GoingUnderground

I was under the impression that Metromodels had moved mostly to 3D printing via firms like Shapeways. So there would be very few moulds to catalogue.

I had dealings with Phil going back to Harrow Model Shop (HMS) days and always found him very helpful and a pleasure to buy from. He kept London Underground modelling alive after the demise of HMS, but must have been disappointed to see EFE and Heljan cherrypick his range with the '38 &' 59 Tube stock and the MetroVic Bo-Bo models. I do hope that it is possible for someone to take over from him. My sincere condolences to his family. He will be greatly missed by all LT/LU modellers.

The flange thickness on Trix fine scale wheels as fitted to the EM1 are slimmer than the wheels on Triang models of the same period. So your comments about the coarseness of Trix wheels only apply to the older models with coarse scale wheels such as the Ruston Hornsby 0-6-0 diesel shunter, or the younger models such as the EM1 with convertible wheels which still had the plastic coarse scale flanges in place. On the coarse scale and unconverted wheels the back-to-back at the wheel tread would be around 12-13mm simply because of the thickness of the flanges. As the 1959 Trix catalogue said about the Ruston Hornsby loco "This model is only available for Trix Track, being fitted with non-scale wheels only.".

I have several Trix EM1s, so the following info may be of use to anyone thinking of buying one. Beware of zinc pest/rot. The first one that I owned was bought way back in the late 1960s, less than 10 years after the loco was launched. It already had a bad case of pest in the motor bogie drive shaft housing but the body was and still is completely free from pest. I have another one where the wheels are suffering from rot as well. These seem to be the two components, wheels and drive shaft housing, that suffer most from pest, probably because they were made by Trix in-house and not bought in. None of the bodies that I have, so far, show any signs of pest, which I put down to them being bought in by Trix from a reputable firm who produced their own range of chrome plated zinc alloy bathroom fittings and who appear to have had excellent quality control standards. I recently replaced some bathroom fittings made by that firm after almost 40 years of use, and they showed no signs of pest. The only reason they were replaced was modernisation of the bathroom.. The 6 individual roof vents are each held in place with a peg that fits through a hole in the body. Treat these vents with care as the plastic does seem to go brittle with age and it is possible to break off the vent by accident and lose it if the loss goes unnoticed. The pantographs seem to be quite robust, but the plastic pegs that locate them on the roof seem to have the same brittleness as the roof vent plastic. The early models had coarse scale plastic wheel flanges applied over the fine scale wheels. The flanges can be removed by cutting them off and once removed cannot easily be refitted. If you think that Triang wheel flanges are coarse then you won't believe the thickness of the Trix ones. There is no such thing as a "3 rail" Trix loco in the way that Dublo locos were 3 rail. The Trix "Twin" system used the centre rail as the common return, unlike Dublo who used both running rails as the common return. On the Trix locos, the wheels are insulated from each other and from the chassis, and so will run on ordinary 2 rail track by adjusting the pickup shoes accordingly. I think on some Trix locos, the chassis was live to the 3rd rail, but that doesn't seem to apply to the design of the EM1, at least not to the examples that I have. The pickup shoes are not terribly attractive, but it is possible to make your own pickup wipers running on the back of the wheels. You will need pickups for each side for the reason described above. The loco is made to 3.8mm scale, so will look slightly small next to true 4mm scale models. The shape of the cab roof and loco front is incorrect. When seen from head on, the front corners of the roof above the cab front windows on the Trix model curve downwards towards the sides quite noticeably, when it should be almost flat like the Triang EM2. It makes it look a bit like it is melting away at the corners. It is possible to disguise this by painting the roofline so that it is horizontal above the cab front windows. But given that the design dates back to the late 1950s when model railways were primarily toys it's very acceptable, at least to me when compared to the real thing. And it has pantographs that will run touching the contact wire without distorting out of shape which is more than can be said for the Heljan EM1, and will run and pick up quite happily under Triang Phase 1 and 2 catenary as well as under the somewhat rarer Trix catenary system. It shares running gear with the Warship, the main difference being that the bogies were further apart on the Warship so it is not possible to do a straight body swap, but it is possible to cut'n'shut a Warship chassis so that it fits into the EM1 body, having done one myself. On this loco I will be remodelling the Trix EM1 bodyshell to convert it to "Tommy", which will probably mean that I will end up removing the curvature of the cab roof above the cab front windows.

