rholgateau

Some good suggestions here, guys - thanks! I've tried Bill's "additional supports", as shown above. Unfortunately, that didn't seem to help... ... just off the printer, with the outer supports removed. You can see the substantial supports bottom left, yet the creasing still occurs. I'll try it again with extra support at the bottom of the cab side but in front of the opening...? Resin holes: yes, I had thought of that. The bottom of the coal bunker is open, and I have a couple of holes between the cab and the coal bunker Z lube: I haven't done this recently, so will add some grease and see if that impacts this. I haven't (ever) lubricated the sliders - they have always appeared quite dry and I had assumed that no lube was required? Light off delay: I understand that a "delay after light off to enable the resin to cool/set before lifting" is a good idea, but Anycubic doesn't support this (at least via Chitubox). The Light Off Delay setting merely specifies a minimum time for the light to be off before turning on again - Anycubic says that it's to allow resin with poor fluidity to reflow"? I confirmed this just now by setting this param to 10s and running a "dry print". It was clear that an extra delay was inserted after the raise/lower of the platform, before the light came on. So, as my raise/lower time is perhaps 5s, this param won't have any effect unless it is more than this. But it is pretty cold here in my shed in Sydney, so I should crank up my resin heater to be sure. Further question: I am using the side "skirts" to support the model: I have seen this recommended under rolling stock to obtain a smooth contact along the edge of a straight-sided item such as this. I have reduced my contact thickness as much as possible, to make the removal easier - it is currently around 0.13mm - is this too small to provide a firm support? So, I will now (a) try even more supports, (b) grease the z-axis, (c) heat the resin, (d) set a cautious Light Off Delay, and try again Fingers crossed!

Bill Thanks for replying so quickly. I had seen the misaligned corner, but am surprised that it might impact up the whole footplate? But, hey, that's why I'm asking for help... I do have a fair bit of support in that bottom corner, but it's on the underside of the skirts that surround the loco. Maybe these skirts are detaching from the loco - or being stretched a bit? I've just put 5 heavy supports right on the loco body around that corner, and I'll run the print again overnight. If that sorts out the creasing, then the cause is confirmed, and I'll have to work out how to provide such additional support on a more cosmetic manner. Back in 24 hours... Rob

I am trying to draw up and print the body for a Johnson 0-4-4T loco in N, to fit on the Dapol M7 chassis as is. (Having to make a few dimensional compromises to enable it to fit on the M7, so purists should look no further.) CAD model is pretty well complete, for 58084 and 58042: The printing is looking pretty good as well, except for one major problem – the model is “creased” on one side – and I cannot work out how to resolve this. Printer is an Anycubic M3 Plus. Resin is Monocure Rapid. Supported and sliced in Chitubox. Model currently tilted at X+20, Y-40, Z+40 degrees. Supports are auto-supported at Density=65%, with finer supports manually tweaked on fine details. Plus extra Heavy supports on the skirts all around the model. Settings are: with the printer UV power set to 60%. Layer height is 0.050mm, but would like to reduce that when the models are looking good enough. General supports look like with the problem area also supported: Results are promising – the pipe-thingy in front of the side tanks is drawn at 0.15mm, yet prints well. The problem is the horizontal creasing on the LHS of the model (well, horizontal when being printed, diagonal on the resulting model). This is on the “downhill” side of the model; the other side is almost perfect. I’ve run numerous tests and cannot get rid of this. Does anyone know what the issue might be (and how it might be addressed)? I’ve been thinking of: · Insufficient supports in that area (I’m trying hard not to have any supports on the outside flat areas of the model) · Wrong exposure settings (but have rerun calibration models) · Wrong/dud resin (and it’s over 12 months old – but Monocure says that it should be fine) · Overly optimistically fine model (the cab sides are approx. 1 mm thick – and other parts of the model are thinner) · Wrong tilt angle (and I’ve tried a lot of different variations) · Or something else I've been printing quite a few other models, but mainly old buildings or tree skeletons, where exact precision is not critical. This is my first item of rolling stock, and my skills are clearly lacking! Thanks Rob

In regards to Simon's query... Some plans (although they didn't always build according to the plan...): and a couple of pictures: (I don't recall where I got all these from, so apologies if they are not appropriate to post here...) Rob (also building MD - in N, in Australia)

