The Bodgerigar attempts to build a Collett 1938 31xx

[The Johnster]

Ok, bit of proper progress to report; I have a loco that looks as if I could tidy it up to become a passable 31xx, albeit best passed at speed from a distance (!), and that runs!  There is, as is obvious from the photos, a lot still to do to both chassis and body, but the basic principle that one can cobble together an acceptable 'layout' Collett 31xx in this way is, I submit, proven.

 

 

Photo 1; Airfix chassis with new 20mm Markits wheels and axles, Hornby 42xx front pony, Airfix solid backed rear pony (actually a radial truck IIRC on the real locos), both temporary place holders, and Airfix cylinders and motion for first test run.  Successful; smooth, controlled mech with the original Airfix keeper plate clearing any obstructions beneath, though by a bit less than it was originally designed to and I would not want to chance it on any but the gentlest of gradient change vertical curve at the top of  an incline!  I had to cut a part of the rear of the keeper plate off to allow sufficient vertical play for the rear pony truck, but all is well.  Chassis/Mech is a runner and can be regarded as job done.

 

 

Photo 2; the body assembly sitting loosely on the chassis.  No problems but the body sits too high at the buffers (by which I mean where the buffers will be when I fit them) because it springs upwards.

 

 

Photo 3; body held in proper postition by finger of god, well finger of Johnster at least.  The proportions of a Collett 31xx are there, though there is a considerable list of stilltoodoos.  The gap between the running plate and the top of the cylinders shows up badly in the photo, and a good bit of matt varnish in this area will subdue the appearance a little. Wheels; axlenut covers, balance weights, and paint for rims.  Body; filling and making good with Milliput, buffers, coupling hooks, new chimney, cut down safety valve cover, restore tank top detail, refit steam pipes, lubricators in from to motion brackets, buffer beam re-inforcing bars, handrails, new handrails to replace Airfix moulded cab and bunker rear, steps for lh of bunker, glazing, coal, and probably half a dozen other things I've forgotten.  Oh, yes, 3100 number plates and transfers, and a crew to hide the lack of cab detail, plus any cab detail I can shoehorn in from a CoT plastic kit backhead.

 

So, it'll be a while yet before she's in service, but I'm not in a screaming frenzy about it; she'll be finished when she's finished.  The real 3100 was the established Porthcawl 'residential' loco, and did not run up into the valley termini very often, so she will be a nice occasional change from 4144 and 4145 rather than an essential component of Cwmdimbath's working day.  I remember her in reality on the Porthcawl, sitting at the buffers in Cardiff General's long gone down side bay Platform 5.  I like large prairies and this one, the most powerful and ultimate design, is my favourite.

 

[The Johnster]

Looks a bit like a 4575 on bodybuilding steroids, doesn't she?

[The Johnster]

Update sitrep; 3100 now has buffers, the safety valve cover has been cut down (top cut off and reattached, the slitting disc gap providing the lower height), buffer beam reinforing bars replaced, a relief as the glue joint for the new front running plate needs all the help it can get until the body is supported by the front of the chassis block, and the chimney has been cut off ready for replacement.  Minor stuff and nothing exciting, but moving in the right direction. 

[Nick Gough]

6 hours ago, The Johnster said:

the safety valve cover has been cut down (top cut off and reattached, the slitting disc gap providing the lower height), 

You must have a steady hand to do that!

 

I would would probably have removed the topfeed instead!

[The Johnster]

It was the presence of the top feed that prompted me to do it this way, and it’s a classic bodge.  I may yet replace the cover! The 43xx boiler has a tall top feed which is wrong for 3100, so I slitting disc cut it below the lip, separating the top.  The top piece was fairly neat but the stump needed a bit of fettling to get it level.  
 

The result is of course a wrongly proportioned safety valve cover an a ugly glue join below the lip; the latter problem can be addressed with Milliput (which is going to be a major component of this loco!) but it remains to be seen if the proportion problem can be lived with on this ‘layout’ loco, or if it’s shortcomings bother me by being obvious enough to go for a replacement. 
 

It might be that a better choice would have been a 42xx body, but that would have had to have been cut between the tanks and the running plate, losing the rivet detail.  