I think Zinc pest or rot is rare on Triang locos of that period in my experience, unlike British Trix. Ex-works flange thickness is probably the issue on points catching the check rail, combined with the depth of the flange being such that it is running on the sleepers on code 100 track. In the 55-65 period Triang was still aiming for the toy market, not the scale modeller, and reliable running on loose lay track on the floor was still paramount.

Rather OT I'll admit, the late Kenneth Horne, famous to my generation for being the "straight" man on the BBC Radio comedy series "Beyond our Ken" and "Round the Horne" was the chairman of Chad Valley.

Regrettably, the pantograph is to the same poor standard as on the Olivias EM1 and EM2, purely for ornamentation and suitable only for running in a fixed position just below the contact wire as it just pulls to one side if run with the pan head touching the wire, or at least that's what mine does.. A lovely model let down by the pantograph. So get ready to lay 3rd rail if you want to look prototypical.

It probably never had a transfer. The lion crest on Miniature Construction and early Trix branded AL1s was a separate plastic moulding, not a transfer. But I would agree that the body side where it would have been is remarkably free of any glue residue. Matthewman describes the crest as "..very clumsy in appearance and was formed from plastic using a brass mould.". The Trix AL1 is widely regarded as a good model, which it is in many respects. However, the early ones were, IMHO, spoiled by the wire pantographs and the awful plastic lion crest. The ones after 1964 were much better with improved pantographs, the lion crest applied as a transfer, and the Trix Express motor..

The bogie casting is never a Triang TC diesel one, not even a very early one where the X.04 motor was used. It looks like the standard Trix one shown on page 189 of Tony Matthewman's book, but I don't know whether that was also used on the Miniature Construction models as well.

What you've got may not be a Trix AL1 at all, but one sold by Ernest Rozsa's Miniature Construction company, which he founded to import Liliput models into the UK. Rozsa designed the AL1 model, Liliput made most of it in Austria, but with "final assembly" in the UK by Miniature Construction to avoid import controls and duties, the "Final Assembly" being fitting the pantographs and couplings. It's not clear whether Rozsa sold the rights to sell the AL1 to Trix or retained the rights himself as owner of Miniature Construction and brought the AL1 with him when he joined Trix. The clue to the age/identity of the loco is the pantograph which is made of wire, not stamped tinplate. If you have a copy of Tony Matthewman's book, look at the picture on page 191 of 4 different "Trix" AL1s, one above the other, and you'll see that the pantograph matches the top image, which is of an "ex-Miniature Contruction model" whilst the lower 3 are all all have a different pantograph which more closely resembles the Dublo Triang one. The AL1 became a Trix model in 1962, but the older style of pantograph, which you have on your model continued in use until mid-late 1964, when according to Matthewman: "A few design changes took place during the following months including the redesigning of the tooling for the pantographs. Gone were some of the rather flimsy pantograph components constructed from wire, and in their place tinplate pressings and a skate assembly purchased from Trix Express. [Trix Express was the German Trix company]. The result was a very efficient assembly with a good resemblance to the full sized version." I have one like yours myself, but it's tucked away at the moment. But your motor is pure Triang Margate X.04. Don't feel too bad about that as Triang/LinesBros. were offered a chance to buy Trix at one point, but Richard Lines turned it down. Just think of it as a "mytabin".

Toys were subject to purchase tax, electrical equipment wasn't. That's why the Triang controllers prior to the introduction of VAT never appeared in the Triang/Rovex catalogues to avoid any inference that they were toys subject to purchase tax. The purchase tax situation was quite complex as regards electrical equipment. I once asked the managing director of a small electrical manufacturing company who made transistor radios why they didn't make a cassette radio - such things didn't exist at the time, at least there were none on sale in the UK that I knew of. His reply was that it was better to sell them as separate units as one attracted purchase tax and the other didn't, sorry I can't remember which did and which didn't. However, if they were combined into a single unit then the whole unit would have been subject to purchase tax. Or at least that's what he told me.

