MikeTrice

They look as if they are X52/53 Hornby TRIANG CROSSHEAD ASSY R/H & L/H EARLY PRINCESS

Bet that was hard to solder!

https://www.ssplprints.com/products/pod1036865

As you can see it has been a few years since they were initially produced. Hopefully the following will help:

Hornby are aware of the issue.

It is one reason why I use Romfords as they add weight.

No, it is one of my Shapeways kits.

When the original Rails version was released I was very critical of various aspects of the model. This release is predominently a repaint into the correct lined appearance and the opportunity has been taken to correct the incorrect lettering of the preserved vehicle. There are however two physical changes that have been made to this release (and I apologise to Rapido if there are others) firstly the wheels have been upgraded to better represent a disc wheel and secondly the vehicle does not have the duckets stuck on which makes it possible to run your model for the period 1928 to 1936. All in all a BIG improvement and well done to Rapido for revisiting them. For reference this is what I believe to be the full timeline for the vehicle. 6th March 1906: First test - R Class 2109. Vehicle is in NER Crimson Lake livery which means both the sides and the ends are in Crimson Lake, and the various panels on the side and ends are lined. Running number 3591. 18th Oct 1923: Last test in NER livery - C11 no 878 9th June 1924: Repainted in imitation grained teak and given a slightly modified form of the NER lining to sides only. Relettered and rebranded "LNER" with running number 3591Y. 26th June 1924: First test run in LNER guise - GN Section 4419 3rd April 1928 - 25th Aug 1928: Gresley type bogies fitted and corridor connections added with end grill protector hinge positions swapped. Number changed from 3591Y to 23591. Still imitation teak with lining present. 26th Sept 1936 - 19th April 1937: Duckets fitted. H M Lane's photo would suggest that the ducket protection grills were in place at this point, however they were not visible at Mallard's speed run. 4th July 1938 - 29th May 1946: In storage. Probably renumbered 902502 with grill protectors added to protect the ducket windows before re-entering service. Once again only the number panel appears to have been altered, the rest of the vehicle is still lined. 30th May 1946: Test runs with locos 484 and 1766 listed as practice runs and for staff training. 26th and 27th Feb 1948: Dynamometer Car sent to Derby to be re-calibrated along with the GWR and LMS cars 8th Apr 1948 - 3rd Sept 1948: Locomotive Interchange Trials. These were the first tests to be reported direct to British Rail. On 29th July 1948 the interior temperature of the vehicle was recorded at 93 degrees Fahrenheit. 4 Sept 1948 - 11 Nov 1948: Given full repaint in imitation teak but without lining. Branded as British Railways E902502 12th Oct 1951: Last test - Hunslet 500HP Diesel. Latter half of 1963: Sent to Darlington for preparation for display in Clapham. Likely to have only involved regraining the panels containing lettering. LNER lettering restored. The "Dynamometer Car" lettering is not quite in the same place, it being incorrect on both sides.

Its last use was 12th Oct 1951 according to the log book.

The method I use is very simple. Eileen's GNR sprung w-irons for the outer axles, a MJT internal unit for the centre axle with a piece of lead as a weight and a length of N/S wire. The centre axle just goes along for the ride:

For PLA Deluxe Plastic Magic works well.

The PDFs seem to have gone the same way as photos from early posts. I will try and find them again and reload. Edit: I no longer seem to have them on my hard drive. If anyone downloaded them and still has them perhaps you would be kind enough to reupload them for me.

The certificate has expired which means if you buy anything via credit card your details are not encrypted. Their site mentions that payment is via a secure payment provider so it is probably web standards being ultra careful.

Thanks. I am already using Cura V5.

Hopefully. It is now a question of going through the various settings to find any that look promising.

As an experiment I thought I would give the free version of Prusa Slicer a go. At the moment it does not have a profile for the Vyper but it is not too difficult to adapt an existing one. For first attempts I have used sides and ends from a GNR fruit van produced by @JCL and @chris p bacon. Really please how they have turned out although they do need some additional tweaking. In all cases they have been printed flat on the bed at a 0.1mm layer height using Anycubic PLA with the print lines running horizontally. The last two images have had a light spray of Halfords' primer:

As with resin printing the model needs to be put through a slicer to produce the machines native gcode. In my case I am using the free version of Cura. Compared to resin printing there are a LOT of potential print settings to use. Cura, on slicing a model, can display the paths the filament will be laid in. Taking the example of my Clayton Trailer side: By default here the top layer fill is diagonal which for railway parts may not be the most desirable. Cura does have settings to all the fill pattern to be configured to different angles but I found the easiest option was just to rotate the model 45 degrees in Cura: The settings on the right hand parameters allow theTop/Bottom Pattern to be changed. Here is ZigZag where you can see the pathfolds back on itself when it reaches edges: Concentric results in the following: Finally here is the same side with Lines: I mentioned I carried out lots of tests using the Clayton Trailer end. Here are a few of them. Closer study reveals the different fill patterns used to asses each: I finally decided that using Lines parallel to the side gave the best result. There is a feature in Cura called "Ironing" which adds an extra process to the extrusion and is supposed to give a very good surface. The end with Ironing enabled certainly had a very smooth finish however it rather destroyed the beading: Following a comment by @JCL I tried printing a GNR carriage side at 7mm flat on the bed with the line pattern horizontal and at the minimum layer height of 0.06mm (the Vyper has a 0.4mm nozzle BTW). Stepping is very evident on the lower panels: However taking just a section ofthe side and elevating it 45 degrees a far more acceptable finish results (yes there is still some tweaking to do): Try as much as I can I cannot completely eradicate the surface markings on the Clayton sides and ends other than filing, sanding and priming. I have read that the quality of the nozzle as supplied with the Vyper is not very good so I have ordered some high quality ones from AliExpress and will repeat some of the tests when they arrive.

