After a few years of dithering, have ordered a 3D printer, in fact a Snapmaker 2.0 A350, which also can laser cut wood and plastics and CNC mill wood, plastic and aluminium.
The laser and milling functions will cut from common image file formats (JPG, PNG, etc.) as well as CAD files .... just as well since I have no CAD skills.
It will be able to work from images I previously prepared for the Cricut Maker (a heavy duty die cutter).
The 3D printing bit is still a total blank so far as my capabilities are concerned but that will now need to be mastered.
The plan is that the combination of the three functions will produce panelled carriages and horse boxes (principally GNR/GCR/LNER and LNWR/MR/LMS stock).
There are plenty of videos about using the Snapmaker on YouTube but has anybody here used a Snapmaker?
Regards,
John.
John,
Examine the Wiki on Cura, LibreCAD, OpenSCAD and Laser Web. Have a play with some of the SCAD scripts. Hopefully I have made them easy to read and understand.
I hope you enjoy being as lazy and cheapskate as I am!!!
Regards
Ralph
It looks to be an interesting machine John.
I've had a 3D Printer (Sovol S01) for just over a year and have found it very useful for some things (all workshop related thus far). Ralph recommended Open SCAD for simple objects and I would agree with him that SCAD provides a low-effort route to designing for 3DP. I've used it for simple things and also downloaded useful objects from Thingiverse that can be 'adapted'.
For larger/more complex, especially multi-part objects, a good, modern 3D CAD system is required and I've been transitioning from TurboCAD 2D to Solid Edge 2020 since last year - partly prompted by now having the Sovol. I can highly recommend SE 2020 (Community Edition) as being both functional and very powerful. It's fully parametric and uses both ordered and synchronous design methods (hard to simply explain but it does make life easier). It's fully parametric and can generate dimensioned 2D drawings from 3D parts very simply. More importantly perhaps, it's also a free lifetime license (and is not Cloud based like Fusion).
I have thought about buying a small CNC machine but I do have a finite limit on the new things I can absorb within a given period and I've quite enough with 3DP and 3D CAD at the moment.
I will be very interested to hear how your machine handles the CNC side however.
Regards,
IanT
Noticed this comment on the G13D Circle John - and thought of you! :-)
"I wanted something bigger so I bought a Snapmaker A250 3-in-1 machine from China. This prints, laser cuts and etches. So far I have only 'printed' but I am extremely satisfied. I like some of the design features. Calibration has a 9 point probe of the bed and then a single point depth measurement. Changing filaments is very easy with direct access to the head drive. If power is switched of the machine restarts where it left of. Software is called 'Luban' and is based on Cura."
Sounds good.
Regards,
IanT
Ian,
Mine arrived yesterday (from Manchester although it was made in China) and I have only had time to check the contents are all there.
It is very solidly made..... the box (it is the A350, the largest version) I was unable to lift!
Encouraging to hear positive reports and hopefully will find time to assemble over next few days.
Regards,
John.
The contraption is still in its box (when assembled it weighs 29kgs.... 63lbs) so I still need to find a suitable base to stand it on, as well as clearing space in the garage.
The first items I wish to print are domes for the LNER (ex-GC) class A5 4-6-2T.
I have prepared OpenScad drawing (as below) but am uncertain about a few aspects.
1) How will the printer handle the "arc" where it sits on the boiler, to prevent the rest collapsing (will it automatically insert temporary supports)?
2) The extreme edges, where it fits onto the boiler, are very thin..... will they print or do I need to "beef 'em up" (not sure how best to do that) ? At present I have (you can just see in image) included a slight (cylindrical) central underhang to raise the edges from the boiler so Isopon filler can be used (if necessary) to replace the missing lip.
The boiler will be rolled n/s ..... I don't feel adventurous enough (yet) to try and print the boiler and dome as one item.....give it time though!!
John.
John,
One of the options that you will find on the control program be ot Cura or similar - is build support structure. This will produce a printed "froth" to fill in the gaps and produce an armature for the main piece to be built on. You then have to remove the froth with a knife or your fingers.
Note : PLA is the strongest and PETG has the best surface finish.
Regards
Ralph
Thank you, Ralph.
Have found the settings.
