"A Wrinkly in 3D Printing Land", a tale of an OAP Vs 3D printers

23. PLA, ABS, PETG - Part 4

An information update.

Just seen this interesting YouTube video on filament 3D printing materials which may be of interest :-

3D printing material: When to use PLA, PETG, ABS, TPU, Polycarbonate, Nylon etc.



Jim

Back to episode list

 

24 - daVinci 1 Pro Bed Improvement

This was a long overdue improvement, there is a limit to what can be achieved even using high quality painters masking tape, including the blue grade used for spraying fullsize cars.

There was a number of issues in particular :-

• Lack of consistency of adhesion by the first layer of filament to the tape
• Time taken to replace the tape and how often
• Removing delicate parts could also be difficult
• An imprint of the tape and joins appears on the print, a nuisance if that face matters.

The original self adhesive pads were some 5 years old by the time I started using the printer and I suspect would be difficult to apply smooth and subsequently replace, they did not last well in storage.

An alternative covering I found to be succesful was using a flexible magnetically attached covering designed for the Ender 3, easily available and easy to cut down in size to fit the smaller heated bed of the daVinci.

The size I bought was the 235x235mm version available from Amazon UK :-

https://www.amazon.co.uk/gp/product/B0C8HZ2CJH/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1

The self adhesive bottom magnetic layer was stuck to the bed of the daVinci and the removeable top layer would attach to it firmly.

NB before using, remember to adjust the z-offset of the daVinci to allow for the extra thickness of buildplate before you start printing as the autoleveling system will not take it into account.

In fact I have had so much trouble with the autolevelling, like many a YouTuber, I level manually these days, often quicker and more reliable.

Now printing is easier, more reliable and delicate parts may be removed safely by flexing the surface it is printed onto, certainly not possible to flex the aluminium plate.



Jim

Back to episode list

 

25 - External Spool Holder

Advantages and Usefulness

This was originally designed to enable the use of 1kg filament spools with the daVinci 1 Pro but has also proved to be useful with the Raise3D E2.

Although the E2 can hold two 1kg spools internally the feed path from spool into the printing area requires the filament to turn through 90 degrees over the spool edge, which is okay for most filaments but it is bad news for some and causes problems.

Two in particular :-

1) The transparent filament by ERYONE is brittle even when dried and also has a tacky feel, which causes the filament to tangle and knot, the E2 can pull hard enough to lift the spool and snap the filament.

2)The colorFabb red brick Stonefill filament which has a rough texture by nature is also relatively brittle even when dry. This too easily tangles, knots and is snapped by the E2.

In both cases this has caused subsequent issues with printing, the forces involved can shift the z-axis offset minutely and cause 1st layer adhesion failures. This took me a long time to diagnose and easily correct. Many of my prints have minimal surface contact due to their small size so the z-offset is crucial for sound first layer adhesion.

First layer adhesion issues can be caused by many factors, that was one factor I did not expect.

So forceful snapping of filament is very undesirable and must be avoided, so in use I align the feed path from the external spool holder to the entry point into the E2 printing area.

A second effect of tangling/knotting is that when the direct drive feed of the E2 heads is confronted with a tangling filament it will grind its way halfway through the filament. This creates a lot of dust which falls on the hot end, messy, and also on to the print job surface, potentially ruining the finish and/or interlayer adhesion.

High friction when pulling filament off a spool is bad news for a direct drive extruder.

Back to the spool holder.

Design Requirements

Initially for the daVinci it was designed to be sat on a shelf near to the rear of the printer, have a clear or at least translucent top, rollers with ball bearings and a feed path which did not pass through the base or lid but via a separate waistband.

The latter resulted from issues I have had with the filament drier, but that is a tale for another day.

The design was developed and printed in stages to ensure that all would fit as required, PLA from Raise3D Premium (came with the E2) and standard HatchBox was used and most parts printed with 10% infill, and 50% where loading was expected.