 

[The Johnster]

Started with the Milliput, filling gaps and making good.  This will be a process of filling, shaping, waiting for the 'put to dry out and go off, and sanding down, then doing it again until I'm happy with the look; it'll be done when it's done.  I'll plug on with replacing details and looking at the handrails and such while all this is going on, a process of 'worrying' the job into submission.  There are things to buy, new handrail knobs and the wheel furniture (balance weights and axlenut covers), then the number plates, and as I'm nobbut a poor pensioner it may be a few weeks until all the shopping is done.  Oh, and a 56xx chimney has been put on, with Milliput shaped to represent the base.

[The Johnster]

Bit of a drawback yesterday; fitting the (now sprayed black) body to the chassis and trial running to ensure clearances, loco derails in a shower of short circuit sparks.  Turns out that one of the Markits tyres has come off the wheel!  Most of the evening consisted of stripping down the wheels and motion, supergluing the tyre back on, and re-assmbling.  Haven't test run the chassis again; it's time for bed now, but probably will do so over the weekend.  I'm basically in the same place I was a week ago, disheartening but not disastrous...

 

Sprayed body looks pretty good, though!

[The Johnster]

Been having a rest from the project with some operating sessions, but have today ordered balance weights and axle nut covers, which will complete the running chassis, which can then be dismantled for painting and final run testing.  Focus will then shift to replacing or renewing body detail; handrails (including new replacements for moulded and not included by Airfix or Mainline), replacement of broken front steps, new whistles, coal for the bunker, and anything else I can think of!
 

3D number plate transfers, which include the crests, smokebox number plate and shedcode, and RA dot/power class, are also on order from Railtec.  I reckon she’ll be in service by December…

[The Johnster]

On 21/08/2021 at 21:23, The Johnster said:

Plan is vague and subject to arbitrary alteration for no apparent reason, but will go something like:-

 

. Remove motion from Airfix chassis for re-use later with new 5'3" diameter driving wheels.

 

. Remove Airfix wheels and dispose.  Easy enough so far...

 

. Fit replacement driving wheels, and fettle and test run chassis with original motor and pickups.

 

. Fit Grange cylinder block and test run with Airfix motion re-fitted.  I am expecting trouble here.

 

. Cut the boiler and cab sides out of the Airfix 61xx body.

 

, Cut the running plate valance horizontally beneath the running plate at the front drop, and through the cab steps to the rear drop, and fettle until the body sits level at the right height.  The cylinders will now be sitting lower because of the smaller driving wheels, so some packing will be needed to get the buffer beams to the right height, which will leave an incorrect gap between the bottom of the running plate and the top of the cylinder block, but I can't see a way around this at the moment and will have to live with it.  Comments will be referred to my Hornby 2721...

 

. Test fit to chassis to establish clearances, level, etc., then remove for futher work.

 

. Fit new no.4 boiler.  Should be ok once there is sufficient clearance for it between the tanks, but it will need a lot of surgery, I mean brutality, to its lower portions

 

. Fabricate and fit front and rear cab spectacle plates, and fit cab detail if possible (rear of Airfix chassis block may need to be cut out for this.

 

. Fabricate new cab roof, possible 42xx part.

 

. Probably needs 42xx chimney.

 

. Fit body to chassis, test run, paint (1951 livery), details, number plates, into service.

 

Visited the first post for a recap; we are nearly 6 weeks in to the project, and progress has been good, but some departure from the original plan, which I did say was subject to arbitrary alteration for good reason...

 

First 3 stages went as planned, fourth was abandoned and the old Airfix cylinder block and slide bars re-used.  This may turn out to be a poor choice, as previous ones have broken at the slide bars, but the Grange block came as just the block with no slide bars.  It has been put aside in the project box with the idea of using it with Comet slide bars and motion at some future date.  Stages 5 to 8 proceeded as expected, but by the time it came to fabricating cab front and rear sheets, it had occurred to me that there was a pretty close similarity in profile between the cab front/rear/cab roof assembly between 31xx and 56xx.  The relevant piece was removed from an old Mainline 56xx body and dropped perfectly into position as if it had been designed for the job, and I had the basic body assembly.