The reel of filament supplied with the machine is 1.75mm black PLA.
There are suggestions in the support forum that the Snapmaker PLA is of variable quality and can crumble or break without much encouragement.
Will order some PETG to try out.
Where do you buy yours?
Regards,
John.
Like steel PLA is strong but brittle. PETG is softer but rubbery compared to PLA. Think of them as steel and aluminium. I get both off eBay
Regards
Ralph
There has been some debate about using PLA in models that will be used outside John - mainly because there is concern that if left in direct sunlight there can be warpage. This has centred mainly around the printing of wagon and carriage body sides (e.g. relatively thin parts) and may be less of a concern for something more solid like a dome. Most of the G13D Circle seem to use ABS for these parts, although other filaments are mentioned too.
https://gaugeone3dcircle.groups.io/g/home (https://gaugeone3dcircle.groups.io/g/home)
Thus far, I've only used PLA+ for my various prints but I have recently purchased some PLA+Carbon filament to trial (but this needs a 'hardened' nozzle). ABS is also more difficult to print than PLA - see:
https://3dprintergeeks.com/abs-vs-pla/ (https://3dprintergeeks.com/abs-vs-pla/)
Regards,
IanT
I live "fairly high up" and wind impact damage is more of a problem than solar heated distortion. The "tiles" I print are 9mm thick and are stuck to exterior grade plywood. I don't use dark colours suchnas slate but mimic the local "Bunter" sandstones and millstone grits. The tiles are printed at 20% fill so there must be little in the way of potential distortive energy. The bulk of my printing is for building work rather than Carriage and Wagon work. The last Brassica station was destroyed by a flying fencing panel on an 80mph gale.
Other than that and squirrel knawing I have had no problems with pieces left outdoors.
Regards
Ralph
Having carried out a bit more research (in the light of Ralph's and Ian's comments) the main problems with PLA seems to be that it is not waterproof and is prone to "compost" if subjected to moisture and is prone to warp if it gets to temp. of 60C . It absorbs moisture in storage (hygroscopic) and can "foam" if water content too great (there are heated boxes available to dry out and contain the reel prior to and during processing).
By comparison, PET-G is waterproof (it is the stuff used to make the drinks bottles for mineral water, fizzy drinks, etc.) and is more stable than PLA at higher temperatures (80C+ without significant warping).
The machine came with 1kg. of black PLA which I shall use to experiment but I have ordered 5kgs of PET-G for the loco and rolling stock parts.
How does one make a thin circular object (e.g an inspection manhole on a tank wagon) wrap around and lay flush on a cylinder....which instruction/command(s) can achieve this?
Latest OpenSCAD creation is a GCR/Robinson loco chimney for the A5 (still needs a few tweaks)
Regards,
John.
Matter Antimatter... Consider the band as a cylinder the centre of which you antimatter another cylinder to make a tube. You then antimatter a horizontal cylinder underneath it . Then you antimatter a tube through the vertical tube at the same radius co-ords as the previous antimatter tube.
This will leave a curved part of the original matter tube remaining. OpenScad is very easy but you do have work some parts backwards!!!
Regards
Ralph
Ralph,
Thank you.
I had a suspicion that would be the solution, not having found any commands which would produce a z-axis arched curve in a 2D/flat object.
One feature which would be nice in OpenSCAD (I cannot see that it is there to be switched on) is a co-ordinate tracker for the cursor, to display the x,y,z position at any point (useful in producing the co-ordinates for smooth curves with the polygon command).
Regards,
John.
Can't think of anything like that in OpenScad unless you can import and convert a 2D drawing into 3D scad(?)
Too late in the day....
Regards
Ralph
The instructions reckoned 20 mins. to assemble ..... since there were at least 60 screws involved (plus 25 pages of instructions just to reach the basic assembly stage) I think that is more than a little optimistic.
Anyhow, by the time it had been unpacked and all the parts had been identified, it took me more than 3 hours!
It went together very easily, it is made of aerospace grade aluminium alloy and the machining/precision couldn't be criticised.
It is quite heavy and looks impressive : Scaled up 60+ times, it wouldn't look out of place at Harland and Wolff.