Hmmm - further usefule suggestions there, guys? Tebee: you are suggesting that maybe it slumps - presumably, after the printing has finished? But I'm a bit surprised that it happens along such a clear line... I've tried some further controlled tests with details removed in the CAD file, so that it is just a "prism" Vertical height of object adjusted between runs solely by extruding the base (i.e. roof angle is constant) Settings are: Photon S, latest firmware, Chitubox v1.8.1, 4x80s base layers, 0.020layers at 6s exposure, list 6mm @60mm/min, Bottom tolerance compensation -0.7mm. Exposure is based on a Photonsters/RERF print with this resin, showing 5s is barely adhering to the base layers, 6s is best for engraved detail, and 11s best for raised detail. (I have roughly sprayed grey primer over the clear resin, to make the results more visible) Apart from that, the only variation between runs is the resin: I topped up between some runs, and didn't on others. Could the resin be settling in the VAT, with different print characteristics at the bottom if it hasn't been stirred?? I have tried three comparative runs: Print 1 and Print 2 are identical, except that #2 have three identical units. I was looking to see if the "point of inflection was the same on both runs, which might suggest a mechanical problem at a particular vertical height. Unfortunately, Print 2 showed the symptom occurring at a different height Print 2 and 3 are identical, except that #3 has the left unit 1mm taller, and the central unit 1mm lower. Print 3 is consistent with #2, in that the feature occurs at (approx) the same height on the right-hand unit as on #2 (for the same height unit), and is around the same absolute height on the shorter and the taller unit, i.e. it is not a constant offset from the ridgeline. This suggests to me that is is not a slumpting characteristic as that would distort to the same shape as the ridge section is identical for all three. So I'm thinking: Bill's suggestion of the lead screw: I'm happy to grease it, but how does one clean it? Do I have to dismantle the unit for this? Is the slicing working as expected? Mike Trice has offered to look at the slicing that I'm getting, and I'll take him up on this... Is the transparency of the clear resin causing problems, as a couple of you have suggested? I'll try some more normal resin - and I'll stir it between runs. And, as a further complication, here's another test piece, hot off the printer (and primer). This one is the same sold shape, but stood up on its back end: Sorry about the grit in the primer(!), but it clearly shows some variation in the lower layers, around the area that has been causing concern?? Apologies for the long post - and thanks for all your help! Rob

Bill, Yes, the (strange) effect is as if the layer heights had increased near the ridge line. But I haven't knowingly done this - and I didn't know that it could be done on a Photon S. I'm running some further tests : one ridge rather than 3 - is the overall area of the print causing some excess load on the Z axis? changing the overall height, to see if the effect is at a constant Z height, or a constant height relative to the ridge. Maybe I have a binding on the vertical drive at that point? and anything else I can think of. I'll report back, in case anyone is interested... Rob

Thanks, guys, for the responses. But I'm struggling, trying to work this out... First of all, I accept that models will, in general, print better when inclined: less sudden change in cross-sectional area being printed, no problems with Elephant's foot, etc. I am reluctant to go down this path, as I find it hard to get rid of all the support marks - and a couple of attempts at this have just ripped off the supports! In fact, I think my best result might be to print it vertically set on the baseplate, and cram on several of these sections in the one print. But ... my concern is that I don't understand what is going on here?? Mike (1) talks about "the suction effect" and "stretching/sagging". My understanding of the suction word refers to a hollow model without any drainage holes at the top - the "upside down cup" effect. But this is a single solid lump - and I thought this printer should be able to print such a model of the base plate without noticeable sagging or distortion? Can we not print a model where the printing area covers a large proportion of the baseplate? I seem to be missing something here? I had also printed a hollow version of this model - just the top surfaces, about 1-2mm thick. I set it right on the baseplate again, with only a few supports inside. It did print - and without the distortion visible that I described above. But, strangely, it also cured most of the resin *inside* the roof space! My guess is that (especially with the clear resin) the UV light on any particular level also tended to set the liquid resin above that level, i.e. inside the model. My exposure was (IIRC) 8 secs, and could have been reduced slightly. BTW I have been trying the new ChituBox feature to reduce the Elephant's Foot effect - quite impressed! Rob

I've been drawing up, and Shapeway-ing, several buildings for my N gauge layout, with some success. However, the increase in Shapeways' pricing, and my purchase of a Photon S a year ago has led me to try to print something myself. I'm drawing up the platform buildings for my station, but need to work out how I will be building the platform canopy - looking something like I guess, especially in a larger scale, one would design a clever etch for the glazing frames, bend it all into shape, and glaze with clear styrene at the end. But I reckon this is almost impossible in N, certainly within the limits of my skills. So, I had another idea... I'm happy that no-one would ever be looking at the underside of the canopy, so I drew this up... ... to be printed on my Photon, in clear resin. The framework would be drawn on afterwards (hence the fine grooves for the glazing bars), and the whole thing dusted with dirt/matt spray, so that it looked translucent yet the depth of the plastic was not apparent. So far, so good. But printing this on my Photon S (flat on the print bed) produced This sort of looks OK, but there are some (repeatable) distortions near the peaks of the canopy. You can see these more clearly in ... I'm lost as to how to remove these. They are not in the CAD .stl file, and I can't see what might cause them. I haven't seen anything similar on other models that I've printed. I tried printing this on an angle and they disappeared (but the support marks on the model were pretty horrible). Does anyone have any suggestions, please? (Modelled in an old Sketchup Make as a solid shape, sliced with latest Chitubox at 0.020 with a 6s exposure, using Monocure Clear)

