Richard Hall

Been dipping into Ernie Brack's magnificent photo archive which has plenty of Wansbeck Valley stuff. I'm turning up all kinds of loveliness, including D5177 (Derby Sulzer Type 2) on a troop train, which is rather handy as I have this very locomotive, and with 2FS wheelsets as well! Woodburn seems to have been a final resting place for elderly vans, possibly used as sack / sheet stores: photos from the early 1960s show two ex North British outside framed box vans, a big four wheeled tranship van still in LNER livery and most remarkable of all, an NER bogie road van in BR grey. Farish produced that one in N gauge a long time ago: I never thought I would find a legitimate use for it. I tried using Templot to overlay some trackwork onto the 25 inch OS map, but the results were not good, partly because the OS map doesn't pay as much attention as it should to track geometry. And partly because the curve through the station will need tightening a fair bit to fit a 7x7 space. With "Stobs" I had enough room to overlay the track plan onto the map with no compression but I can't do that here. So I think I will start again, draw a 3'3" radius circle of track and go from there. I'll have to fudge the scenic break at the eastern end, a typical NBR cattle bridge won't look out of place. Richard

That is very good advice and very much what I tried to do with "Stobs". I think I have seen just about every published photo of the Waverley Route and some unpublished ones as well. One of the reasons I have drifted away from the idea of modelling Billingboro is that the Bourne & Sleaford did not attract photographers or writers: information about the line has turned out to be very hard to find., especially on the operational side. West Woodburn on the other hand I already know a fair bit about. I have a shelf in danger of collapsing under the weight of Borders railway books. The area has changed very little since the line finally closed in 1966 and most of the structures and trackbed are still there. There is still plenty of information to discover: I didn't know until last night that there was a shed housing a Wickham trolley just beyond the road bridge, with a short siding leading to it. It isn't marked on the 25 inch OS plan that I was using to for the track layout so I suspect it was added post-war. There comes a point where you have to accept that you will never be able to model everything that takes your fancy. Better perhaps to just stick to doing one thing and try to do it well. My distant ancestors were Border Reivers from upper Redesdale and would have known West Woodburn, if only as a good place to steal cattle from. Perhaps that is why the railways of the area have always fascinated me. Richard

This keeps happening to me. I find a nice little station to model, get as far as posting my intentions on the Internet, then start having second thoughts. Last weekend I was up North and had a couple of hours free so I walked along part of the trackbed of the Rothbury branch. My interest in the railways of this area goes right back to 1977 and seeing Ian Futers' "Longwitton" at an exhibition, after reading about it in the Railway Modeller. I had forgotten just how much I love this part of the world. A couple more nudges to derail my plans: Sovereign Colliery Junction in this month's RM, and seeing Ballyconnell Road (3mm Irish and gorgeous) at the Spalding show. Now I'm pondering circular layouts and wondering whether West Woodburn (on the Morpeth - Reedsmouth "Wannie Line") would make a good subject, seeing as the entire station and goods yard curves through ninety degrees. Main disadvantage is that like my currently stored and unfinished Waverley Route layout "Stobs", a circular model won't fit in the small house where I live. With "Stobs" I made the mistake of building the baseboards so big and heavy that the thing needs a crew of roadies to transport and erect it, which killed my idea of working on sections at home and then taking them to my workshop for running sessions. Time to start looking seriously at lightweight board construction. A Templot session beckons... Richard

Finicky little things, these 2mm mechanisms. I just repositioned the motor on my Pannier chassis, corrected the height on the compensating beam, and now it runs like a three-legged donkey and I can't see why. Bob, I feel your pain. Couple of pages back you mentioned problems fitting crankpins to wheels. I ran into the same problem: my nice Mitutoyo digital caliper measured the crankpins at 0.53mm diameter... Association Yearbook tells me the answer is a No. 75 drill bit but being a crude bodger and not having a set of numbered drill bits I just put each crankpin in a pin vise and filed the inner end down a bit. One day I'll actually remember I have a lathe which does stuff like this. Good luck with chassis number two: this is much too lovely a body not to have an equally good chassis under it. Richard

A six coupled chassis built on Sharman principles has some of the dynamics of a Reliant Robin, which is not a good thing I feel. In the same way that a three-wheeler handled better with a couple of bags of cement in the boot, I suspect part of the trick here is to get the C of G back towards the fixed (driven) axle. The Sharman system is delightfully simple but has its quirks as I am now finding out. The twin beam system is intriguing though trickier to build. Richard