The three photos show the basic setup (ready to receive either the 3D print, laser cutting or CNC milling bases and machine heads); the printer heated plate held in place by 22 screws.... which need to be removed and replaced when switching between 3D/laser/milling.... bit of a bore; the machine set up for 3D printing.
If it works as well as it looks, I shall be happy!
Watch this space.
Regards,
John.
(https://i.ibb.co/DkY9QMc/basic1.jpg) (https://ibb.co/DkY9QMc)
(https://i.ibb.co/G0D6yQn/22screws.jpg) (https://ibb.co/G0D6yQn)
(https://i.ibb.co/18M8k3m/setuptoprintfinal.jpg) (https://ibb.co/18M8k3m)
Welcome to the world of applied laziness! Although my reprap is undergoing servicing I am looking forward to sitting with a pie and a pint while it does the work.
Regards
Ralph
Machine now "run in " as a 3D printer.
4 x A5 (GCR 4-6-2T) tank fillers and a dome .... total print time 12 hours (set at 80 microns and 50 percent infill).
In addition I have printed sound card, whistle relay and Revo module mounting trays, to enable the PCBs to be mounted directly onto insides of brass side tanks and bunkers of a number of locos under construction.
More detailing parts for LNWR Coal Tanks ( cab and tank fittings ,etc.) in preparation.
Not tempted to use laser or CNC milling yet as machine is in the study and fumes/smoke would probably set off the alarms and not be popular with "the boss"!
Regards,
John.
(https://i.ibb.co/jhkm2cZ/A5-fillers.jpg) (https://ibb.co/jhkm2cZ)
(https://i.ibb.co/QcdkfrV/A5-dome.jpg) (https://ibb.co/QcdkfrV)
Well done John, watching your progress with interest, you seem to be winning the CAD battle!
Mike
Having, for a few years, been debating how to best produce LNER panelled coaches of various types, I have experimented with die-cutting but have not got much further than making test pieces (it does work but is very noisy). I have not discounted it but it may be more efficient/effective when combined with other methods e.g. applying 3d printed detailing onto die-cut base.... I have prepared the artwork for GWR 6-wheel Siphons and various horseboxes but have not got so far as cutting the parts.
The arrival of the Snapmaker opens up possibilities for combining die-cutting, 3D printing, laser cutting and CNC milling to produce components more effectively.
My latest test project is to 3D print a Gresley coach module, a pair of luggage compartment doors. It took a full day to prepare the OpenSCAD file and I suspect it may take almost as long to print the subject.
It may not be practical to 3d print individual coaches in their entirety (takes too long) but the printed modules may be used to produce moulds for resin casting.
More news soon.
Regards,
John
(https://i.ibb.co/WGDdxkp/luggage-doors.jpg) (https://ibb.co/WGDdxkp)
Well, I set up the printer and it gave 10hrs 17 mins as time to complete.
It started at 06:05 and has now been running for just over 2 hours.
It now says 10hrs 14 mins to run but more worrying is that less than 10% has been processed..... suggesting a more realistic time of 20+ hours overall. If that is so, it will not complete until tomorrow at around 03:00 BST.
Whoever said watching grass grow or paint dry was tedious hadn't tried 3D printing!
I set resolution at 80 microns .... it will go down to 50 but at that rate I may not live long enough to see it complete!
John
(https://i.ibb.co/Bts8vN7/printtime.jpg) (https://ibb.co/Bts8vN7)
It sounds as if your device is working on the % of the vertical height of the object to be printed, rather than the cubic volume printed. But yes, I have had print times in days... You can alter the print speed aka frame rate up to about 125% without too much degredation. It is possible to start at 100% amd do the last few layers at 150% which gives a really rough finish - like hewn stone. Perfect for building walls with!
My initial 2015 thoughts were that the machine was producing too perfect pieces then I found out how to make them more naturalistic.
Regards
Ralph
In the end it took 10hrs and 4 mins ..... why didn't it just stick with its first assessment which was only 10mins different!
Being black, not easy to photo but best effort below.
There are some minor striations on the lower panel which run in the direction of the teak grain.
I will give it a light "rub down" and decide whether to use as a pattern for a mould or print sufficient for a few carriages.
John.
