The Bodgerigar attempts to build a Collett 1938 31xx

[Barclay]

Have you tried turning out the lights so you can see exactly where the short is occurring? If something touching the frame is an issue then I assume the wheels on one side are live? If this is the case then if a coupling rod grazes the wheel rim on the non-live side there is the chance of a short through the wheel itself or the motion, and a spacing washer might fix this. Don't give up - one thing at a time until you have it cracked!

[1466]

1 hour ago, Barclay said:

Have you tried turning out the lights so you can see exactly where the short is occurring? If something touching the frame is an issue then I assume the wheels on one side are live? If this is the case then if a coupling rod grazes the wheel rim on the non-live side there is the chance of a short through the wheel itself or the motion, and a spacing washer might fix this. Don't give up - one thing at a time until you have it cracked!

Dear Johnster

i like your choice of engine and admire your analytical, methodical approach. We share the same humour.This thread is very entertaining and  I always look  forward to new instalments but I’m as mystified as you by the intermittent shorting .

I’m trying to be helpful so Grandma and eggs time . Do you recall you had problems with a tyre detaching from a Markits wheel ? And you superglued it back on ? Do you think  that may be the source of problems?

I know Araldite is non conducting but don’t know how superglue works .

Ken 

[The Johnster]

Both valid and cromulent points, guys. 

 

I have tested each wheelset individually for shorting by placing them on the track and applying power, and looking at the overload light on my Gaugmaster.  They are all fine; all wheels are insulated, and superglue is an insulator, but, being Markits, the insulation is between the wheel and the rim, whereas the Airfix wheels insulate between the axle and the wheel.  This means that it is possible for the uninsulated spoke part of the wheel to contact the chassis block or the pickups, and this may be a part of my problem.  It was with this in mind that I made the insulating side plates, but their presence may be forcing the pickups too firmly against the wheels.  I will also have to keep an eye on the more limited sideplay on my tighter curves, which may also induce shorting!

 

AFAICT, the Airfix chassis is not a 'live return' type, as it has pickups on both sides and an insulated supply of current to the motor from both sides.  It is therefore essential that no current gets to the chassis block, and it looks is as if there is contact between the pickup strip on the upper face of the keeper plate and the bottom of the chassis block, which makes no sense and suggests that there is something I don't fully understand going on here.  This is the root of my thoughts concerning an electrically dead chassis block a la Lima, in the form of a home made block or Kitmaster construction kit frames with bushed axle holes.

 

The visible fireworks are between the rear wheels and the block or pickups, and seem to occur despite the provision of the insulating side plates, which suggests that the culprit is an interface that shouldn't be there between the wheel spokes and the pickup.  This is steering the various synapses and flickers that pass for thoughts in the ruins of what was, they tell me, a finely tuned precision instrument, my brain, towards replacing the pickups, perhaps with spung plunger types.

 

I will hold off actually doing anything to 3100 for a few days, perhaps until the weekend, while all this thinking brews and ferments, possibly towards a constructive conclusion!  If nothing else,  the project is developing into a mental challenge as I struggle with the problems of sending electricity where it is supposed to go and nowhere else while maintaining a smooth and free running mechanism.

 

The motor, despite it's age, is a very smooth and quiet runner that has a good bit of potential to provide a decent performance and, looking at the amount of carbon left on the brushes, should last me out at my age!

 

One thing at a time until it has me cracked...

[The Johnster]

Despite promising myself I'd leave it until the weekend, and in line with a lot of past broken promises involving refraining from doing things with trains, I had a bit of a poke around yestereve, and have identified and cured a short.  It may not be the only short of course, but it is at least a box ticked and a step along the road to an operational chassis.  With the driving wheels removed, I placed 3100's chassis on the track supported by the front pony and trailing radial.  Short!  Investigation showed that these wheelset, insulated at the axle/wheel interface, are only insulated on one side and I'd got a set in backwards, hence there is a short running from one side of the pony throught pony frame to the block and at the rear of the blockh through the radial frame to the other side.

 

Now, this is a further puzzlement because the loco had been running as 4145 on the original 23mm driving wheels for some time before I began the rebuild, but there is now no chance of a repetition as I've reversed one set of trailing wheels.  They are in any case going to be replaced by better Markits versions; can't see me living with the Airfix solid backed ones for long!  But I'm more and more convinced that the root of the problem is the very narrow rear of the Markits rims on to which the pickups must bear without straying into inner wheel territory.  The pickups will need very precise locating to be effective and they will go out of adjustment far to easily, and may need some thought; I really am going to leave this to the weekend!