If I remember correctly, Smith Corona typewriters used a "Golf Ball" type head, not individual lever-operated keys which is the sort of Typewriter to which I was referring. Space would have been at a premium on the Golf Ball, so the use of a single character for the capital letter "O" and the number 0, and the capital letter "I" (or was it the lower case "l") and the number 1 isn't surprising. Other makes of "Golf Ball" typewriters may well have done the same. And it is also a question of the design of the typeface whether any distinction is made between O and 0 and between I, l and 1 And from what I have read, I understand that when speaking Germans call it "Halb-Null" sometimes shortened to "Ha-Null" meaning "Half Null" or "Half Zero" which is why they use the number 0. Whilst Null exists as a word in English it is hardly in everyday use and most folks may only have heard of it in the context of something being said to be "Null and Void". The German version of Wikipedia https://de.wikipedia.org/wiki/Nenngröße_H0 makes interesting reading on how the origins of H0 looks from the German side.

A but late to reply, but the terms "positive" and "negative" when used with DCC are meaningless because DCC isn't a constant polarity system like conventional DC. Just think of it a being similar to (but not the same as) AC, but that the polarity changes at a much, much higher frequency than AC mains. And, of course, even on a DC system the polarity is reversed when you reverse the loco so positive and negative would be equally meaningless on a DC controller with a reversing switch or bi-directional throttle knob. It's reversing the polarity on DC that makes DC locos run in the other direction on DC layouts. You need the letters because when you add a booster the wiring from it must have the same "orientation" as the ECoS. or CabControl, and of course to ensure that you keep the same orientation (ESU do call it "polarity" in their manuals) throughout the layout. The B and O matter much more when used with locos equipped with older MM or mfx decoders and Maerklin K83 and K84 accessories may not work correctly on 3 rail systems if the layout isn't wired with the "B" wire going to the centre rail/stud contact and the "O" going to the running rails, note plural "rails" as in this context I'm talking 3 rail not 2 rail. In the 2 rail context all that matters is that you keep the orientation constant throughout the layout. And for the record, "B" and "O" do not refer to colours. They were used by Maerklin in connection with their wiring for their 3 rail system where "Bahn" meaning "Train" was used for the B wire that goes to the stud contact centre rail, and "O" meaning "0 volts or Zero volts" or in English the "Ground" or "Earth" for the O wire that went to the running rails. With thanks to Tom Smit on ESU's Forum who provided that information most recently on 29 November 2021.