(https://i.ibb.co/q73wgnT/gresmodule1.jpg) (https://ibb.co/q73wgnT)
I would suggest you SCRAPE the areas flat with a sharp broad chisel as this is how I make the smoothest flatest surfaces for parts.
Regards
Ralph
The Snapmaker has been pounding away almost continuously over the past week.
On the Gresley coach front, there are three printed modules with a fourth currently printing (23hrs 30mins to go) and two more "queueing".
The printed parts will be used as patterns from which to produce the sides for three types of PBV (61ft 6ins gangwayed, 52ft 6ins gangwayed and 51ft 1.5ins non-gangwayed.
Different combinations of parts cast from the six modules can produce any of the above (as well as the van sections of several passenger brake coaches). The Doncaster (GNR) approach to modular coach design is helpful to modellers.
The three modules already printed are shown (arranged in no particular sequence) : Double luggage doors; 3-panel section (for the 51ft 1.5ins and 52ft 6ins vans); guard's door.
Regards,
John.
(https://i.ibb.co/t3PnGhh/3-modules.jpg) (https://ibb.co/t3PnGhh)
John,
Are you going to donate the files to the G3 Wiki?
Regards
Ralph
Ralph,
Whilst I am happy to allow individuals free access to print for personal use, I have concerns that the files could be "misappropriated" for commercial use, since carriages are the commodity most absent from the G3 marketplace.
It may be that the way to deal with this is by illustrating the modules and providing a link for individuals to request access upon an undertaking not to disclose to other parties or use the patterns for profit.
Regards,
John.
P.S. I have already been approached by someone offering a "deal" to secure a set of castings.
John,
Looking at the jpg shows something that is "dead easy" to recreate using OpenScad... It is a simple collection of "translate(x,y,z)" and "cube([x,y,z])" statements. It is of the same complexity as writing a .scad for a five barred gate!
What is really funny is that I a former commercial programmer with a PhD donate .scad files to the G3 WIKI and everyone ignores them. Whilst you write a "doodle" and people go crazy over them...
(Life!!!)
Regards
Ralph
This is the window frame for the Chris Barron 2-NOL loco. The print is 1.5mm thick and I stuck it to the plywood side of the loco.
(https://i.ibb.co/ryq9HXZ/shot-2021-03-30-01-20-39.jpg) (https://ibb.co/ryq9HXZ)
If you buy a reprap for £115, feed it with decent filament £16/Kg and sit down to learn OpenScad then you can have anything you want.
Yes I am that lazy. Yes I am that cheapskate...
Regards
Ralph
Ralph,
I suppose the value lies in the time involved to anyone looking to make money from their designs, .... which you and I are not.
The Gresley modules do include rather more than just cubes and cylinders.
As example, to get the curvature of the tumblehome and beading overlay required other components such as linear_extrude, rotate_extrude and polygon were employed.
It took time and a lot of juggling of the co-ordinates to get it looking just right..... remember this was done only a few days after I started with OpenSCAD. I have not done any serious programming since 1980, when I dragged the law firm I was working with into the 20th Century..... the microcomputer operating system at that time was cp/m!
Regards,
John.
It sounds like the .scad for this is one huge program? Personally I would have done this in three pieces and assembled them into a single module.
To me this breaks down as:
The upper door frame made from cubes.
The lower door frame (tumble home) printed as a section of a cylinder printed vertically.
The frippery which are simple half cylinders.
When I started in 1973 the O/S was UNIX sys III on an HP computer the size of my freezer it had 64k of ring core memory. Nowadays my humble machine has 12Gb and 16 processors.
I agree with you about the profit motivation, but that has never driven me. I do have to design some Greasly articulateds for my EE-1 to pull. So I will probably donate the .scad files for them to the G3 WIKI.
Regards
Ralph
Ralph,
I will email the file for the guard's door for you to see what I have done.
One aspect of OpenSCAD I have not yet had time to investigate is how to (if it is possible) import a file into another. By that I mean, having separate modules (as you mention) and then grouping them in a single file ..... It sounds as though that may be possible from what you have said? A bit like calling a sub-routine into a main program.
Regards,
John.
John,
You should be looking at CURA to collect your modules in a print.
Regards
Ralph