Rollers

The rollers are wide enough to take a generous 1kg spool, and spaced so that small 0.5kg spools and full size 1kg spools are easily accommodated.

Using SketchUp Make 2017, my CAD tool of choice the design was built up as an assembly starting with the rollers, aided by a models of spools some of which I have used to print smaller spools, cheaper than buying them and a simple representation of a 1kg spool.

The rollers were made in two halves with a male spigot on one half and female half with a socket into which it is a push fit, a few sets of roller bearings were bought via Amazon to see which were the best width and diameter.

Test pieces were printed to determine the best size hole in a roller end to push fit the bearing , and the best size stationary axle on which they would rotate.

Once finalised the rollers were printed and assembled.

Image 1 shows left to right, a complete roller assembly, two roller halves, and two test pieces for checking the fit of the bearings.




Roller Cradle

Push fit static axles with a hexagonal head to help twist the axles into place were fitted to retain and support the rollers in the cradle.



The roller cradle is made detachable from the casing base for ease of printing, assembly and any further changes required.



These were printed in the daVinci.

Case

Case parts were printed in the E2 as a larger build volume was required.

With the roller cradle and a spool included in the model assembly the base could be designed around them.

An x-ray view of the simplified 1kg spool on the rollers :-



Pegs and holes locate the roller cradle firmly to the base, further holes between them in the base enable the two to be separated with the aid of a screwdriver from below.

Model of the base :-



Feet are moulded plastic retained with brass screws and nuts to the base.

This was followed by the waistband which includes the feed hole by which the filament is conveyed to the printer by a PTFE tube.

Base, cradle and waistband assembled :-



The lid was printed by the E2 with a transparent PLA, the thin silver waist band was part of a test exercise but I left it in the design because it looks good there and stays attached to the top.



The windows in the top are one filament thick and although tests proved the E2 was capable of stacking many single width layers (0.4mm nozzle) the slicer would not include them as part of the inner surface of the lid despite being recessed. The result was that the connection between the single filament 'pane' and the case was only sound on three edges, but it stays in place even if it is not airtight.

Final Assembly

X-ray view of the CAD model.



Assembled unit with a 1kg spool ready to load into a printer.



Top in place.



Result

A further refinement would be a simple tensioning brake to stop undue unwinding of the filament especially during retraction movements, which would help maintain a well coiled filament on the spool.

Overall a pleasing functional add-on tool which works well with both the daVinci and the E2 which has provided a valuable insight on how to use these filament printers to create larger useful objects.

Jim

Back to episode list

 

26 - Fume Extractor Hood

Advantages and Usefulness

When I bought the DaVinci 1 Pro back in 2017 the ability to print with ABS seemed like a good idea for functional parts as that did not seem viable with PLA then, however I know it is possible now especially with a higher performance PLA such as PLA+ from HatchBox.

But ABS parts remain a very resilient option and certainly where elasticity is important.

The big problem with ABS that many 3D printer manufacturers seem to ignore when it comes to domestic users is that the fumes emitted when heating ABS is toxic as well as being an unpleasant odour, it certainly gives me (as a quite healthy wrinkly) a sore throat after an hour or so of printing. Particulate emission is also unhealthy.

Requirements

The Raise3D E2 bought in 2023 supposedly had filtration to print ABS safely, alas that aspect of an E2 is useless due to the position of the fan, all the fumes escape via the unsealed top cover. So much for a low end professional machine!

However as the print volume of the E2 is larger than the DaVinci I decided to use spare parts for the E2 to make an extractor hood for the DaVinci.

At least the filter should be the right type even though their fan is rather overpriced to say the least, anyway that was the starting point.

The DaVinci has a large vent on the top onto which I could sit an extractor hood, and vent through a filter into the room rather than just pumping unfilitered fumes out of the window which is not an environmentally friendly thing to do.

So with the printer vent, fan and filter dimensions known the design could proceed.

