D869

First work out the geometry of the flare on a dome. That seems somewhat tricky on its own. I struggled to do it with a file and some brass, never mind telling software how to do it. At the risk of diverging well off the OP's question... I suspect it's do-able - at any point when viewed in cross section it's probably not unlike a circular fillet but the radius is small on top of the boiler and a good deal larger at its lowest point. Not all flares are created equal - some seem to flare so that they end up almost tangential to the boiler cladding, others hardly seem to bother much and have a fairly noticeable angle where they join. So it seems possible to generate circles of varying radii and position them around the base of the dome in a manner not unlike my tank filler skirt rivets. Joining them up as 3d objects might be trickier... generating lots of spheres and subtracting them from a 'fat' dome might be easier.

+1 for OpenSCAD (but you knew that already) If you are happy getting your hands dirty with code then it's ideal - most of the tank barrel and rivets below were done with code - like spacing the rivets equally around the barrel and doing the 3d trig required for the rivets on the filler skirt. The top of the filler is a separate bit BTW. It's not all code - the vent and syphon bosses are rotated DXFs as are the domed ends on the barrel. Horses for courses. I also use a lot of 'if' statements to allow me to do an assembly preview model pulling together DXFs from etch artwork with the 3d printed bits... and then use the same OpenSCAD model to render the parts (on sprues) for sending to the printers. I suspect something similar is possible with Blender, Fusion 360 etc but once you know one tool there is less benefit in figuring out how to do the same things in another. I do use a bit of Blender too - for posing views of rendered STLs and setting up lighting and camera angles to replicate prototype photos (and overlay the two for comparison). I find this a useful way to check dimensions when drawings are unavailable or don't fully answer the question. I have to refer back to my notes to remember how to work Blender every time though - I find it a pretty strange beast but its 'sweet spot' is more artistic 3d rendering, perhaps of organic shapes and also animation so perhaps it's no surprise that it's a bit odd to those of us more used to CAD tools.

As a rule it is wise to have a sheet of something under the job. Ply, aluminium, MDF, Tufnol etc. This can be bolted or clamped to the bed without the clamps interfering with the cutting. Double sided tape is very useful for holding the sheet being cut onto this. For metal I use the Diall stuff and use a hair dryer to get it unstuck afterwards. Maybe something less grippy for plastic but I have not done much of that.

I think the best advice is not to worry about the number of flutes and just get on with doing some milling. One day you might need to care but for now you really don't. Whether a cutter can plunge does matter, of course and it's worth buying ones that can do this. Did you also know there are single flute cutters?.... and reasons why they work better for some jobs... but again, to get started with an MF70 it really doesn't matter. Bigger diameter mills snap far less frequently, so use the 2mm and 3mm cutters while you are learning. Make sure you have some spares in hand when you start a job that needs 1mm. Cutter strength is related to cross section, so a 2mm is 4 times the strength of a 1mm and a 3mm is... well, figure it out All of 'em will snap if you try hard enough though. I'd reiterate Rich's comment about safety glasses. You can't tell when a carbide cutter is about to let go. I've had one ping off just by switching on the mill - it was nowhere near cutting any metal but it had done plenty of taxing work before that. My theory is that they have a 'stress lifespan' and when you've put them through enough stress cycles they will let go regardless of what they are doing at the time. They are also bl**dy sharp, as, quite often, is the swarf that they produce.

The X and Y axes are finished and have done plenty of good work cutting out bits for the O2. @justin1985 the reason I made my own motor mounts was to allow the MF70 to be swapped quickly between manual mode and CNC. It works as intended and I can swap in about 5 minutes. For quick, simple milling jobs I prefer manual mode. Happy to share the details if you are interested. You may need another machine to make the bits though The Z axis is a lower priority and is not done yet. I did do a mockup using an offcut of laminate flooring and bits robbed from the other axes to find out whether the NEMA 17 motors had sufficient grunt for the Z. I was not able to get the motor to miss any steps during my tests - both moving the axis quickly and slowly, short distances and long. GRBL on the Uno is highly optimised and pretty much beyond any further development - the Uno is full so you need to move off the Uno is you want backlash compensation. As Rich said, I used an Arduino Mega 2560. I had a CNC shield and found that I needed to switch from GRBL to Marlin firmware in order to keep the I/O pin assignments required by the CNC shield. Marlin is intended for 3d printers and is a little odd for CNC but it does the job. Alternatively there are GRBL versions for the Mega with backlash compensation but I'm not sure what hardware can be used to carry the stepper drivers if you go down this route.