 

The expected gap between the running plate bottom and the top of the cylinder block turned out to be nowhere near as chasmatic as I'd feared, and is 'liveable with'.  I experienced a good bit of faff in securely attaching the new dropped section of running plate with the buffer beam ahead of the smokebox, and was assisted by replacing the buffer beam re-inforcing bars; 3100 should now be able to survive a front end collision of medium severity, which the prototype apparently failed to do after a heavy impact on the buffer stops at Porthcawl in 1957, presumably after an over-enthusiastic approach...  Not that I intend to bang the loco about like that, but sh*t happens. 

 

Don't know why I thought a tapered 42xx chimney with a cast iron top would be ok, and have given it the chimney off the Mainline 56xx, which is a little too wide in diameter at the base.  A new base has been shaped up out of Milliput, and looks ok after a rattlecan coat of Halfords eggshell acrylic black has hidden a mutlitude of sins.  I used Markits 20mm wheels, a scale 3" diameter too small, because Wizard were out of stock of 21mm and recommend 20mm for this loco, and I am not entirely happy that the proportions of the loco are right; she looks a little too much like a fat 4575, so I may replace them with the proper 21mm some time.  OTOH, once painted they may be less obvious and I'll be able to live with them.

 

I've well exceeded the hoped for £50 price tag, into the low 70s as of this morning. 

[The Johnster]

Modelling mojo a bit low, so not much progress to report, but I did manage to pick the job up and look at it last night, resulting in handrails, whistles, and tank ventilators appearing and a second coat of Halfords rattlecan acrylic matt black covering the blemishes.  The transfers are on the way from Railtec, and will be applied as soon as they get here; 3100 will be in her final livery, plain black with medium size unicycling lion.  A coat of matt varnish will seal them, and some coal for the bunker, and the main handrail around the smokebox and front of the boiler remain to be done.

 

Progress on the chassis is less noticeable, and I am having a bit of a rethink about the wheels, which should of course be 21mm not 20mm; went for the smaller size because I couldn't source 21mm Markits at the time but have now found some.  Finishing the body should make me focus more on the underpinnings, as I am starting to see light at the end of the tunnel of finishing this project.  By finishing, I mean bringing the loco to a state in which she is complete in the essentials and good enough for service; there will always be improvements possible! 

 

I'll take some body shots when the transfers are on...

[The Johnster]

 

Transfers are on, so the loco now has an indentity, but I've just realised that the 43xx dropped running plate ahead of the smokebox is too long for a tank loco, and a bit of a rethink of this area is needed.  The position of the RA/power class spot is odd, but photos of 3100 show it to be correct.  Plenty still to do on the body; smokebox handrail, detail on running plate ahead of tanks (lubricators?) and the boiler support at the front edge of the tanks.  The handrails are a big improvement to the old Airfix 61xx body!

[97xx]

Really interesting build!

 

Photos I have seen of 3100 show it with the 'large' early crest?

[The Johnster]

Yes, the crest is too small and is a placeholder until I replenish my supply of transfers.  I mistakenly ordered the wrong size crests from Railtec, my bad.  Google has a photo of the loco at Abergwynfi in 1951 in the June 1948-August 1949 'BRITISH RAILWAYS' Gill Sans livery, though this had been replaced by the large unicycling lion by the time of a May 1955 photo in the Hodge/Davies book taken at Porthcawl, which portrays the loco in a fairly dirty condition suggesting that it had been in service for a while since it's last overhaul.  The loco was involved in a 'heavy shunt' with the stop blocks at Porthcawl in 1957, and apparently damaged the frames at the front end, being sent to Swindon shortly afterwards for assessment; the decision was taken to withdraw the loco but there are several photos of it on the recep roads at Swindon as late as 1960.  There is no visible damage at the front end in these photos.

 

My layout, Cwmdimbath, is set to a nominal time frame of 1948-58, and 3100 was allox Tondu in 1946 for the restoration of the daily Porthcawl-Cardiff 'residential' commuter service that had been a Bulldog job before the war.  It was presumably overhauled and repainted between June '48 and September '49, and overhauled and repainted again between the beginning of September '49 and a bit before May 1955.  It would be interesting to find out exactly when, but either livery would be suitable for my purposes.