The torque from my 50 pence Chinese can motor is ridiculous, I wish I had bought twenty of them rather than just ten. I think part of the problem here is that the Farish Pannier body carries most of its weight high up. I also managed to mount the motor slightly lopsided so that it catches on the body and stops it sitting entirely square on the chassis, which combined with my slightly off-centre compensating beam is making the whole thing a little unstable. All fixable, and it will be interesting to see how I get on with applying the principles to a nice new chassis rather than a much-bodged one. Mike Sharman's ideas continue to provoke debate and experimentation: I suspect he would have been delighted. Couple of random thoughts: firstly, wondering if I can get a very small vertical slide for my very small lathe. Then I can mill the frame bushes into slots. I tried doing this with a file but the results were horrid. Secondly, the compensating beam might benefit from a couple of guide slots to keep it dead central in the frames. Or even just one slot at the forward end. Richard

Not quite sure whether moving the pivot point was worth the effort. I think my first compensating beam was a fluke: much harder to get it central on the axles and the ride height correct, second time around. I'm now getting an oscillation at certain speeds which rapidly increases until the loco derails. So it still needs a bit of fettling, but here are a couple of bad videos for now. First up, the naked chassis, completely unweighted. If the low speed control looks a bit unimpressive, bear in mind that the controller is an H&M Clipper. https://www.youtube.com/watch?v=z9q4UrAFsJE And secondly the same chassis with Pannier body tackles a ridiculous gradient. 1 in 5, maybe a bit steeper. https://www.youtube.com/watch?v=r9QM0VtImxM Still not 100% sure all this will end up leading anywhere useful, but it's fun. Richard

Thanks for that Bob. Yes, it does have split rods. I'm with you in thinking most compensated chassis have far too much articulation, you probably only need 0.1mm above and below the centre line to get most of the benefits, unless your tracklaying is especially wonky. The problem for me at least is building the thing square enough to achieve that tiny amount of axle travel equalised over four wheels. Having said which, your suggestion is basically using the rods to restrict travel and I can see that might work. One other thought I have had is to move the pivot point for the compensating beam, so as to increase the proportion of total weight borne by the centre wheelset. I haven't seen that done but it will be easy enough to see if it works.

Things are starting to get interesting. I fitted vertical scraper pickups to the leading wheelset which work beautifully, and tried a simple adhesion test, gradually raising one end of my test plank until the Pannier slipped to a halt. This little engine will tackle a 1 in 4 gradient! The benefits of distributing weight evenly across all the driving wheels seem fairly plain. The only problem I have now if a tendency to lift each centre driver alternately under heavy loads. Because each wheel is only bearing one-sixth of the weight, the drive from the rear axle through the rear coupling rods can be enough to push one wheel upwards clear of the track. That little 8 x 15 flat can motor has a lot more poke than I expected. Not a problem in 4mm where (all other things being equal) you have eight times the weight on each wheel. Adding more weight to the body would help, but this chassis was only intended as a rough prototype for the compensated "03" that I actually want to build, which will be much lighter. Incidentally the Pannier runs sweetly and without hesitation as a bare chassis, in which form it weighs about the same as a sweet wrapper, but the wheel lifting is very evident at anything above walking pace. I tried fitting lightly sprung wires bearing downwards on the centre wheelset but it seemed to run worse. Richard

Quick update on the compensation experiment. Now fitted with a shiny new set of Mk 5 wheels (thanks Jon for excellent service as always) and with an old Farish Pannier body dumped on top for a bit of weight, it runs very smoothly indeed. It steps neatly over a bit of 0.5mm wire laid over one rail, and all is rosy in my 2mm garden except for one problem - poor conductivity between the axles and the U-shaped guides that take the place of axle bearings. I was expecting problems here. I can't use conventional Simpson springs (apart from the driven axle, where I have already fitted them) as they would interfere with the vertical movement of the axles. I think I'll be looking at some kind of scraper pickup, as fine and light as possible, most likely mounted vertically to bear on the back of the flanges. I foresee much fiddling with 0.3mm phosphor bronze wire: fortunately I have lots of this. Whether all this effort is worthwhile, I know not at this stage. It will be interesting to see what kind of load the Pannier will handle, compared to my other two 2mm locos. Right now I'm just pleased the thing runs at all. Richard