Likewise - mine does the job I need. There is a pretty big jump in price and size when you look at alternatives so I have no plans to change it. Incremental improvements, however, are well worth a look.

Not weird at all. Same person doing the asking My standard spindle has about 2.5 thou runout measured on a cutter shank in the (Proxxon) collet. There are other reports online of circa 3 thou runout on the Proxxon spindle. The problem with this is several-fold. Firstly you get a cut that is oversize (3 thou is a fair percentage for a 1mm cutter). This error does manifest itself if using CNC to cut out shapes (and can be compensated by fudging the cutter diameter when generating g code). Secondly there is a high probability that you will only be cutting on one flute instead of the whole cutter. It also stresses the cutter more which is not a good thing with carbide. They do break and if you have significant runout then that will make them break sooner. There was a long discussion started by a chap on the Model Engineer mag forum a few months back. He ended up bolting a whole new Chinese motorised spindle to his MF70 which is a bit extreme (and not cheap) but it worked for him. Having measured sub 1 thou runout inside the collet holder taper, I tried a tip from another contributor to the same thread - to make a new collet with a much tighter fit in the parallel part of the holder. This needs some accurate lathe work and it took me two attempts to make one satisfactory 3mm collet. My runout came down from 2.5 thou to a tad under 1.5 thou. I was hoping to get under 1 thou but it's still an improvement... plus of course making a new collet only improves matters for the collet size that you make. Another comment on the same thread was to the effect that the Proxxon cutters appear optimised for cutting non ferrous metals and also stick out a long way from the Proxxon collets (which doesn't help with either runout or stress). I've now bought a 1mm stub length cutter from Arc Euro trade which should stick out less far and has a more pronounced flute spiral. I can't comment on its performance yet because I have not needed to do any 1mm cutting since buying it. Regards, Andy

I'd suggest starting with a rewheeled 'N' loco and saving the full loco kit build until later. Then you will have something up and running quickly with which you can test your track and electrics.

Could you do the black circles with a fine marker pen with the wheels turning in a lathe/drill/loco ?

Good stuff Pix. Glad you are keeping up the work rate. I especially like the DMU nose - always good to see how people tackle tricky design issues and those cab window surrounds coinciding with the angles on the nose definitely come into that category.

I reckon if you shift over to a different L&B then you will keep the RTR folks at bay for a while longer https://www.youtube.com/watch?v=ZjBKQlr9fSs The donkey at 1:12 is *NOT* happy! I do like a tin turtle too Oh... don't mind me... the randomness is quite an accurate reflection of the contents of my real workbench anyway.

Thanks Tim and thanks for the postings about your loco builds which have definitely helped me with the O2. I do like a bit of L&B and am particularly glad that a short stretch has reopened and we can ride in restored coaches and admire the Exmoor scenery. I like my own narrow gauge models in 16mm scale with actual steam though

You have the 16T kit from the Association - plastic body and etched chassis. Personally I'm a fan of this kit and have built several of them in different variations (top doors or not, bottom doors or not, different brake designs etc.). I generally scrape the insides of the sides to taper them a bit so that the top edges look thinner and add some 5 thou plastikard fillets on tops of the corners to complete the corner strengtheners. The slots in the sides are a nice touch. I'm not 100% sure about their later condition but I suspect you should probably delete the tiebars between the axleboxes. The chassis that comes with the Association kit covers several options including vac fitted china clay opens, hence the tiebars and 4 brake shoes when you probably only need two for a 16 tonner... but again check what the later survivors looked like because I'm more au fait with their earlier condition.

Gosh... five! That's a vote of confidence. I hopefully have one on the way with completely wrong bogie centres which will need some stretching. Ian Morgan's post on t'other thread reminded me of another thing... I wonder if we will get to pay extra for import taxes and then service charges on top to for the privilege of having said taxes collected from us. I've had that happen before now with an order from the US but it seems pretty random.