Design

First stage was to build the housing to hold the filter and fan, the filter would slide into the housing as it does in the E2 for inspection and/or replacement and the fan would be attached to the outside by machine screws into threaded inserts.

It was decided that this fan/filter housing would couple to the hood by sliding vertically into place and be retained by the hood top.

Test pieces were printed to check the dimensions of the hood and housing for a light friction fit before committing to the housing itself.

Printed on its side, closed end of filter slot on the build plate would minimise the need for supports.





Then confirming that the fan/filter housing printed as required dimensionally the hood base part was designed. This proved particularly tricky due to the compound angles of the sloping lid, I have still plenty to learn about compound angles in SketchUp.



The base was then printed in the E2 due to its size, the next two images show the CAD model of fan/filter housing attached to the hood base.





Once checked that dimensions were as expected and the fan/filter housing fitted as required the lid was printed complete with the holes for the retaining screws. I had to resort to using pan-head screws as trying to create countersink holes would be too demanding on my SketchUp skills and probably too fine for printing with a 0.4mm nozzle anyway.



In practice I found that the opening in the fan/filter housing was a little too large allowing leakage so a ring labeled "fan inlet ring" was printed to make a better seal between the fan/filter housing and the filter.



An x-ray view of the extractor hood assembly :-



Seals

A thin flexible strip was sandwiched between the hood top and base, foam draught extruder strip was fitted to its base to give a better seal between hood assembly the and printer top.

Screws

Self tapping pan head screws are used to secure the lid to the base of the hood, the holes in the base being pilot drilled to suit.

The fan is retained to the fan/filter housing with steel cheese head screws which engage in cold fitted brass threaded inserts (ref 26.1).

Fan Speed

As fitted to the E2 it is speed controlled although it only seems to run slowly so initially I bought a simple PWM controller temporarily connected, to test the fan at various speeds.

Testing with prints of an hour or so proved that the extraction hood was working well when 12V was applied to the fan at 100% pulse width, so as the full 12V was required (for no horrible pong) I replaced the PWM controller with an old mains adapter with a regulated 12VDC output.

The temporary connection was replaced with a small junction box in which I wired a conversion from a standard tubular power plug to square pin posts to plug into the connector of the fan.

Connecting in this manner means if I need a replacement fan in the E2 in a hurry I can easily move it from the extractor hood as the original cable and connector on the fan is unchanged.



Continued in part 2

 

26 - Fume Extractor Hood - Part 2

The Printed Result

Fan, filter with junction box.




Filter removal.




Underside view showing fan inlet ring and base seal, by this time I found that the feet of the fan/filter housing needed to be higher so extra height was gained by printing and gluing square saucers over the original feet.






Finally in place on the DaVinci printer.




Summary

This has proved to be another successful accessory which works effectively with the DaVinci printer, it is relatively quiet, the printer makes a lot of tuneless electronic sounds which almost drowns it out at times. Particulate and fumes are staining the inlet side of the filter which is good news, less for me to breath in.

No noticeable reduction of print quality due to the fan drawing air through the printers enclosure.

Modifying the lid of the E2 to accept this extractor hood is on the cards, however some printing I have done using HatchBox PLA Pro+ would indicate I can get just as strong prints as I can with ABS but without the pong.

Also an advantage of using ABS was that it was easier to cement parts together than PLA but I have found that DeLuxe Materials Plastic Magic (at present the older version, no longer available, but current is supposed to be effective) creates strong joints too. I prefer this to using cyanoacrylate any day.

References

26.1 Threaded Inserts - Brass - Part 2

https://platform1mrc.com/p1mrc/inde...le-of-an-oap-vs-3d-printers.6395/#post-105317

Jim


Back to episode list

 

27 - Filament Dryer Upgrade

The FixDry filament dryer I described last year (Ref 27.1) has proven to be useful for static drying but rather awkward to use when feeding the filament to a printer whether heated or not.

It is provided with multiple exit points around the case to which a tube can be inserted for guiding the filament to the printer.