As I pointed out earlier, Fred highlighted the fact that scale was a bit of a moveable feast at the beginning with no common standard. My own view is that it is not possible to say with any certainty exactly when OO gauge first appeared in the UK only that it was popularised by Hornby with their Dublo range, and the launch of Dublo marked the point when OO separated, or evolved if you like, from what had been a somewhat confused situation across the board in 00. You might like to mull over the words of that eminent modeller Dudley Dimmock over the origins of OO in his foreword to Michael Foster's 1987 magnum opus on the Dublo system "Hornby Dublo 1938-1964 The Story of the perfect table railway": "The word Hornby Dublo, of course, is coined from the phonetic pronunciation of "00". The early Trix trains were really 'H0' and to understand how these symbols came to be used one must delve into the history of model railways. To differentiate between the various big gauges of the past, the trade gave the symbol '0' to trains that ran on 1 1/4" gauge track and '1' to those that ran on 1 3/4" gauge track, and so on. As the measurement of the gauge '0' track was then 33mm, when Trix started their small gauge they called it 'Half 0' which became abbreviated to 'H0' and the trains ran on a track of 16.5mm (half the gauge '0' track). As an early modeller and builder of small locos, I found as did countless others that 'H0' used a scale measurement of 3 1/2 mm to the foot which made it very difficult t oget small motors of that period into the body of the loco. Because of this and other reasons of wheel clearance and bogie swing on locos, we decided to model in 4mm to the foot whilst still retaining the same gauge of 16.5mm which was used on the 'H0' 3 1/2mm scale system. This gave us bigger loco bodies and answered nearly all our problems and saved us making and creating a separate track system for for the 4mm scale which which would have entailed expensive jigs and tools. To differentiate between the 3 1/2mm H0 scale and the 4mm scale with their larger bodies, it was decided to call the new 4mm system '00' gauge or phonetically 'Double 0 gauge' and Hornby seized upon the idea because of the increased body size which enabled them to fit in their motor and coined their own name of 'Dublo'." There are some factual errors in the above as Trix and Maerklin actually called their 1935 models 00 not H0 or OO in 1935 as is shown by their German catalogues from that date, and that H0 was only adopted as the universal name for 3.5mm:1 foot models running on 16.5mm track after WW2. Also the initial track gauge was 16mm not 16.5mm. However, Dimmock has explained the origins of British OO, (or 00 if you insist there is no difference), as 4mm bodies on 3.5mm track. However, in the late 1930s calling the British 4mm scale/16.5mm gauge system as '00' would not have differentiated it from the '00' name already used since 1935 by Trix and Maerklin in Germany for their 3.5mm scale/16mm (later 16.5mm) system. That is only possible if the acronym for 4mm/16.5 is OO. The German Trix 1938 katalog https://www.conradantiquario.de/content/katalog/trix-1938.html Describes their system as being 1:90 and calls it '00'. The German Maerklin Katalog for 1938 https://dermodellbahnblog.de/maerklin-katalog-1938-pdf-download.html calls their system '00' as you will see if you scroll through the pictures in the link. British Trix called their 3.8mm scale on 16.5mm gauge track '00'. Looking back, the 3.8mm scale was probably their solution to the big motors and tight curves vs small bodyshell problem. Meccano went for 4mm on 16.5mm track for Hornby Dublo, for the reasons described by Dimmock and coined the name 'Dublo' even though '00' had been in use since 1935 by Trix and Maerklin in Germany. Meccano would have known that Maerklin were using '00' as there seem to have been a flow of information between the two companies since the early 1920s, when anti-German sentiment after WW1 was still running high, through to 1938 when Dublo launched with tinplate track that closely resembled the Maerklin 3 rail track of the time. I leave you all to draw your own conclusions on when 00 meaning 3.5mm scale on 16 or 16.5mm gauge track, and OO as 4mm scale on 16.5mm track each started.

I have seen references previously in other threads where Maerklin were said to have used 16mm track, and I think that someone has already said earlier in this topic, or it may have been Fred in his booklet on Scale and Gauge, that the gauge was increased to 16.5mm when they stopped making rails from folded tinplate and moved to square section brass rod, the increase was to allow the wheels to run on the flat part of the wheel, not on the curved radius to the flange.

In spoken English, yes because we, and I include myself, habitually use "Oh" to refer to both the numeral Zero 0 and the letter O. But your computer doesn't, it treats the number Zero as totally different to the letter O. That is why in the world of computers where numbers and letters can be used in the same string the numeral Zero is written with a forward slash (/) through it whilst the letter O is written simply as the letter O. Edit: Traditional typewriters have different keys for the number 0 and the letter O with the number 0 usually being of a narrower width than the letter O, as you will have seen when you read this. The phonetic alphabet uses "Oscar" and not "Oh" for the letter O. I've not used the phonetic alphabet for numbers, but when I have to read out combinations of letters and numbers to people I always use "Zero" for the number 0 and "Oh" or "Oscar" for the letter O, that way there is no confusion. Edit End. In German, and the gauges were originally named by Maerklin, 2 - 1 - 0 are a numerical sequence, not 2 numbers and a letter. Hence 0 meant Zero to indicate that it was less, i.e. smaller, than 1 gauge. Having got down to Zero how do you name a gauge that is smaller than 0 that is approximately half the size of 0 gauge? The solution seemed obvious to Trix in 1935 - add another Zero as I think that was the first time that 00 was used to describe their 1935 products followed rapidly by Maerklin.