[Nick Gough]

The Airfix prairie radial truck acts as a power pick-up on one side

I believe this is to compensate for the traction tyre on one of the driving wheels.

 

I've just investigated erratic running on my grandson's example. I was surprised to find that the driving wheelset was fitted to the loco upside down, compared to my ones, causing the traction tyre to be on the wrong side. Rectifying this has made a significant difference to the running.

[The Johnster]

That makes sense, Nick. 

 

Nothing to report in the way of actual material progress, but I am increasingly certain that the shorts are a result of the Markits wheels being isolated at the rims, while the Airfix originals are isolated at the axles.  The flashovers and sparks are at the rear of the wheels.  As I noted previously, this is going to be an issue unless I can devise some means of adjusting the pickups so that they do not impinge upon wheel centre territory, and I have history with pickups; me and pickups do not like each other and I prefer to set them up and leave them alone.  A chassis on which the pickups need constant adjusting is a 'mare, as are the shorts.

 

So I've been putting some thought in, which I should arguably have done weeks ago.  It is beginning to look as if the idea of markits wheels on this chassis is not the flyer I thought it would be.  I've been looking at the Wizard Comet kit, which would need a motor and gears to complete; the Airfix motor and gears are no good for it as there is no means of meshing the worm and cog; it will need a suitable motor/gearbox combination. 

 

I've also thought of using the Kitmaster plastic kit frames with brass bearings, perhaps assembing it for split pickup, but the same issue of motor/gearbox arises.  Another possibility is to adapt another RTR chassis that has 5'3" wheels at an axle spacing that is not too far off for a layout model, remove the cylinders and Walchaerts if there is any, and graft some sort of mounting for the cylinders/motion bracket and the body 'top'.  3100 is a 'layout model' at best, and I am not attempting museum accuracy.  If it looks tolerably like a 31xx, I'll be happy enough.

 

The big advantage of this approach is that I would be building from a starting point of a functioning chassis and mech, on which the pickups are already working and which is free of shorts.  The big disadvantage is cost, of course, but probably not much more so than getting a Wizard/Comet to operating condition and much less likelihood of me messing it up (or 'differently improving it' as I prefer to call it...).  Cost must be measured against the point that I have a locomotive which is a paperweight at the moment...

 

Candidates so far are:-

 

.Ivatt 4MT mogul

.BR ditto.

.BR 3MT prairie tank.

.LMS Fowler 4F.

 

Axle spacing should be spot on for the first 3, though I'm not sure of the 4F, and the coupling rods would have to be reversed and filed into a fishbelly profile if the spacing were different to the GW standard.  The Fowler 4F has the advantage that it would not need to undergo surgery to remove cylinders or motion.  If anyone can add to the list, I would be interested to hear your thoughts!

 

Onwards and upwards (or more like widdershins and backwards in my case)!

 

 

[DCB]

I haven't been following this thread but have been struggling to make two good Airfic 61XX from four.   There is no way to stop Romford / Markits wheels shorting with airfix pickups.   The standard wheels have insulated centres.  The tyre pushes on from the inside of the wheel and has a flange to stop it pushing right through as well as the flange to keep it on the rails so about 2mm of rubbing surface for the pickups which only rub insulated plastic if they rub the wheel centre.   The Markits have nearer to 1mm and short if they touch the wheel centre. Short them tyre to centre enough times and they develop a conductive track through the insulation, see my Comet Chassis woes thread.   The  wobbly flimsy construction of the pickups lets them flop around far more than the Markits can tolerate.  The Airfix axles are a very floppy fit in the chassis which does not help. My fix, copper clad PCB fitted along the bottom of the plastic keeper plate with wire  pickups coming out through the brake gear to rub the Markits flanges.   With the floppy centre axle the middle wheel pickups probably need to rub the tyre tread, my Wills and Farish Prairies have run with with centre pickups rubbing the tyre for 30 odd years with no real problems.

The RTR 21mm wheels mentioned have much shorter crank throws  as do your Markits, It really needs   30" stroke (5mm throw) and not the generic 4mm throw type to get that long stroke GWR gait.     

 

Edited November 6, 2021 by DCB

[DCB]

Have you thought of turning down the Airfix tyres.  The Airfix wheels are around 22mm and have a nice long crank throw. The Traction Tyre groove must be around a millimetre deep and I measured the base of the groove at just under 21mm or a scale 5ft 3" which is spot on for a 1938   31XX.    I reprofiled some 14XX traction tyre grooved tyres once, not sure why, using a set of needle files. It needs a mandrel to take the tyre but once that's made they can be done in a black and decker drill chuck.