However seating the spool on the rollers whilst the filament passes through the lid is very tricky, a third hand would be very useful, one each to hold the lid, the spool and guide the filament through the tube.

The solution to this problem was to introduce a waistband layer between the lid and base of the filament dryer through which the filament would pass. Design was a little tricky as the joint between lid and base has radiused corners and getting the curves to the correct radii and allowing for print shrinkage proved challenging.



My first attempt was almost right but the top inside radius where the lid sat on the waistband was of too large a radius, i.e. the curve cut the corner, so the lid would not sit loosely in place, in fact it was just too tight to fit at all.

So as the rest of the dimensions were good and the print sound I decided to just cut off these corners as shown.





The rubber grommet into which the feed tube fits does so by virtue of a recess on the inside of the waistband to locally reduce the waistband wall thickness.




End result is much more practical as I do not need to grow a third hand.



References

27.1 Filament Dryer

https://platform1mrc.com/p1mrc/inde...an-oap-vs-3d-printers.6297/page-2#post-103824

Jim

Back to episode list

 

Thanks, just as well you don't actually hear me speak, I'm awful!

Jim

 

28 - Raise3D E2 Bed Improvement - may also be useful for other printers

Most of the parts I print are quite small as would have been seen in my Arkitex project.

The build plate of the Raise3D E2 is a plastic coated flexible steel plate, magnetically held to the print bed by strong magnets.

An object is printed onto this plate and when finished and the object has cooled the plate is removed from the bed, it is then flexed and the printed object pops off.

Well, this works fine if the object has a stiff solid base, but what if it just bends with the plate.

Bend the plate too far and it is creased and then useless, an expensive part to replace.

Many of my Arkitex and other model parts just flexed with the plate, scraping off even with the supplied scraper risks damage to the plastic surface of the plate, replaceable but still costs and sticking a new one on flat and squarely tricky I would imagine, I have enough trouble with a smart phone screen protector.

Or, the printed object is damaged.

The Alternative

For just over a year now I have tried out a :-

AJOYIB 3 Pcs Flexible Magnetic Heat bed 3D Printer build Sheet Heated Bed Cover 310x320 mm Compatible with CR-X, CR-10 pro V2, Ender 3 max, CR-10 V3

which I bought on Amazon :-

https://www.amazon.co.uk/gp/product/B09VBMSDLC/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

This is a soft flexible magnetic plate which comes as a kit of two magnetic print surfaces and one base layer which is stuck to a printers non-magnetic bed. Available in different sizes, so choose the smallest you can cut down if yours is not listed.

Only the top surface parts are required for this printer and are placed on top of the original flexible steel plate, there appears to be little issue of magnetic polarity, they just stick.

It can be readily cut with scissors into two parts with little wastage as shown below, the guide lines printed on the surface aid cutting.



I positioned the joint to the area where it will least be used, but in practice it hardly shows on the base of an object, and although I rely on the build plate surface for the outside face of many of my parts, they are rarely big enough to reach that far.

It can be used, if necessary with bed adhesives such as Magigoo and Geeetech JT16.

Cleaning

Typically when Magigoo remains are to be removed I just rub it lightly with a soft sponge wetted in warm water and washing up liquid. Rinse and then dry off with a clean cloth, I use a microfibre cloth. Refit in the printer and heat the bed to 50C for a few minutes, this is quite adequate to fully dry the surface.

Geeetech JT16 is also water soluble.

In Use

When used on the Raise3D E2 the level sensing system seems to accept it without issue, so no changes have been necessary in Z-height calibration when switching between using the additional layer or not.

Just make sure that the whole build plate is covered or not.

Summary

This machine can be used for printing small delicate parts with this 3rd party covering, I bought an extra kit of this product but have yet to wear out the first piece despite the large amount of prints I have done for the Arkitex project.

I have used it for PLA, ABS, PETG and recently TPU successfully.