Zero progress on the layout itself, but I did manage a field visit today while passing through the area. Station is now used as a transport yard and is remarkably intact although the trackbed either side has been obliterated by new houses. That yard office is gorgeous with its bargeboards and finials. I might contact the owners and see if they will let me have a prowl round with a camera. Richard

Bob, I am following this build carefully and learning a huge amount from it. Even little tricks like the plastic sleeves over the crankpins to hold the rods on for initial testing. Thank you for posting in so much detail, absolute gold here. Were the 05s scrapped early because there weren't enough drivers thin enough to squeeze through the cab doors? They can't be much more than a foot wide. Richard

Thank you for confirming that I wasn't just imagining that the crank throws were out. I should have worked it out when one wheelset didn't fit snugly into the quartering press. I'll stick the errant wheels in a box and wait for the day when I have a decent mill/drill machine. Never throw anything away. Richard

Last post on my experimental chassis for the moment. It now has the smoothest, sweetest motor and geartrain that anyone could wish for, but despite endless fiddling I couldn't get it to run without binding up and lifting the centre axle, even after using Association quartering plates. After much measuring I think I have found the problem: one of the wheels has a crank throw 0.2mm less than the other five. I've measured the throws several times in half a dozen ways and always come back to the same result: left hand rear wheel is different to the others. Probably a bit late to send them back (I think I bought them in 2002), so I'll order up some Mk 5s and see if that cures the problem. Richard

Rescue operation under way. Things weren't quite as bad as I thought, it started to free up quite quickly and after half an hour's running there is only one small sticky spot left. If this doesn't quite clear it I have a small pot of jewellers rouge at work doing nothing useful so I might try that. Motor is a Chinese 8 x 15 flat can, first one I have used from this batch and it seems very nice, quiet and plenty of torque for its size. Richard

Rather inevitably it has all gone wrong, but not for the reasons I expected. As mentioned above I had to relocate the first stage gear assembly so that it would drive the rear axle rather than the centre one. What I should have done: look up the gear meshing data, precisely mark the new bearing location and carefully drill it out in small steps. What I actually did: measure the hole centres as best I could with a Vernier caliper, then spotted an etched hole on the chassis in roughly the right place (probably intended for brake gear), said "that'll do" and very carefully drilled and reamed the new bearing holes about 0.2mm too close to the axle, so when I fitted the 1st stage gear I got a load of tight spots. To make matters worse, for some reason I thought I must have disturbed the rear axle quartering and spent ages fiddling around to no avail. Finally I took the rear rods off and found that the rear wheelset still locked up once every revolution. It's not dead yet, but the odds just lengthened. Richard

I wasn't entirely serious about only needing one set of wheels for pickup. Personally I think Mike Sharman's theory is sound but there are some major challenges making it work in practice on very small models. I'm not just chasing reliable pickup but better haulage power as well. On my J39 I was able to use the tender to apply a bit more weight to the rear axle, but that isn't an option for a small diesel shunter. I have been using Kato four-wheeled tram chassis in N gauge for a while now. They have a very simple compensation system with a pivoting beam on one side of the chassis, and as a result they run far better than a four-wheeled mechanism has any right to. Coreless motor and flywheel too, and all for £30. I have used them as tender drives (28mm wheelbase equals 7' + 7' with a dummy centre wheelset) and put a couple under van bodies. I have one which I want to try converting to 2mm when I get round to it, but the wheelsets are a similar design to Dapol diesels (split frame, gear muffs and pinpoint outer ends) so I'll need to do something with the lathe that I bought. The wheels supplied are 6mm which is a bit dinky for anything other than a powered van. Richard