On my Airfix 61XX I fitted spare non grooved tyres to replace the traction tyre tyres on the rear wheels,  Just the tyres, I  take a wheel off, remove the tyre, replace it with one from a spare wheelset and push the wheel back on, that way it goes back on the same spline and the quartering is either spot on or way out, changing wheels makes getting the quartering right a bit of a nightmare, however the process would work in reverse with turned down tyres, ideally starting with 6 grooved ones which everyone usually throws away.....   

[The Johnster]

The Airfix wheels are, assuming they are to scale, 23mm and the 31xx scales out at 21mm, which would need that the Airfix would need to be turned down by 1mm radius to reduce the size to the correct one.  A brief visual check, involving lining up a 20mm Markits and an original Airlfix suggest that turning down the Airfix wheels is indeed possible, that there is enough 'meat' on the Airfix rim.  Not sure what the look of the finished wheel might be, but I have nothing to lose by having a go, and it would be a viable workaround for the shorting problem. 

 

Thank you, DCB, you may have hit on the solution to this.  My only reservation is that it would be destroying any plating on the tyre surface of the Airfix wheels, and would probably need a good bit of polishing to get a reasonably smooth tyre. 

 

It also occurs that wrapping the ends of the Markits axles in clingfilm to isolate them from any current that strays into the inner wheel, the spoke web, at that point could help as well, but this smacks of desperation a bit!

[jcm@gwr]

Have you considered mounting the pick-ups on top of the wheels,

so that they bear on the part of the tyre that is in contact with the rail,

but at 180 degrees from the rail. I've done this on a small 0-4-0 in 7mm.

It should be almost invisible as it'll be within the splasher area.

[The Johnster]

Thinking about that; perhaps with the pickups bearing downwards on the tops of the flanges.  A plastic pickup plate shaped to accommodate the motor and gears sitting on top of the chassis block and glued into position, keeping all the current as far away as possible from the wheel centres, which are indubitably the problem at the moment.  
 

I’m less sure now of DCB’s idea of turning the tyres down.  How would I ensure all the wheels were exact same diameter after turning, essential for good running, and what about the plating on the tyres?

[The Johnster]

Ok, I've gone back to basics, stripped everything off the chassis block and started from scratch, with the aim of somehow picking up current from the tops of the wheels with top mounted pickups.  With this in mind, yesterday evening's session involved putting additional plastic isolating plates on the downward facing part of the block above the wheels, running from just behind the cylinders to the bottom of the curved half-arch under the front of the cab (when it is put back on!).  The plan is that the pickup or pickups will be glued to the underside of these horizontal isolating plates to bear on to the top of the driving wheel flanges, probably a single piece pickup bearing on all 3 wheels each side, and largely hidden when the loco body is on. 

 

Insulated feed wires from these will run around the forward ends of the horizontal plates and then to the motor terminals.  The chassis should then run, smoothly and without shorts (as this build has shown, this could be 'famous last words'), the finishing off process can begin, and 3100 can start earning her occasional keep.  She was between 1946 and her withdrawal in 1957 the usual loco for the daily Porthcawl 'Residential', a commuter service that stopped at all stations between Pyle and Cardiff General on the South Wales Main LIne, but photos prove that she did venture into the mountain fastnesses now and then.  High pressure no.4 boiler, 5'3' wheels, a large prairie on steroids; there can't have been many Tondu jobs that she wasn't too powerful for, even in those steeply graded valleys!  Her ability to get away very promptly from the main line stops with the Residential, 5 gangwayed coaches, was much appreciated!

 

She was withdrawn following a 'heavy contact' with the stop block at Porthcawl, which allegedly bent her frames, though this is not visually apparent in photos of her during the 3 years she was at Swindon awaiting scrapping.  She spent a couple of weeks as the yard pilot at Ogmore Jc before being sent to Swindon for assessment, less than 20 years old, but of course parts of her were older than that having had a previous life as a 3150, and the decision was taken to scrap her.  There were only another 4 in the class, all gone by 1960.