 

Another mistake is the positioning of the route availability/power classification spot; I took it from a Swindon recep road shot which shows what looks like this feature above the bunker handrail where I have put it in this photo, but it is clearly below this handrail and between it and the number plate in the 1955 Hodge/Davies book photo.  The number plate should be set slightly lower to allow room for it and coincides with the horizontal centre line of the unicycling lion.  These are matters that will be rectified before the loco goes into service.

[97xx]

I've seen a photo of 8102 with the RA spot just slightly above handrail level but between h/r and cab opening, and 6125 even more oddly placed - at the top corner of the number plate.

 

Not sure we're referring to same photo, but in one I can see dated 27Jul58 of 3101, 3100 and 3102 inside Swindon shed, the RA spot on 3100 is pretty surely where you have it, above the handrail, centre to the plate?

 

Other details for 31xx I'm aware of are front lamp iron on smokebox door, and the 'three steps' (with grabirons) on fireman's side of bunker post 1954.

 

Edited October 11, 2021 by 97xx

[The Johnster]

Don't think I've seen the 3x 31xx in Swindon Shed in 1958 photo.  I'm aware of the smokebox door mounted lamp iron on 3100, which the loco seems to have had from new, a feature shared with another Tondu loco, 5797, which I've modelled by converting a Bachmann as it was a 'no top feed' example during my period.  Thank you for dating the fireman's bunker steps to post 1954, as this is a possible clue to the repaining in large unicycling lion livery.  It suggests that the BRITISH RAILWAYS Gill Sans would be a better choice for me, as it would mean that the loco carried this for 5 or 6 years post 1948/9 as opposed to 3 years in service before the Porthcawl impact incident, 1954-7.

 

According to the Hodge/Lewis account, 3100 was used for a short time as the Ogmore Jc. yard pilot while the damage was assessed locally, and then sent to Swindon for a fuller examination.  As the loco was 'only 19 years old', it was expected that a full overhaul would follow but Swindon decided to withdraw it, and Hodge/Lewis suggest that this may have been accounting error resulting from an assumption that the loco was actually a 3150.  I'm not sure about this; it may have been the belief at Tondu, but the loco was actually a rebuild of a 3150 and not really 'new' in 1938, and the frames were the original Churchwards, and getting on a bit.  All five of the Collett 31xx were withdrawn within 3 years, as well as a good number of 3150s, though some of the latter survived into the early 60s. 

 

A lot of photos of 3100 were taken by enthusiasts of visits to Swindon Works while the Tondu branches were less photogenic, including Porthcawl, though Hodge/Lewis has photos of the loco at Porthcawl, broadsied on and in full late afternoon May sunlight presumably after working the evening return leg of the 'residential', and on shed at Tondu, as well the previously mentioned Abergwynfi shot which is useful as it is bunker first, and proves that the loco was not exclusively used on the 'residential'.

 

Modelling 3100 in the Gill Sans livery would enable me to 'get away' with not providing firemans' bunker steps as well!

[97xx]

Whilst at Tondu, "The former... (3100) ...frequently worked the fast morning business train from Porthcawl to Cardiff"

 

Another snippet, although too early for your modelling, "some confusion has been caused by a first official photograph with the group letter E painted on in error".

[Weaselfish]

On 11/10/2021 at 15:57, The Johnster said:

According to the Hodge/Lewis account, 3100 was used for a short time as the Ogmore Jc. yard pilot while the damage was assessed locally, and then sent to Swindon for a fuller examination.  As the loco was 'only 19 years old', it was expected that a full overhaul would follow but Swindon decided to withdraw it, and Hodge/Lewis suggest that this may have been accounting error resulting from an assumption that the loco was actually a 3150.  I'm not sure about this; it may have been the belief at Tondu, but the loco was actually a rebuild of a 3150 and not really 'new' in 1938, and the frames were the original Churchwards, and getting on a bit.

I wonder if the relatively early withdrawal dates for the 31xxs was to do with them having higher pressure versions of the standard 4 boiler, 225lb psi rather than the 200lb psi of the 3150s, 43xxs, 42xxs etc. If the 31xxs were becoming due for work on their non-standard boilers perhaps it was decided it wasn't worth the cost. There were plenty of other large prairies in the 5101, 61xx, 41xx and 81xx series available, particularly with DMUs taking over more of the local passenger traffic.