Jim

Back to episode list

 

29 - Bowden Tube Release Tool

This is very much a universal tool, designed for releasing tubes from the industry standard pneumatic tube couplings adopted by 3D printer manufacturers for connecting PTFE and similar tubes to machines and accessories.

Here is a simple but useful tool I made for these couplings, aimed at those for 1.75mm filament, it could probably be scaled up for larger couplings, presuming that larger ones are required for 2.85mm filament.

I was finding it increasingly awkward releasing these connectors on both of my machines, no issue for agile fingers and young eyesight but becoming increasingly problematic for me. I tried using pliers or spanners but inevitably they would slip off with an accompanying cuss.

So in the end I came up with this solution, no doubt someone has already created this but here is my version for which I included dimensioned drawings so you can print you own if you wish.

Almost any rigid material will do, I used basic PLA and printed off a few, because like paint tin openers if you only have one you can never find it!

General View




Plan view




Side view




In Use

I generally use it as in the side view, but at times using it the other way up is better, depends on the access available.

Just slide it sideways into position, press down and the extract the tube with the other hand.

Easy.

Connector on my Raise3D extruder




Below in position to release a tube




On my modified DaVinci 1 Pro




Jim

Back to episode list

 

30 - Raise3D E2 - 0.2mm nozzles - PLA - Nylon

(Part 2 of DCC Loco Conversions - 53 - Mainline 43xx Mogul 1978)

This is in three parts.

1. Setting up the printer for printing with 0.2mm nozzle
2. Printing the parts to repair a Mainline 43xx Mogul
3. Summary

The information does overlap partially with the DCC conversion article but keeps the two articles self contained.

1. Setting up the printer for printing with 0.2mm nozzle

After completing the design and printing of the belts for the TTR Elevator Conveyor I switched the nozzles from 0.4 mm to 0.2 mm to find out how detailed I could print using these finer nozzles.

Now Raise3D sell 0.2 mm nozzles and provide support in the E2 firmware for them but that is as much support as you get, yes that is essential, but nothing else was available.

So when the nozzles were installed they could be configured for use singly and together using the firmware of the E2 to set their Z height and relative offsets, fine, but printing profiles*?.

*Profiles : Those blocks of data used by their slicer, IdeaMaker to match the characteristics of a particular filament to the characteristics of a machine.

Profiles for using 02 mm nozzles were not available for the E2 at the time of writing this introduction to using them on the E2.

Looking around there was two options I found in their community forums almost as passing snippets.

PLA or similar: Copy an existing high quality PLA filament profile, modify the appropriate parameters, e.g. nozzle diameter and layer thicknesses to be proportionally smaller for a 0.2mm nozzle to use as a standard quality profile.

Nylon: either as above or use a community supplied profile for another Raise3D printer e.g. a Pro 2 and modify the necessary parameters and the G-code sections of the profile.

At this stage it became necessary to keep the profiles for 0.4 (I'll skip the mm bit from here on) separate from 0.2 profiles.

The way I did this was to create a custom printer in IdeaMaker named similarly to the default E2 machine but with a reference to 0.2 attached to it and apply the necessary default perimeters which related to a standard E2 mechanical spec, e.g. print volume dimensions etc.

Then to create new profiles and add them to this new machine, save the profile I wanted from the default E2 printer as the starting point to local disc, swap to the new custom printer and import and rename from local disc so this copy was now part of the profiles only for the 0.2 E2.

Yes, it is a bit involved, hopefully I have remembered it correctly!

Profile Options

For option 1: this was relatively straightforward for PLA (as used for my improved EPC Coupling Assembly), however for nylon there was two options, first, base the profile on PLA and then apply the data from Polymaker to suit their CoPA.

Polymaker Site : https://polymaker.com/product/polymide-copa/

Or option 2: Modify a Polymaker CoPA profile for printing CoPA on a Raise3D Pro 2 printer.
From the community forums it was possible to download this directly into IdeaMaker (I am currently using v5.1.4) with the 0.2 E2 printer selected.