 

Tondu, in most aspects a rather ordinary workaday medium sized South Wales Valleys shed, had form in terms of slightly unusual left field motive power.  The Porthcawl branch, developed from a tramway, had very tight curves and speeds were restricted, which resulted in the 44xx small small prairie, 4'1" driving wheels, being allocated.  These went back to the shed at lunchtime every day to be turned, to even out flange wear.  They were withdrawn in 1953 when 4575s for auto working were allox TDU.  In pre-war days, the Residential was hauled by Tondu based Bulldog 4-4-0s, not your typical South Wales Valleys shed fare either, 

Edited November 8, 2021 by The Johnster

[Halvarras]

Far less likely to get a build-up of dirt under the pickups bearing on the flanges instead of the tyres' running surface. That's the pro, the con could be wear on the pickups due to the concentrated contact area - but unlikely to be a concern if the model would never travel long distances?

[The Johnster]

As a con, it's not a deal breaker; 3100 will be an occasional performer and Cwmdimbath is small and runs at lowish speeds.  I'll plug on with this, current sitrep is investigation of suitable pickup material and how to spring load it against the flanges for reliable contact without acting as a brake, but I haven't ruled out the donor chassis idea completely yet either!

[The Johnster]

Plan B, or is it C, has been put into operation.  Not sure at all that it was the best thing to do but I've been an' gorn an' done it now; I've picked up a Bachmann Ivatt 4MT running chassis on the Bay of e.  The idea will be to remove the cylinders, piston, crosshead, connecting rod, motion bracket and valve gear, and cut the chassis block to whatever extent is needed (the frames ahead of the smokebox will have to go, for instance).  The Ivatt's wheels are the correct size and at the correct axle spacing; I'm not going to obsess about the number of spokes!  The Ivatt's injector plumbing and some other chassis detail will have to go as well.

 

Then, I'll remove the coupling rods and replace them with the fishbellies from the original Airfix 61xx, or new ones from Wizzo/Comet, and devise some sort of mounting for the new cydinders, also probably Wizzo/Comet but possibly Hornby spares.  Forget all that nonsense about pickups and isolating plates, though the rear part of the Airfix chassis will have to be retained to mount the rear radial pony.  Cost is through the roof, up around the £90-100 pound mark now, probably more than I would want to pay for the loco in one  hit, and we're not finished spending yet (new buffers, handrail knobs, cyldinders/slide bars/xhead/conrod_ but it's done now and no going back!   Wish me luck...

 

'That's so crazy it might just work'.

Edited November 8, 2021 by The Johnster

[1466]

12 minutes ago, The Johnster said:

Plan B, or is it C, has been put into operation.  Not sure at all that it was the best thing to do but I've been an' gorn an' done it now; I've picked up a Bachmann Ivatt 4MT running chassis on the Bay of e.  The idea will be to remove the cylinders, piston, crosshead, connecting rod, motion bracket and valve gear, and cut the chassis block to whatever extent is needed (the frames ahead of the smokebox will have to go, for instance).  The Ivatt's wheels are the correct size and at the correct axle spacing; I'm not going to obsess about the number of spokes!  The Ivatt's injector plumbing and some other chassis detail will have to go as well.

 

Then, I'll remove the coupling rods and replace them with the fishbellies from the original Airfix 61xx, or new ones from Wizzo/Comet, and devise some sort of mounting for the new cydinders, also probably Wizzo/Comet but possibly Hornby spares.  Forget all that nonsense about pickups and isolating plates, though the rear part of the Airfix chassis will have to be retained to mount the rear radial pony.  Cost is through the roof, up around the £90-100 pound mark now, probably more than I would want to pay for the loco in one  hit, and we're not finished spending yet (new buffers, handrail knobs, cyldinders/slide bars/xhead/conrod_ but it's done now and no going back!   Wish me luck...

 

'That's so crazy it might just work'.

Who dares , wins !

[The Johnster]

Or in my case, more like, who cares who wins!

 

Thanks 1466, I needed a bit of validation at this point!  I've spend more money than I wanted to on a scheme which is by no means certain of success; what could possibly go wrong?  Anyway, we won't find out until the Ivatt chassis arrives in a week or so; stay tuned for the next thrilling episode of...

 

...The Bodgerigar attempts to build a Collett 31xx!  (cue Dick Barton music.., dern, diddly dern, dern, diddly dern)

 

It was the right decision in one respect; I am feeling more confident about the outcome now and stressing less.  I feel an operating session coming on!