I'm going to play "Devil's Advocate", so be warned, ymost of you may well disagree with all of the following observations. What no one has pointed out is that in the beginning it was 00 (Zero-Zero) to indicate that it was approximately half the size of 0 (zero) gauge, which itself was so named to distinguish it from the larger Gauge 1 and its even bigger sibling Gauge 2. I believe it was Maerklin who first created the gauge numbering of 2 - 1 - 0 to distinguish between the progressively smaller scales. The first 1922 locos appear to me to be generic designs rather than models of specific prototypes and that at least for me makes the scale actually used a bit academic - how can you scale something that doesn't exist. The German language version of Wikipedia makes interesting reading describing the Bing 1922 models as 00 (Zero Zero), not OO (Oh Oh), being based on half the scale of gauge 0 (Zero) with the models being "strongly focussed on the English market" despite the Bing company being based in Nuremburg. So in that sense, there is a case for saying that 1922 was the beginning of not of 4mm scale OO (Oh Oh) gauge but of sub-0 gauge models in general, and 3.5mm scale H0 gauge in particular. I have a copy of the Trix Twin Railway "Permanent Way Manual" 14th edition dated January 1952 originally written by Henry Greenly in 1937. You might find this section on page 9 relevant: "SCALE AND GAUGE The gauge of the railway is 5/8inch (16m/m) and is commonly known as "00". The scale, which is that originally determined by the writer, [meaning Greenly himself] for this gauge is approximately just over one-eightieth (1/80) of a full-sized railway, the measurements of the details being, as far as practicable, made in this proportion." Please note, the book does use "00" not "OO". Taking Greenly literally, the relationship between scale and gauge is already broken as 5/8 or 16mm track gauge is 1/90 but he describes the scale as 1/80. It could be argued that 1/80 is closer to 1/76 OO than 1/87 H0, or that Greenly was only talking about the British Trix system with its 3.8mm scale, which is 1/80, but he seems to be taking credit for creating this scale and gauge and it could be argued that he was referring back to his involvement in the 1922 launch. The Trix system was arguably the descendant of the Bing 1922 system as Stephan Bing was also involved in the German Trix company who launched 00 proper in Europe in Spring 1935, shortly before Maerklin announced their 00 models. The Bing family lost control of the German Trix company in the mid 1930s and Bing emigrated to the UK in 1938 where he became involved in the British Trix company which was closely tied to Bassett Lowke at the time through the Bassett Lowke Twin Table Railway which historically was the German Trix system with UK outline bodyshells. As Fred has pointed out in his booklet, the actual scale used for 00 varied between manufacturers and over time, eventually settling on 1:87 which itself is based on the scale of 16.5 mm gauge track, and changing the nomenclature to H0 which more accurately describes it as "Half 0" in German, to distinguish it from the conflicting UK and US use of OO in connection with 4mm scale. Nearholmer has pointed out in past posts on RMWeb that OO existed in various forms prior to the launch of Hornby Dublo in 1938, but personally I would regard that as the launch of OO proper, i.e. it's adoption by a major toy maker of 4mm scale on 16.5mm track. I don't want to reopen a lengthy debate over the origins of OO (4mm on 16.5 track), that's water under the bridge, but wanted to raise information from relevant sources that others may not have seen.

