like the HevORT beyond the extreme dynamics of the Z axis and thats the way they've developed a system for calculating the size of the machine. A spreadsheet which you enter your build area requirements and then you get your BOM all sized accordingly!  SLICK! 
Also... I admit I have zero Klipper experience. Perhaps I or Whitney need to get one and get that experience so that we can compare it to the Duet and RRF experience.  Regardless, every aspect of the Duet/RRF has been brilliant for us. I did have one board die on me while under warrantee otherwise I feel really spoiled and really would rather not use Marlin.. 

[Ryan Baird]

Just a quick link to one awesome YouTube channel on HevORT development.

https://youtube.com/c/MirageC

[Daniel Rosen]

Look at the speed of that thing!

[TobyCWood]

uhhh... infill percentage is kinda low! I can do a 20 min 3DBenchy on the CR1, but to get the quality I put the infilll at 10-15% and it's usually a fast Honeycomb pattern.
Still... thats a sweet machine! As I wrote in another thread I also see the same kind of effects at the higher speeds. The plastic wants to just keep on going.

[TobyCWood]

So imagine... you start the print with the bed leveled to the X/Y gantry... then at some predetermined point in Z, you switch the bed to a 45 degree tilt and then print at 45 degrees using gravity to control bridging. ooooo!

[Luke Hartfiel]

Better yet, be able to tilt the hot end, which would do the same thing, but not affect bed adhesion.

You would never need to print with supports ever again…

On 30 Mar 2021, at 4:26 pm, TobyCWood <andyc...@gmail.com> wrote:

So imagine... you start the print with the bed leveled to the X/Y gantry... then at some predetermined point in Z, you switch the bed to a 45 degree tilt and then print at 45 degrees using gravity to control bridging. ooooo!

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/9afba3d1-16d3-40f3-b21a-a47489816099n%40googlegroups.com.

[Ryan Baird]

Oh well if we are going that far let's tilt both the bed and the print head. 

[TobyCWood]

Right now the HevORT's HW can tilt the bed during the print. There's no slicer SW solution yet. You could do an external command in S3D to do the tilt but S3D does not have a 45 degree print capability like the slicers the iFactory uses. 

[LukeH]

I’ve been thinking about this while building my big delta. I think that if the bed was built on a turntable, so the model could be rotated in any direction, and the head could be rotated from 90 degrees to the bed to 45 degrees in any single direction (it doesn’t need to rotate in all directions, since the rotating bed would take care of that), then you could probably print any surface in any direction without supports.

I suspect that would be the lowest complexity 5-axis printing. 

The hardest part of course would be the slicing algorithm.

[Edward Simpson]

Can't remember where the link went, but there was a large delta featured on Hackaday a couple years ago that had a simpson platform as the bed mount. I think it was some enginering final paper, those sometimes get mentioned over there but I recall it could tip the bed about 40 degrees up from level in any random direction, upper half was just a stock Kossel setup.

[Luke Hartfiel]

That would work - you could run the linear rails all the way down and have the bed supported on a second set of carriages. Then you could angle it in any direction using the three points (positions of the carriages on the towers) to describe the plane of the bed.

Once again though, the challenge of 4-axis, 5-axis, or 6-axis 3D printing isn’t the hardware it is the firmware control and the slicer.

On 31 Mar 2021, at 9:46 am, Edward Simpson <r2a...@gmail.com> wrote:

Can't remember where the link went, but there was a large delta featured on Hackaday a couple years ago that had a simpson platform as the bed mount. I think it was some enginering final paper, those sometimes get mentioned over there but I recall it could tip the bed about 40 degrees up from level in any random direction, upper half was just a stock Kossel setup.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/23ba9f0b-faef-497b-8a96-c1a02682c7e8n%40googlegroups.com.