Edited November 8, 2021 by The Johnster

[The Johnster]

And yet another setback, I'm thinking of calling in a priest to exorcise the curse on this project.  This is a familiar tale I know; the running chassis has been delivered by Hermes, who have posted me a photo of it in my porch, but there the story ends; it has vanished off the face of the earth, no doubt into some druggie chav's swag bag. part of the general fun of the mean inner city.  Hope he drops it on his toe!  A replacement has been sourced and paid for, but the cost is now beyond all reasonabl bounds for a project of this sort

[The Johnster]

Cost, by the latest reckoning, is now around the £160 mark and we're not finished yet, but perhaps £62 should be deducted by anyone thinking of attempting a similar rebuild, as this is the lost chassis replacement cost including delivery.  Hopefully not by Hermes leaving it in the doorway and taking an expensive photo...

 

What this project needs now is a bit of good news, and it arrived this afternoon, the replacement eBay Ivatt 4MT.  LMS 3001, early model if the expanded polystyrene packing is any indicator, and missing some of the detail pack parts, a cab window, and the tender coupling pin, but this is of no matter as the bulk of the loco is not going to be used.  Quick test run reveals perfect running and no problems with any of my turnouts (there is always a question mark over the Hornby 3rd/2nd curved in the fy).

 

It's a joy to watch the Walchearts flolloping around and the loco carries a good bit of detail, including the monstrous Ivatt double chimney which apparently badly affected the steaming, and the character of the Flying Pig is well captured.  I don't find them ugly, more 'function over form', and always consdidered that they looked exactly like a post war steam loco should do in the UK, and better than the 'styled' BR 76xxx version.  So it's a shame that all this Ivattry has to be removed, everything except the chassis block, motor, pickups, and wheels, at  which point I'll address how to mount 3100's body and the new cylinders and motion.  The balance weights on the wheels are a bit less subtle than on a 31xx, but given that this 'representational layout model' has numerous other departures from fidelity, I'll live with it. 

 

According to some of the reviews of the Baccy Ivatt 4MT, it is a (flying, or something else starting with f) pig to take apart, and I will try my best not to cause any irreversable damage so that it can be re-assembled as an Ivatt 4MT in future should that be required.

[The Johnster]

I've now dismantled the Ivatt, not a difficult job but there is a bit more involved than I am used to with GW prototypes.  A methodical approach pays dividends.

 

I am now assessing the work needed to fit 3100's body to the running chassis, and it looks as if it should fit, but there will be very little room between the top of the motor and 3100's firebox, and some plastic may have to be removed from the inside of the firebox.  The cast Ivatt chassis includes the cab floor and front end frame and buffer beam detail, so these will have to be cut off.  There is a lot of Ivatt detail included in this casting, and I am hoping most of it will be hidden by 3100's lower running plate and tanks.  The 'DCC ready' chip has been removed along with the cast metal platform that it sits on, which in turn sits on top of the crosspiece of the cylinder block, all of which has been taken off and put in the bag with the other bits.

 

Plan of attack, subject to arbitrary alteration to circumvent problems, is to cut the cab floor and front end off the Ivatt chassis block and fit the 3100 body to it.  This may well involve made up mounting points, blocks, and plates, not to mention a degree of out and out bodgery; those of a sensitive modelling nature will be requested to look away until it's all finished...  Once that has been satisfactorily achieved, the next job will be to mount the GW cylinder block, and it is not yet decided if this will the Hornby Grange set I originally bought for the project or Comet fold ups.  I will probably use Comet connecting and coupling rods.  I may have to revisit the GW front end, as I made the mistake of assuming that the proportions of the large prairie front are the same as those of a 43xx, and they're not, the level section between the drop and the buffer beam is smaller.  The aim will be to make it look as right as possible on the Ivatt chassis in relation to the pony wheel.

 

I've just noticed that the Ivatt wheels do not have enough spokes to represent 3100's, and I am not going to lose any sleep over this, either.

[The Johnster]

Sidetracked a little by the latest thing in my railway life, the colliery project (see layout thread. 'South Wales Valleys in the 1950s'), but I'll probably get as far as measuring and marking out where to cut the Ivatt chassis block early next week.  Basically, the cab floor has to be either removed or about 2mm taken off each side of it to clear into the prairie body, and the front end has to be removed ahead of the cylinder locating cutouts.  There will probably follow a confused period of fitting and faffling to get the prairie body to sit level and in the correct position vis a vis the wheels, and this will be determined by the position of the cylinders, which have to be aligned with the steam pipes and the chimney. 

 

This means that the body must fit to the chassis before the cylinders can be correctly mounted, but so far as I can see, the Ivatt cylinder position will be ok.

[Halvarras]

On 18/11/2021 at 14:47, The Johnster said:

I will try my best not to cause any irreversable damage so that it can be re-assembled as an Ivatt 4MT in future should that be required.