Apologies for telling you things that you already know, but if you ever revise your booklet, you might like to add in the rather odd scale that British Trix used. In their sales literature it was described as OO, and the track was 16.5mm same as its German counterpart, but as more realistic locos, coaches and wagons were introduced were built to an intermediate scale of approximately 3.8mm to the foot, 1:80, neither true OO nor true H0. The exceptions were the Bo-Bo AL1/Class 81, which was't designed by British Trix but came into their range when Ernest Rozsa joined their company, the A3 Pacific "Flying Scotsman", the A4 Pacific "Mallard", and the A2 Pacific "A.H. Peppercorn". The result was that the 3.8mm scale British Trix locos and coaches were visibly smaller than the rival products made by Hornby Dublo and, from 1952, Rovex under the name Triang Railways. This, and the fact that British Trix retained AC power and very coarse scale wheels and track into the mid 1950s, had a significant impact on the ability of British Trix to compete in the UK market. In the UK, there is no "coupling salad" in OO. Through the 1950s and 1960s there were two rival systems, the Peco coupling, used by both Hornby Dublo and British Trix, and from 1952 there was another coupling system introdiced by a new entrant into the market, Rovex trading as Triang Railways. In 1959 Rovex introduced their Tensionlock (or Triang Mk3) coupling. The failure of Hornby Dublo in 1964 and their acquisition by Triang, coupled with the fading away of British Trix in the early 1970s, left Rovex, who were trading under the name Hornby Railways from 1972, as the dominant player in the UK and their Tensionlock coupling as the de facto "standard" British coupling for ready-to-run models. For some time before Trix ceased trading their models were made so that owners could fit the Triang Tensionlock instead of the Peco type Trix coupling. G&R Wrenn, who became a subsidiary of Triang in the mid 1960s and started reissuing the former Hornby Dublo models altered the tooling so that they also used the Tensionlock but with the option of substituting Hornby Dublo couplings if the owner so desired. Slimmed down versions of the Tensionlock were subsequently introduced by competitors, notably Kader trading as Bachmann Branchline in the UK, but they are still recognisable as the 1959 Tensionlock design. Hornby Railways also came to use this slimmed down version as it is less visually intrusive than the original Tensionlock design. You only briefly touch on track, which I find surprising given that you mention the breakdown in compatibility at OO/H0 level when today there are two competing methods of delivering power to locos: 3 rail, nowadays in the form of stud contact, and 2 rail, whilst Trix both in the UK and Germany until the 1960s used their hybrid "twin" system which allowed 2 trains to run on track with a centre conductor rail by isolating the running rails from each other and using the centre rail as the common return. In the UK 3 rail died with the end of Hornby Dublo although it does still have a following in Hornby Dublo enthusiast circles, and those who model mainland European railways using Maerklin locos and track, a very specialist niche of the UK hobby. Its passing is so complete here that most UK modellers, including myself at one stage, don't realise that it is still very popular and readily available to buy from new in those countries where Maerklin is the market leader with manufacturers making both 2 rail and 3 rail versions of their locos. But thank you for the link, it made interesting reading.

And just to add to the havoc and confusion, Maerklin used to call their 3.5mm models 00, and only changed to using H0 in the late 1940s. And I think the USA had OO, but theirs ran on 19mm gauge track.

To be pedantic, it was going to be called the Fleet line when the original route via Aldwych and Ludgate Circus was under consideration.

Did I say that the sole reason for HS2 was speed? No, I didn't. My point is that it's not worth building HS2 as a high speed line if it's only going as far as the West Midlands. I other words by all means build it to increase the capacity along the WCML corridor, but at conventional speeds as were used on parts of its route when that was used by the Paddington-Birmingham and GCR London-Rugby services. I never talked about the GWML, only the secondary route via West Ruislip which was closed under Beeching as it duplicated the Euston-Birmingham route. And my comment about the state of services on the WCML was made to highlight the irony that whilst HS2 is primarily about capacity rather than speed, the actions of some TOCs have resulted in the existing track capacity becoming underused because of cancellations. I am all in favour of HS2, just not the cut-down version that we're in all probability going to get.

Further to my last post, zinc corrosion is perfectly normal, and that is why galvanising, coating steel with zinc, stops the steel rusting. When in contact with steel and moisture, the zinc, being more reactive than the iron in steel, will corrode first and in the process stop the steel from rusting. The zinc is used as a "sacrifice" metal, being the one to corrode first. Eventually enough of the zinc galvanising will have corroded so that it can no longer protect the steel which will then start to rust. That is why all galvanised steel products eventually rust. They don't disintegrate as the zinc galvanising is only a surface coating and it is the underlying steel that gives it its structural integrity at least until the steel itself corrodes to rust.