Hi guys I've just seen this thread and was fascinated to see your interpretation of this iconic location. Very impressed - and also, very depressed, as you have achieved in a couple of years what has taken me some 15 or so years for my Millers Dale layout! Mine is in N gauge, so I have a bit more freedom to include (for example), most of the station trackwork, and more of the buildings in the MD village. But, to go from Chee Tor to Headstone tunnel, I have had to cut out a couple of the quarries, Litton and Cressbrook mills, and Monsal Head station - sad, but something had to go. Scenic work is now well underway (thanks, Covid!), but there's a lot more required on the station area - platform 1 is done, but I'm still trying to get more info on the other two platforms... Anyway, keep it up, and keep posting here, please? Rob

A couple of further points... You mention the slight grade separation of the two viaducts? I also thought that it was hardly worth it - I've ignored the grade on the line, thinking that it would be invisible to an observer, with the scale of the rugged terrain. But one further thought has occurred to me: not only are the viaducts separated, but (I think) that Platforms 2 and 3 are slightly separated at the south end. There's certainly a dividing partition between the two halves of the platform there, and a couple of photos also indicate the separation of heights - for example, p 143 of Through Limestone Hills. Could be an interesting feature, esp as the platforms are at the same level at the north end? Re the subway - it was obviously the official way to cross between platforms, but I reckon that most passengers were likely to have walked over the tracks at the south end. But, in the 1980s, would imagine that would be frowned on? Guess that you'd need the subway, or even a footbridge?? I agree that the Platform 2/3 buildings are interesting and unusual (as far as one can see of them!). They were originally the main station buildings, but became the island platform after 1905 - hence the lack of windows on the Platform 2 side. TLH p140 has one of the few photos that I have found - others just have very distant or partial shots. And, yes please - would love to get a copy of the balustrade CAD file for the viaducts!! (how best to send the - do you want an email address?)

I haven't been on to RMweb for over a year and, as others have said "wow - this thread has blown my socks off". But there's a more particular reason in my case. I am also modelling Millers Dale in N (nearly 10 years so far), and it's fascinating to see the design, the approach, the techniques that another has taken. Cav - have no fear that your efforts are being challenged. I'm modelling 1951 (i.e. mostly steam), and am quite some distance away (Sydney Australia!). And my abilities and standards are nothing like yours that are demonstrated here! I have somewhat more space - it's an L-shape, approx 14' by 9'. So I've included (on the Manchester side) Bridge 75 leading to the Chee Tor tunnel. And on the London end, I have the two MD viaducts, then I've skipped Litton and Cressbrook mills, but I have then included Headstone viaduct, leading into Headstone tunnel. (I'm frequently unimpressed by layouts where the river disappears into the backscene - it's very hard to make this at all convincing, IMHO. So, by including the downstream viaduct, the river will curve forward and exit at the front, if you know what I mean.) I will have only a token building at the front below the station (the ex-pub cottages), but I am including the village on the other side of the railway - the Anglers Rest, St Annes church, the mill on the river, and a couple of cottages. To fit this in to an L-shape, I have had to build the MD viaducts on a curve - you can imagine how this affects the dimensions of the 15 curved spans!! My approach was to draw these up in CAD and have a "skeleton" for each span laser-cut - and then manually built up into girders, etc. I also had the detailed 1905 strengthening overlay of the inner spans cut as an addition. I built the bulk of the baseboards in a similar manner to yours - plywood girders, and multi-level structures on the baseboards where the track is elevated. But the bridges are only now being filled in. The Headstone viaduct and the new MD viaduct were done a couple of years ago, the new MD viaduct was largely done earlier this year (I'm envious of your handrails!), and am now about to start painting Bridge 75. Landforms are largely in place (and river now poured, including floating weed), road in place, but vegetation (lots of trees!) not yet done. Photo of the MD viaducts loosely in place (without railing) is attached. Just FYI - please, no comparisons to Cav's efforts!! A comment on a couple of earlier points: I don't have detailed pictures of the buildings on platforms 2/3, and haven't started on them yet. I'm attaching the best plan that I have found (two halves plus poor quality, I'm afraid - original is in the Derbyshire Records Office in Matlock). Otherwise it will have to be the best guess from the various pictures that people have posted and various other similar ones that I have found The "large rhubarb" is a type of Gunnera and is large enough to be distinctive, even in 1:148. I'm planning (eventually) to draw up a sheet for etching - unless anyone else has one available? Will keep monitoring this thread!! Regards Rob