From https://en.wikipedia.org/wiki/British_Rail_Class_121 - 8'6 bogies and 46'6 pivot centres I notice they have some Green Max in stock too. Historically this brand had better answers for 1st gen DMUs. The product numbers seem now to be in the 57xx range and most of the listings dont quote the bogie wheelbase but maybe there is hope for a 17mm or 17.5mm wheelbase bogie one day? ... or maybe they have moved on to making GWR Pannier tanks now?

MinerChris needed to install a replacement Tramfabriek motor in his 122 to get it to run properly. They seem to do quite a trade in them so maybe factor two new motors into your sums

Lovely job Jim. Good to see it is now in revenue earning service. Onwards to the next project now then

Stephen Harris' kits are excellent. As you say they are 2mm scale. That is a 3% size difference from British 'N'. Unless you have an identical vehicle to 2 1/16 mm scale alongside then you are unlikely to notice... and with wagons I suspect you'd be pushed to tell even if you did have two identical ones. The Association shop also does a number of BR era wagons... it just depends how 'modern' is your idea of 'modern'. http://2mm.org.uk/products/shops.php?shop_num=2 You will also need some 2-027 wheels from the shop to go under your kits - these are 'N' profile but have the right axle length for the kits.

Brakes... I knew this was going to be quite an involved job and sure enough it was... and so is this post (sorry!) The O2 brakes are pretty prominent and even the cranks by which they are driven are on full display. MinerChris does have one of Mr Higgs' chassis etches in his gloat box so I could have pinched the brake etches from that but that seemed rather a waste. The Dave Holland keeper plate design complicates matters because the brakes must be removeable if the wheels are also to be removeable. I spent a whole evening just thinking about how to make all of this fit together. The eventual design is pretty similar to the one recommended in Pete Wright's book but I did consider two alternatives before arriving at this. The key 'lightbulb moment' was the idea to solder short lengths of 0.5 x 0.3mm tube into the brake hanger pivot holes in the frames. This gives just over 1mm of length for location and support rather than relying on just the 0.5mm thickness of the frames. Spoiler alert... the finished article Before starting on the real thing I made a fairly basic test piece to find out if it was actually possible to solder this thing together and also to see if the hangers could be poked up behind the coupling rods when assembing the chassis. This was a very useful exercise, apart from proving that I could put the thing together it also showed me that I needed to make the hangers longer by 0.5mm to avoid the pull rods fouling the coupling rod bosses at bottom dead centre. The pull rods are 0.4mm MSE point rodding strip. Height-wise this is pretty much spot on with the GA drawing but no doubt the real things were not square in profile. They are also nice and strong. A couple of pieces were soldered with some temporary plain rail spacers at the correct spacing. The cross pieces are 36SWG phosphor bronze bent into an 'L' and soldered behind the pull rods. For the hangers I made a drilling jig using the MF70 to get the hole spacing correct and then drilled some 1mm wide scrap etch and filed it to the general outline. 36SWG wire was soldered in for the top pivots. Having taken a break from CNC for the hangers I decided that manual filing was overrated and went back to CNC for the other bits. Because of the tiny parts involved the blank was soldered to a backing plate rather than relying on sticky tape. One of them still came unstuck though (lack of solder coverage). An O2 brakes kit... Soldering the shoes onto the hangers involved a bit of trial and error and some swapping things around on the final assembly to find the best combination. The assembly was then fitted to the chassis and the hangers soldered to fix them to the cross shafts. Some aluminium hairgrips were used as heatsinks to prevent the cross shafts coming unstuck from the pull rods. Some small bits of 0.4mm PCB were gapped and used to join the cross pieces in the middle. The temporary spacers could then be removed. Which just left the connections to the cranks to sort out. On the real thing these have an adjuster so I wanted to represent that. I used some bits of 5 thou N/S strip that I bought very cheaply many years ago and am now running out of. This was bent around a drill and 0.4mm brass wire soldered into the 'U' for the 'pin' into the crank. In fact this is just soldered to the face of the crank rather than being pinned. Once cut to length the adjuster will quite happily grip the pull rod, can be adjusted to line up with the crank and will stay put while being attacked with a hot soldering iron. Nice . The photo was taken before soldering. I got a bit silly with the bosses at the ends of the cross pieces. These are 0.8mm brass rod with a the end filed down to leave just a 0.2mm thick sliver to sit on top of the pull rod. A 0.5mm-ish disc (plus the 'sliver') was parted off in the watchmaker's lathe and the resulting bit soldered onto the pull rod (heat sink deployed again). I think I must have made at least 8 and failed to file the 'sliver' several more times. Once attached to the pull rods, the bosses were filed back to about 5 thou thickness. Once all was assembled the loco was taken for a test run through South Yard's 'interesting' radius pointwork. It made it through but there was some 'rock and roll' due to a clearance issue. Can you spot it?