I think I saw your comments somewhere about most compensated chassis having far too much articulation and I tend to agree: if you really need the ability to run over a match laid on one rail, your tracklaying probably isn't up to much. Once you move one axle too far above or below the centre line the geometry starts fighting you as the rods try to move in an arc. The axle travel I have on this chassis is the difference between the inside and outside diameters of an Association frame bush and that is probably too much. I should probably have bought lottery tickets for tonight as my chassis appears to roll freely with the rods fitted. One advantage of split rods is that I was able to quarter the chassis as a pair of 0-4-0s: normally I'm not too bad at quartering, but this chassis has wheels of considerable antiquity and on a couple of them the crankpin holes have been drilled fractionally out of line, just far enough I couldn't quarter them by lining up the spokes. I could have used the Association quartering press that I bought, but I was so excited by the prospect of a rolling chassis that I forgot all about the press. A couple of wheels ended up loose in the muffs due to excessive fiddling, hopefully drilling and Loctiting the muffs should hold them in place. Motor next, and I've found an old Farish Pannier body to provide a bit of weight. Wheelbase is all wrong, but it's only a Great Western engine so I don't care. Pondering on electrical conductivity: Simpson springs will only work on the fixed drive axle so I might have to arrange some vertical scrapers behind the front two wheelsets. On the other hand, if Flexichas is as good as its proponents claim I should only really need one set of wheels picking up... Richard

A bit of experimentation on a quiet Saturday: I have twice tried building a compensated chassis in 4mm and both times it ended badly. So why not try one in 2mm instead? This all started because I want a little "03" diesel shunter for Billingborough. I have got as far as buying the chassis kit, then started contemplating the problems of a very short wheelbase and minimal room for adhesive weight. The Mike Sharman "Flexichas" approach distributes weight evenly over all six driving wheels, but can I make it work? I needed a guinea pig. This Fencehouses Jinty chassis has been sitting half finished in a box of bits since my previous foray into 2mm some twenty years ago and I don't really have a use for it, so here goes. I had to move the first stage drive gear so that I could drive the rear axle rather than the centre one. To provide vertical movement on the front two axles I bent some 0.45mm nickel silver wire into a U shape, exactly 1.5mm between inside faces, and soldered it to the outside of the frame bush holes (bushes omitted). On a compensated chassis the outer ends of the axles don't bear any weight, but it remains to be seen whether my minimal bearing surfaces will cope with the longitudinal thrust loads. So far it seems to work, daintily stepping over a diagonal piece of wire across the workbench without lifting any wheels other than the one crossing the wire. But I haven't fitted any rods yet so it is probably all about to go wrong. Soldered two piece rods are easy enough to split for jointing purposes, but at that point I'll find out how accurately I have positioned my axle guides. The digital caliper says they are near enough but we'll see. I have a strong suspicion that this will turn out to be one of my heroic failures, but if so I can just remove the wire guides, refit the frame bushes and go back to rigid frames. Richard

Grrr couplings. Have spent an unproductive day working on different designs for the lifting latch. I have something which almost works, but there isn't quite enough metal around the pivot "ears" to be able to drill a couple of 0.4mm holes without breaking through the edge. I have played about with the standard DG latch but I really don't like it at all and would much prefer a more "engineered" solution. I have one more idea to try for now but it's a bit fiddly. Keeps me busy anyway while I try to sell some unwanted N gauge tat to raise funds for Billingboro's baseboards. Cost of living crisis and all that: my modelling is going to have to be self-funding for a while. Richard

Looks just as lovely in blue as in green. As you say, a good example of how you can still get plenty of operational interest into a small space. I have been carrying around an idea in my head for several years, for an inner London small terminus vaguely inspired by the Kings Cross suburban platforms, and set around 1970. Blue 57' non-corridor stock, Cravens and Derby DMUs, bit of parcels traffic and maybe a through freight line running along the back for some inter-yard trip workings to add variety. Soot, decay, cigarette adverts and that feeling that hung over North London through the 1970s, that it had never recovered properly from the war. The DJM "Baby Deltic" would have been perfect, pity that never happened. Someone should build it even if I don't. Richard