[The Johnster]

Modelling mojo took a few weeks off but I had a look at 3100's chassis rebuild last evening, and have started tarting up the wheels with balance weights.  Unfortunately, I bought the wrong size axle/crank covers and there will now be another delay while I order new ones.  The body now has 'BRITISH RAILWAYS' Gill Sans June 1948-September 1949 lettering as per the 1951 Abergwynfi photograph on the left side, and a unicycling lion on the right, bit of a cheat but you don't see both sides at once...  I've also soldered new wiring in to a keeper plate with the pickups and clear plastic backing in better condition ready to solder to the motor when the chassis is re-assembled; I think this may have been the root cause of the earlier shorting!

[The Johnster]

Still having problems with shorting out between the wheel rims and the chassis, so the chassis is dismantled again while I think of a solution.  The Markits wheels are insulated at the join between the wheel and the rim, and the fact that they are smaller would, one might have thought, meant that they were less prone to short to the mazak chassis, but they are anyway; a rethink is needed.  An answer might be to insulate the wheels from the chassis by shielding the chassis sides with sellotape, then hiding it by painting it matt black/rail colour, and I think I might be needing new pickups.  It's a bit frustrating at the moment but I'll plug on after a few days of considering the problem.

Edited October 28, 2021 by The Johnster

[The Johnster]

Eliminating short by process of, um, elimination.  Chassis now dismantled to component parts and short provisionally allocated to motor wiring, but the failed test runs produced a firework display between the rear lh driving wheel and the chassis block as well, so the evening's work has been to mark out clear plastic side pieces to overlay on the sides of the block to prevent any chances of this.  These will be painted matt black when they are glued in position.  The wheelsets have been individually tested out of the block and the insulation is ok. 

[The Johnster]

Taking things very slowly now and checking for shorts at each stage.  The new plastic insulating plates for the sides of the chassis block have been cut to shape to clear the bottom of the motor and superglued in place, and cutouts provided to match the axle channels.  I have given them a coat of matt acrylic varnish prior to painting them black later today.  I will then 'try fit' the wheelsets and test again for shorts, and check the solder joints between the keeper plate eyelets where the feed wires are attached to the pickup strip. 

 

The insulation plates will slightly reduce the sideplay of the driving wheels, but I am not expecting this to be a problem.  If it is, it will manifest itself on the 2nd radius part of a 3rd/2nd curved turnout in the fy.

 

Then we'll fit the keeper plate and wire up to the motor, which should then run when I place the loco on the track and apply power, by which I mean that the motor should turn and drive the leading axle.  If it doesn't I'll have a much better idea and a smaller search area to locate the short.  I will be able to check for polarity as well assuming it runs; this could be done and dusted by late this evening if all goes well!

[The Johnster]

And, back to square one.  Ok, here's what happened.  I inserted the rear wheel set and sat the whole thing on the track, and turned up the volume, I mean voltage; no short.  Waggled it about a bit, still no short, so, the new insulating plates are working.  Next job, offer up the keeper plate and test for short; no short.  Next job, solder the feed wires from keeper plate to motor connections, sit the loco on the track with just the rear wheelset, and crank up the volts.  Motor runs perfectly, no short, o frabjous day, calloo callay.

 

Right, now we're making a bit of progress.  Let's try putting the front wheelset in and see what happens and if I need to reverse the connections for correct polarity.  Shorts out.  B*gger.  Now, this wheelset does not short on it's own, and makes no difference to the way the loco is set up and wired when it is inserted, so I'm scratching my head a bit.  Ok, remove it for investigation and test for short with the loco in exactly the same state as the successful motor run earlier.  Short.

 

One step forwards two steps backwards, we are making reverse progress!

 

Rethinking the project and looking for possible alternative measures.  Early days but:-

 

1). Build new chassis, Southeastern or Comet.  Pros; choice of motor and gears. cons; cost, a serious issue for a poor pensioner.

 

2). Use alternative RTR chassis but replace the wheels with the Markits I already have.  Obvious one is the current Hornby large prairie, but this is almost as expensive as option one.  If Dapol do their 43xx chassis as a separate item, that might be worth investigating.