I'm not so sure. My (early) Hymek notes say that the wormwheel is 5.8mm OD and has 20T. This doesnt work out to a whole mod number but maybe that is the 'fudge' to allow it to pivot. The spur train in the bogie works out to mod 0.3 from my measurements - the wormwheel moulding is a compound gear giving reduction between the wormwheel and the spur train.

I'm not sure how you would get the quartering 'in sync' with 3 axle gears. if your axles fit the bearings well and the crankpins fit the rods well then having a gear on just one axle is the usual approach. RTR models with fully geared wheels usually have tons of play around the crankpins which may be an option... but once you open the holes out you can't close them up again so I'd treat this as a last resort. I've usually had to tweak the quartering a little after using the quartering tool. It's still worth using because it presses the wheels into the muffs squarely but don't assume that the quartering will be perfect.

I've never had much success in getting wheels out of muffs non destructively. Most folks think it's better to trash the muff and avoid any risk of damaging the wheel. Sounds like you have a methodical approach. We're probably repeating ourselves but check for machining burrs on the gears (and file them off if you find any) and also make sure that the gears are not catching on anything at some point in their rotation - both of these things will mess up your running in potentially intermittent ways.

There are probably two questions getting mixed up here - one is how to motorise an MU... which may involve cannibalising an RTR thing or creating something to make life easier, perhaps 3d printed as discussed or something etched along the lines of things that have gone before such as Mike Bryant's offering or TPM's kit that cannibalised a US Bachman A/B unit to get two for the price of one(ish). The other question is how to build a custom diesel chassis where the reason for doing so may be the absence of something suitable to cannibalise or a feeling that the RTR option doesn't run well enough or won't last long enough to put underneath something that we may be investing a lot of time in. I've got as far as doing a CAD drawing of the drive train in one case but on that occasion figuring out how to hold the (ball race) worm bearings and arrange a bogie pivot with something that I could make myself was as far as I got. I agree that brass gears will be noisy but so far the plastic ones that I've found cost more than the brass ones from the Association. I never minded the sound of metal geared Farish or Minitrix drive trains in the past and so far I have never had a split gear with those... but quiet and plastic is probably a better option nowadays if it can be achieved without being too costly.

I was having another look at the gear question. My notes of past research say that Farish locos have 0.3 mod gears... which ties in with what Julia said too so if you want to have the option of using the Association drop-in wheels then 0.3 mod seems like the thing to have. My notes also say 0.3 mod for my Dapol Hymek which is one of the early ones so things may have changed. I haven't found anything that beats the prices from the 2mm shop. If I understand correctly the gears are now being done in house. Whether the in house person has the capacity if we all start building diesel bogie spur gear trains with 2mmSA gears is another question.

Indeed. I've been down that route. An MP15DC is a good match wheelbase-wise for a class 22 and the Atlas chassis runs very well. Sideframe detail is a challenge because it looks very non-British and the sideframe is needed to hold the bearings/pickups in place. I've approached this on both the class 22 and the Kato DMU chassis by butchering the overseas chassis detail. Finding a suitable wheel diameter can also be an issue with US chassis and most 6 wheel US bogies have non-equal wheel spacing to fit around the traction motors. I even invested in a US book of diesel loco drawings so that I could find the key dimensions when I need them. I have thought about how I might build a 'see through' class 22 bogie with spoked wheels and without a chunky gear train to spoil the view... and then thought I'd go and do something less difficult instead.