Downloaded as Copa_BVOH_LQ

From https://www.ideamaker.io/detail/1457-pro2-series

NB - you must change the G-Code in this Pro 2 profile to that supplied for the E2, as it was I forgot but it was flagged up as an error when I tried to perform a print and the machine was stopped by its firmware because there was a collision likely, I am not sure where but the G-Code are different, but, anyway the printer had to be re-started. After which I cancelled the job.​

Having got that tee-shirt I copied and pasted the G-code sections from one supplied for a 0.4 E2 and the print proceeded okay.

Nozzle temperature can be reduced as the lower flow rate through the nozzle means it is heated for longer, this is also a minefield and can easily lead to jams or stringing.

There is not a lot of information available on the web or YouTube about printing with nozzles below the default 0.4 size, generally 0.4 and larger, printing small functional parts with a filament printer does not seem to be very popular.

Who cares, I will find a way!

2. Printing the parts to repair a Mainline 43xx Mogul

Whilst trying to convert to DCC running a Mainline 43xx Mogul I found that this loco had Mazak rot which had destroyed the two idler stub axles between the motor and the geared driving axle.

The parts required were :-



Rather than just jump straight into 0.2 printing with CoPA I decided to first develop the parts I needed and print with HatchBox PLA Pro+ a tough version of their PLA which was already in the machine from the EPC Coupling work, a bright yellow, light/bright colours are best for prototyping parts, easier to examine and mark up when an iterative process is required.

So after a few iterations I had workable parts in PLA which proved it was possible to print them with sufficient definition to be viable, shown fitted.



However these are not strong enough for real use, this I confirmed in a few minutes on my rolling road, when the upper stub axle sheared off. It is also essential to print with 100% infill because even a 1.8mm axle can be printed hollow with a 0.2 nozzle.

Printing with CoPA

After drying I mounted the spool of CoPA in a Kingroon box with desiccant which I found worked very well when printing TPU belts over a few weeks without needing to further dry the filament. Here in the UK the humidity tended to be around 50% RH late last autumn when I was working with TPU.

As the TPU did not need a heated bed I printed with the top of the E2 open and fed the TPU directly over the machine to the extruder as shown below.





Nylon filament does not like draughts and requires a heated bed, even the extruder fan to be off, so this time the lid was down and I fed the CoPA from the Kingroon box via PTFE tube to the extruder by removing one of the flexible rectangular bungs separating the internal spool containers of the printer from the printing space.

Adjusting the parameters for a string free print (after drying the CoPA for 8 hours at 70C, which really isn't hot enough, I need a higher temperature dryer) I got a decent print but somewhere whilst adjusting the multitude of parameters it suddenly went stringy and I could not recover from it so I dried it again with no improvement.

The axles would end up badly distorted, all was a mess, along with extruder jams.

Had the CoPA become hopelessly damp, meaning that the drying I performed which was below the recommended 80C did nothing and now it was in too bad a state to use?

At this point it felt like a bad parameter mix had occurred so I decided to swap to a profile via option 2 as I could not at this time dry it at 80C.

This printed well enough straight away to produce the functional parts I required, the CAD dimensions were almost perfectly replicated in the printed part, hooray!

Annealing

Nylon parts just printed are not as tough and temperature resistant as they can be so require annealing.

From the Polymaker site.

"PolyMide™ CoPA displays outstanding mechanical and thermal properties when fully crystallized. The printed part will not reach full crystallization after the printing process, an additional step is required: Annealing.

You can anneal PolyMide™ CoPA by placing your part in the oven at 80_C for 6h."

That 80C again, will these parts be strong enough without annealing, a key issue when filament printing is layer adhesion, if poor, a stub axle which is a tower of cylinders is probably more likely to shear than break the CoPA itself.

Enhanced heat resistance is not required in this application.

Moisture absorption after printing

It seems that nylon is always 'thirsty' and in some cases prints are given a soaking to improve elasticity after printing and then annealed. I did leave my prints a couple of days to re-absorb some moisture before using them.