Which makes me wonder whether they could have achieved the same result by restoring the GCR to Rugby, or the GWR route from Paddington to Birmingham and settling for more capacity at lower top speeds. High Speed lines are more effective the longer they are, and hence I agree with HS2's opponents that High Speed London to Birmingham is a waste of money. Don't get me wrong, having travelled on the TGV services in France I support HS2, but HS2 in its entirety, with purpose-built extensions beyond the West Midlands to the East Midlands, Manchester, Leeds, Glasgow and Edinburgh, not what looks to be a truncated high speed service just to the West Midlands and then running on existing tracks to destinations beyond the West Mids. Do it right, or don't do it at all is my view, and always has been on HS2. But right now trying to get to the NEC for the Warley show by train later this month is a nightmare. I gave up trying to see times and tickets a week ago as did my brother who was also going to travel by train. We're both going to travel by car with all the consequent pollution and greenhouse gas emissions. Capacity doesn't seem to be an issue with so many trains apparently not running on the WCML

Triang published instructions in their "first 10 Years" on how to remove these sleeved wheel axles, but "use a suitable pin" to push out the axle gives the wrong idea about the amount of force required, at least on older models. I've found that many of the 1950s-mid 1960s wheelsets, with their metal bar and half sleeve plastic axles have problems when removing the axles because either the axle has surface rust inside the sleeve and expanded or the plastic sleeve has shrunk slightly and tightened on to the axle, in which case there is no wheel that rotates independently of the axle. Ideally you shound push out the axle from the freely rotating wheel side as the crimping on the axle only holds one of the the sleeved wheels in place on the steel axle. That way you are pushing against the crimping for the shortest possible distance But if you can't rotate one wheel independently of the axle than you don't know which way to push. This seizing -up problem seems to be quite common, especially in models showing signs of rust on the couplings, an indication that at some point they have been stored in dampish conditions. Fortunately, in my experience, there is often just enough "give" left in the "free" wheel that you can rotate the wheels independently of each other of you try hard enough. When you do this, keep an eye on the end of the axle, and you should see it rotate with only 1 of the wheels. letting you work out which is the "free" wheel, and you push out the axle from the free wheel end. Take care as if the bogie or chassis is plastic it may have become brittle with age, as the youngest example is likely to be coming up to 60 years old. I use a spare axle as a drift, and place the other axlebox into a shallow recess in a small block of wood with a hole drilled through the wood. That way when I tap on the drift it can push the axle out as the bogie side which is holding the wheel in place is supported and is less likely to crack or in the worst cases shatter. The early axles of this type had flat ends, whilst the later ones were domed, so I use an early axle for my drift as it will sit on the end of a later domed one, whereas with dome to dome contact, on striking the end of the drift, the drift will try to slip sideways round the axle and between it and the bogie and if it is plastic this often will break the bogie - personal experience. Also if you do use an old flat ended axle as your drift, make sure that the crimped end on the drift axle is closest to your hammer, or you risk getting the crimp on the drift stuck in the "free" wheel's sleeve and you'll have to resort to pliers to pull it out again. Here is my kit for removing these old wheelsets. The flat ended drift axle is sitting in the hole in the block. The wheelsets to the left of the block have dome ended axles and you can just see the crimping next to the wheels, and also the crimping on the drift axle. I don't tend to replace these older axle & wheel combinations with pinpoint axles and bearings, but when I find a seized example I do remove them as freeing off the "free" wheel will let the coach or wagon go round curves more easily as seized plastic wheels exert more drag on curves than modern metal wheels fixed to their axle.

The wheels on the Bogie Well Wagon are 10.3mm in diameter, so you could use Peter's Spares PS33 10mm Container and Lowmac Wagon replacement wheelsets. I've used them to replace missing wheelsets on the Triang container flat with good results, the metal wheels have a bit more mass than the plastic originals which gives it a little bit more inertia and it seems, at least to me, to sit on the rails better due to the slighty higher overall weight.