A bit of reading for a quiet Bank Holiday weekend. When I came back into the hobby about ten years ago I had very little space available, so I set about constructing an N gauge branch terminus in the traditional manner. Straight away I was into problems with couplings. I wanted to be able to operate the layout "hands off" and no way could I do that with the standard Rapido coupling. I tried Microtrains knuckles and found that although they work very well with nice big American bogie boxcars they aren't so great with the standard 10' wheelbase British four wheeler, especially in N gauge where the vehicle can yaw some ten degrees either side of the centre line. So I set about designing a coupling that would meet most of the following needs: automatic uncoupling including delayed uncoupling, easy fitment to N gauge models without major modification, easy to assemble, robust, inconspicuous and cheap. Over six years and seven redesigns I went from the coupling on the left (my first attempt, with hand-filed hook, which somehow survived by hiding in the bottom of my scrapbox) to the one on the right. I named it the "Magpie" coupling because it steals design features from just about every tension-lock since the days of Hornby tinplate. The buffing bar and central hook are obviously DG-inspired, but the coupling uses an open ended loop, fixed delay bar rather than a lifting latch, and is designed around a 2mm square plastic shank which can be shaped to fit just about anything. This close-up shows the main features including the delay bar and the drop arm with a few turns of fine iron wire round the end. This version turned out to work very well, even in exhibition conditions. Not 100% reliable, but certainly better than 90% for first time coupling / uncoupling, and it was very rare for it to fail to work on the second attempt. Problems were usually caused by couplings being slightly out of spec: drop arm angle, delay bar length and the shape of the lifting loop were all critical. I found that the fairly heavy, large diameter (0.7mm) nickel silver loops very seldom suffered from the dreaded "loop clash", so I could have loops both ends of all my wagons and plain hooks on the locomotives, which is ideal for a terminus - fiddle yard layout and avoids needing to find clearance for a drop arm on the pony truck of something like an Ivatt 2MT. Early on in the development process I ran into problems with the first or second vehicle in a train randomly derailing. I traced this to the pivot point for the loop being lower (relative to the horizontal) than the drawhook: when put under significant load the loop would try to achieve a straight line pull from hook to pivot, and in doing so would lift the front wheels on a small, light wagon clear of the track. That was the biggest redesign, putting the pivot in line with the hook. The loop rests on the buffing plate just a few degrees above horizontal. Like many coupling designs, this one requires a small "shuffle" to set the delayed uncoupling function: the two vehicles are drawn a fraction apart over the magnet to flick the loops up above the delay bars, then pushed back together. Not ideal visually but I could live with it: the problem was that any steel content in the vehicles (including axles) would result in the magnet pulling them back together. As it happens most N gauge wagons have had non-magnetic axles for some years but I still had to get rid of steel ballast weights. However, 2mm Association wagon axles are mild steel... So I had a choice: yet another redesign to incorporate a DG-style lifting latch, or stop being awkward and just use DGs like everyone else. I ordered DGs, started reading the instructions, noted the geometry (pivot point well below the drawhook), contemplated trying to solder steel wire to a fine brass loop and wondered whether "loop clash" was as big an issue as some people say. Then I looked at the DG etch and thought "there's a lot of meat around those pivot holes, I wonder if I can fold the pivot "ears" up rather than down". This being the result: I have attached these to one end of two wagons and so far they seem to work pretty well although in need of minor fettling (compare the drop arm angle on the shock open with the 16 tonner above). Next stage is to build a few more and see if they work on "Longframlington". I want to try them with steel axles as well. For now, the main problem I have is with the lifting latch. The loop cross shaft restricts the latch movement, so the latch has to bend through two angles and the end has to be cut quite short. I am having problems getting the latch to drop down under its own weight. Opening out the slot for the latch even a tiny fraction makes it waggle around too much. I might try soldering a small brass strip on top of the latch to add a bit of weight. I realise these couplings aren't as delicate and "fine scale" looking as the DGs but the photos make them look more obtrusive than they are, and the one above would look better with a chemically blackened loop. Richard

That is an absolute monster of a layout and would be a pretty ambitious project in N gauge, let alone 2mm. Wonderful to see it on video. Richard

"Belstone" was my first attempt at building something after very many years out of the hobby. It was a steep learning curve but quite a nice little terminus to operate, apart from that kickback siding. I only had room for a J39 and two wagons beyond the crossover, which meant a lot of shuffling around to shunt the coal traffic. Your semi-curved diamond crossing has Templot all over it, lovely piece of work. I'm glad you enjoyed Longframlington. I'm not sure what to do with it now, might see if I can do another couple of shows while I build Billingboro. Richard

I was wondering when that diamond crossing would come into play. Your trackplan is very similar to a little N gauge layout I built some years ago (apologies for display of horribly coarse and clumsy modelling below) and I suspect you have exactly the same problem I had: you can only shunt that kickback siding two wagons at a time due to insufficient distance between the crossover and buffer stops. One of those ideas which look great in terms of making best use of space, but whose fatal flaw I didn't spot until I started running trains. Layout planners take note.