Boy is this material complex fun!

3. Summary

My first foray into using 0.2mm nozzles and CoPA has been a headache with the messy prints and the extruder jams I experienced when adjusting option 1 as it slid into chaos for reasons unknown.

However I have proved that I can print small functional parts in a modest engineering grade material that work, for how long, well, time will tell.

So far the CoPA printed parts have survived about 30 minutes of test running on my layout pulling a couple of coaches (typical train length for this small loco) and at least 5 on the 1 in 40 incline from a standing start. Promising, but I am not deliberately going for destructive testing as dismantling this loco is complicated and tedious. If the stub axles break I will fit annealed parts, otherwise I will just use it.

I must get a suitable dryer that can also be used for annealing, I have looked at some high temperature (90C) dehydrators but think that the new SUNLU FilaDryer E2 is worth a shot. Then I can develop printing with nylon filament and even possibly stronger engineering grade materials like carbon fibre filaments for tools.

Oh dear my wallet says, a printer for carbon fibre?

Nothing ventured, nothing gained.

Jim

To Part 3 of DCC Loco Conversions - 53 - Mainline 43xx Mogul 1978

Back to episode list

 

31 - Tierwag End Ramp

One of my Tri-Ang Tierwag car transporters, bought second hand that was in good order except for a missing ramp, needed attention so I decided to dismantle it to aid design of vehicle wheel chocks (Ref 31.1) and in passing create a replacement ramp prior to reassembly.

Dismantling

Remove the central lifting deck section by lifting it half way up such a finger may be inserted via the lifting section to go under the static upper deck next to it.

Then gently ease the static deck part upwards whilst springing the body sides with your other hand so it can be un-clipped from the main body of the wagon, easier done than said.

Once one static deck is removed the lifting deck part may be easily unhooked from the static deck at the other end and lifted clear.

The other static deck can then be removed easily.

Both static decks were removed, one to measure up the remaining ramp and the other ready to insert the new ramp.

Design

Straightforward but required orientation to enhance appearance as supports were required which can be quite unsightly when printing small parts with a 0.2mm nozzle on a Raise3D E2 filament type 3D printer.

The ramp is basically flat but, has angled surfaces, which surface needed to look the best as a supported print surface will not look very good. In practice this was answered by which faces are seen the most, noting that most of the time the ramp is stowed for running.

The only deviation from the original design was replacing the two small lugs which support the stowed ramp with an entire surface, this way the exposed when stowed surface does not suffer from support 'scarring' because the model can be set to place this complete face on the bed.

CAD

The design




Orientation notes.




Result

Printed in Hatchbox PLA Pro+, Cool Grey which is quite close to the original colour, closer in fact than some of the age discoloured wagons and their ramps which tend towards a green.

The parts.




New ramp in place, ensuring the guide pins are in the vertical guides.




Static deck re-fitted, the fine steps on the top of the ramp are not unreasonable, as I suspect they would have been ribbed for grip on the full size wagon.




Outer face, relatively smooth on the underside of the ramp.




To complete, re-fit the lifting deck, then the original ramp and finally its static deck.

Summary

A simple print which looks much better than the piece of Plasticine that was there before and operates as an original ramp, a few spares were printed while convenient.

Jim

References

31.1 Minix - Tierwag - Upper Deck

https://platform1mrc.com/p1mrc/inde...-tierwag-and-cartic-2.7084/page-2#post-115731

Back to episode list

 

32 - Supports for Delicate Parts when Printing with a 0.2mm Nozzle

Part 1 - Design and Print

This arose from the need to provide effective supports for small parts without using soluble filament which was impractical with a 0.2mm nozzle on a Raise3D E2 printer, even the support materials are specified for 0.4mm nozzles and larger.

The models being printed were wheel chock frames for the lower decks of Tri-ang Tierwag and Cartic 2 car transporters, (refs 32.1 and 32.2)

The Solution

After quite a bit of experimentation and frustration the following approach has proved to be effective.