 

This is now sounding a little unlikely TJ! But it was only a secondary consideration, wasn't it?

I admire your perseverance - I thought I had oodles of patience but this summer I was at times tempted to see if one or two uncooperative projects had the capability of flight from an upstairs window! But, like most modellers trying to make silk purses I reckon, we have visions in our mind's eye of the finished product and steadily work towards it (the financial investment often means we have no choice but to see it through, as here). Some paths are relatively smooth, and others........

But we do this for fun.......don't we?!

[The Johnster]

Yes.  That’s right.  Fun.  What fun…

 

Reassembly of the Ivatt will not of course be possible following the mutilation of the chassis block.  This is probably not a sensible way to build a 31xx, and it has lost the benefit of being cheap, but as long as I am reasonably confident that the result will be a locomotive that a) looks reasonably like it could be a 31xx one day and b) works, I’ll plug on regardless.  
 

The danger of this approach is that I will end up in a hole but fail to stop digging.  Tjey way to avoid this is, I think, to accept that what has happened has happened and the money has been spent, but that I now have to adopt the approach that I’m not going to spend any more on this massively overbudget project except for minor purchases.  The new colliery project will absorb most of my railway money for the next few months at least; I reckon about £50 on track and turnouts, I’ve already blown £30 on a pithead, and I’m thinking about another £55 & postage for a Walther’s Keystone ‘New River’ colliery kit to adapt, so budget limitations for 3100 will be enforced by neccessity anyway. 
 

It’s not perseverance so much as really wanting a 31xx and Tondu had one…

[Halvarras]

It was my overriding desire (also known as sheer desperation) to have a model of a Brush Type 4, having begun to see real ones at my local station, that led me to do unspeakable things to a used Tri-ang Brush Type 2! This was back in late 1968, I was 15 with many years of plastic kit building experience but I had never attempted anything like this before. But I learned a great deal from it - I had never used plasticard before (all I had was one sheet of black 20thou) and had to invest in a few tools to tackle the butchery. I would probably never had attempted it at all, had I not been handed a copy of Railway Modeller, January 1966, in which somebody had scratchbuilt a Brush Type 4 out of plasticard and had conveniently included 4mm scale drawings, so I could have a reasonable stab at the roof detail (suitably compressed of course).

The RM job btw was outstanding for its time, built onto a stretched Dublo Deltic chassis with bogie wheelbases also extended by drilling new outer axle holes. It was finished as D1722 - I wonder what happened to it....?

PS I still have my conversion, to remind myself how ingenious I was back then!

[The Johnster]

Kudos for having a go, Halvarras.  At about the same time, and a year older than you, I was committing unspeakable atrocities based on the premise that as all GW locos were built of standard Churchwardian components and that old (even then) Triang Black Princess running chassis would 'do' for some of them.  There was a 43xx made out of the BP mech, an Airfix kit CoT boiler, and Airfix 61xx bits, motor hanging off the back of the footplate, a BP based 61xx, a Jinty (actually the awful 08, so no great loss) based 56xx from 61xx bits and plasticard tank sides, and a 42xx from a HD 8F and similar overpinnings to the 56xx.  Then I had a go at a Urie S15 from an Airfix Schools with a plasticard cab and a tender built up on a couple of Rovex Triang shorty coach bogies, another pair of which went under a Booth Rodley 35ton Airfix crane as an attempt to model the one at Radyr PAD.  I revisited this idea some years later using a Triang Hornby FS tender chassis, which fitted perfectly.  And a cut down BP Jubilee inspired by an article in Constructor describing how to convert a Black Princess into a Black 5.

 

Motors stuck into cabs, proportions were completely off, Airfix motion struggled but almost worked after I read in Constructor that pencil graphite could be used as a lubricant, the GW tanks had hopelessly misshaped roofs, my attempts to Swindonise outside cylinders were a disaster,and none of it was my proudest modelling moment.  I cobbled together a reasonable 82xxx from a Triang body and Airfix kit 76xxx underbits, but this was a non-running shed lurker. 

 

This sort of slapstick taught me a lot about modelling, mostly not to attempt such abominations, and has certainly 'informed' the 3100 project.  It conditioned me to a lifetime of wondering what RTR chassis can be utilised under different RTR, kit, or scratchbuilt bodies, and it is probably just as well that Cwmdimbath doesn't need a GW rebuilt TVR 04, or I'd likely have a go at one from a Lima or Baccy 45xx/4575 body and an Oxford N17 mech.