I suspect it will be very difficult to match the print quality of a resin printer especially in the smaller scales. One thing I can state is I seem to get through far less filament printing things than I did resin printing things.

Time to get serious. On Thingiverse I found someone had posted a model of a Sentinal Railcar and set to printing it: You will note I have some surface blemishes which need to be addressed. Unfortunately having printed out all the parts I discovered they were to the wrong scale and did not really reflect any LNER variant of the Sentinels. After the disappointment with the downloaded Sentinel I decided it might be interesting to produce my own model of the Clayton Railcar Trailers that were supplied to the LNER. My rational being that I could use the ends to experiment with slicer settings and try and improve print quality. I came up with the following parts: The vehicle features recessed doors and I wanted to print the sides flat-on-the-bed so had to produce the doors separately. The sides had strengthening strips added top and bottom over the door cutouts to hold the sides in place. Tabs were added to the corners of the sides to help stop the thin sections warping during printing. Assembly was carried out using Deluxe Plastic Magic. The doors were fixed in place, the droplights fitted and the first side glued to the floor: The second side could then be added: When the solvent had set the top and bottom alignment strips could be razor sawed off: First attempt at the roof with some warping evident in the corners: Based on the first prototype I carried out lots of experiments in an attempt to improve the surface quality. I also took the opportunity to revise the original 3D model to improve it: Experiments took place to try and eradicate the printing lines on the surface. This was acheived using a combination of knife scraping, sanding and spraying with Halfords' Filler Primer: Print quality does depend on print orientation. For the roof I ended up printing it vertically which gave the best print finish. The rainstrips being so small struggle to print well so I decided to take them off and leave grooves to allow strips of styrene to be added:

I should also point out that the Vyper comes with an arm to support the filament roll however I decided to buy a roller carrier to stand the filament on as seen in the first photo above. The now followed a period of playing, mainly printing bits for my Grandsons. A T Rex tooth: Velociraptor claws: T Rex skull: I also printed some filler strips to cover the grooves in the printer bed to stop bits falling into them: Finally, and just for the hell of it I had a go at printing a GNR 6 wheeler. OK, needs work (a lot):

I have recently purchased a new Anycubic Vyper FDM printer: Why you may ask when I already have 2 Anycubic DLP printers? DLP (Digital Light Processing) printers use a technology that exposes uv sensitive resin (in my case) layer by layer building up a 3D object over time. The majority of resins are extremely brittle (as Tony Wright discovered on my V2 bodies) but can be mixed with more flexible (and last time I looked) and more expensive resins. The finished prints have to be cleaned in Methylated Spirits or IPA and generally smell unpleasant. The resins are toxic and gloves need to be worn. Care must be taken when washing prints so that resin does not go down the sink. Putting a machine to bed on finishing can be a messy business and takes time. So to answer the question why, I found using the resin printer too much hassle. An FDM (Fused Deposition Modeling) printer on the other hand prints by laying down "cables" of molten filament a bit like a clay coil pot. I had managed to pick up two reels of SUNLU PLA (Polylactic Acid) Silk filament in a fetching shade of pale green from a charity shop so was able to start playing when the machine arrived. Assembly was very straightforward and I was soon able to print off the supplied test print of an owl: The owl is quite a coarse print so layer lines are very evident. I decided I really should do a proper job of it and calibrate the printer. The first and most important is to check and configure the e-steps. FDM printers feed filament through a heated nozzle at a specific rate. If the software requires 100mm of filament to be fed through the nozzle but the printer only physically feeds 95mm then the filament will be under extruded. Likewise feeding too much filament results in over extrusion. My machine as supplied was under extruding so I had to reconfigure the machine to correct it. In order to measure and configure the e-steps setting you need to connect the printer via the supplied usb cable to a computer and use a control program to remotely control the machine. There are many articles and videos on how to do this so I will not go into any further details. With the e-steps correctly configured I then printed a calibration cude to check the dimensional accuracy of the printer: I then downloaded and printed a temperature tower from thingiverse to test the effect of different nozzle temperature on the PLA (Polyactic Acid) filament I was using: This determined the best temperature for the Anycubic PLA to be 190 degrees. I then decided to print a benchy (well one has to really): The last test which I have only printed today is to test various overhangs to determine at what point overhangs need support: Unfortunately I rescaled the downloaded overhang test so the lettering is unreadable.

4D Model Shop

The cornice (the strip along the side of the roof that covers the join between the canvas and the side) should be teak, not white:

My workflow after adding supports in Chitubox is to save the file and supports as an stl file. I have been known, on occassion, to load the supported stl file into blender, select the original model and delete it then import the revised model and orientate accordingly against the supports then re-export as stl, slice and print.