First remove the missing layer to stop printing in the air which is the default for use with a 0.4mm nozzle, printing in the air does not work when the first layer of the model includes a corner, the extruded plastic just cuts the corner, which is messy when the model part is only a few layers thick. The result is malformed and weak.

Second, minimise the contact between the top layer of the support and the first layer of the model by other means, this can be achieved with a minimal infill density of 10% from layer 2 of the support upwards, layer one needs to be solid for good bed adhesion.

I use the rectilinear infill pattern. Ensure that the angle of laying the plastic is always alternating, I use +/- 45 degrees, sometimes the default settings angle the last support layer and the first model layer the same way. This will guarantee problems as some finer pitch model layers will partially drop between the sparse support top layer depositions, a mess can ensue again. When at right angles to each other short bridges are created which is much more solid.

The horizontal gap between the support and a vertical model face is kept minimal, just enough to insert a scalpel blade, e.g. 0.2mm.

CAD

Example part requiring supports between the supporting chocks.




Preview of print, layer 1, high density first layer of the support structure is shown centrally.




Preview of print, layer 9, last of the main support structure only, central, model layers surround it.




Preview of print, layer 10, end cross members start to be printed




Preview of print, layer 11, longitudinal beam starts to be printed




Preview of print, completed




Result

Spot the deliberate mistake, the resulting print shown below is an earlier version than the CAD drawing and slicer previews but, the technique is the same, (update 06/1/2026).

During printing somewhere between layers 2 and 9




A pair of parts as printed.




Continued in part 2 - Post Processing

Jim

References

32.1 Minix - Tierwag - Lower Deck

https://platform1mrc.com/p1mrc/inde...-tierwag-and-cartic-2.7084/page-2#post-115782

32.2 Minix - Cartic 2 - Lower Deck

Link to be added here when the posting is made shortly.


Back to episode list

 

Part 2 - Post Processing

Now comes the tricky part of removing the supporting material without destroying the part we want.

The support should now have a dense skin where it was first laid on the bed followed by alternating angle of 10% infill creating a lightweight mesh to which the required part is printed directly onto.

Removal of the support can be performed relatively easily with a sharpish scalpel; a really sharp one would have to be used carefully to avoid damage to the print or your fingers.

First image shows two identical parts just removed from the bed, left showing dense base, right 10% infill rectilinear for remainder of the support.




Prising up the centre of the skin with a small flat bladed screwdriver until it breaks or cut midway with fine end cutters to reveal the interior of the support.




Peeling back and then using a scalpel to trim it down to the model.






If required used a 240-grit abrasive paper to smooth further.




Test fitted into the lower part of the lifting deck to ensure it grips okay, and observing that there is not any support residue preventing the chocks sitting on the deck.




Further Refinement

As I moved onto designing similar chocks for use with the Tri-ang/Hornby Cartic 2 I found it better to include the last layer of the support as part of the model and slice the support off by cutting between the penultimate support layers once the hard skin was removed. As this face is not seen it is not worth smoothing it down, besides, only a fine pattern is left behind.

In addition, to save breaking off the centre posts a small piece of wood was drilled with a hole that the post could freely enter to make working on the underside of the frame easier by allowing the frame to be pressed against the wood. This also reduced the frequency of stabbing my fingertips with the point of the scalpel.

The best sequence of support removing depends on the part and the arrangement of the support and your dexterity, when many parts are processed an optimum process develops.

A variation of this technique was applied to my 3D printing of a 4mm scale skip truck platform, which is somewhat larger than these chocks but, still a relatively small print, Ref 32.3.

Summary

This solved an issue I have had with using supports on delicate parts when printing with a 0.2mm nozzle, I have not read about this approach anywhere else, but then very little is published about this scale of filament printing anyway. I suspect most just keep quiet about it.

Jim


References

32.3 1960s Skip Truck

Link to be added here when the posting is made.

Back to episode list