 

3). What else has 20/21mm wheels, the same axle spacing, and preferably GW fishbelly coupling rods?  A Mainline/Bachmann 2251, that's what!  3100's body and running plate should fit over it, some sort of mounting block for the cylinders would have to be fabricated (or they can be mounted to the underside of the running plate), and holes drilled to take the leading and trailing pony frame pivot bolts, which would compromise the split chassis setup.  And it's a Mainline split chassis, with all the potential problems that entails.  Anything else?  Well, actually, yes, but not with the correct wheels, the Bachmann 57xx/8750 lines up perfectly.  We are back to replacing the wheels with the Markits, though.  I looked at the 56xx, which would have the rear radial pony already supplied, but the axle spacing is a bit different on this loco.

 

4). Build a new chassis block out of electrically dead material, i.e. plastic, possibly using components from the Kitmaster plastic kit prairie with the axle holes bushed, or alternatively as a scratch build to fit the Airfix motor and gears.

 

Ok, let's regroup.  I'll have another go tomorrow but if I can't make progress I'm not going to persist in banging my head on the project.  I have a spare 57xx chassis block, so plan B will be to see if anything can be done with that and the Markits, but as things are it looks as if I will abandon the idea for now.  Put it back in a the project box and maybe in a few years the price of 2h Hornby 5101s or Dapol 43xx will have come down enough to make adapting one of those chassis worthwhile, or I could go the Southeastern/Comet route, already have the wheels and perhaps can use the Airfix motor and gears.  The cost can be spread to make it affordable, but both these plans are playing the long game...

 

Next step, though, is to see if I can get the spare 57xx Baccy chassis block to work with the Markits wheels.  If this can be done, it is worth plugging on bit by bit, perhaps as a hybrid with Comet cylinders, slide bars, and motion.

[The Johnster]

Number 3 won't fly, because althought they appear identical, the Markits axles are thicker than the Bachmann ones, so the Markits wheelsets will not fit in the axle grooves in the Bachmann chassis, or presumably a Hornby or Dapol one either.  Perhaps some unholy union involving a Kitmaster 61xx chassis, but I'm starting to sound desperate even to myself now...

[Halvarras]

At the top of Option 3 you mention "Mainline/Bachmann 2251" - true, the original Mainline model had the infamous split chassis, but the later Bachmann 'Blue Riband' version received an upgraded conventional chassis (with motor mounted vertically in the firebox I believe). However, are you saying that you have a spare Mainline one to hand and therefore free, whereas obtaining a Baccy one would involve the reluctant handing over of readies?!

[The Johnster]

Thanks, Halvarras, I'd forgotten the Bachmann 2251.

 

I don't have a Mainline 2251 chassis to hand, and a Bachmann one would be a better idea all round as a basis from which to build up a 31xx mechanism.  I don't have a Bachmann 2251 chassis either.  I am reluctant to hand over readies because I am on a pretty tight budget, and 3100, while an interesting prototype, is a long way from being absolutely essential to the Cwmdimbath scene, so I can't really justify spending the sort of money needed to finish a Comet or Southeastern chassis. 

 

I've just been browsing the Comet large prairie chassis pack, £37.50, which is borderline justifiable.  I've already got a motor and gears, from the original Airfix 61xx, and the wheels and axles, so if I could build a running chassis using the Comet pack with the old Airfix motor and gears, there would be no additional expense, and this is looking like the least unlikely way of continuing the project at the moment.  But I am by no means certain that it would be possible to mount the Airfix motor and gears in the Comet frames. 

 

Clearly the best way to proceed would be with a setup involving a motor/gearbox/driving axle assembly, which would sit neatly between the Comet frames and is what they are presumably designed to be fitted with, but the cost ramps up quite quickly when you start considering this sort of thing.  Worth thinking about if I can find a cheap, perhaps secondhand, such assembly; messing around with attempting to mesh gears without a gearbox and trying to make the motor fit at the right angle and position to enable this sounds like a hiding to nothing to me.  If I can't find something that is already built to put an axle in, it isn't worth the bother!

 

I will investigate, by which I mean browse eBay, sourcing a Bachmann Blue Riband 2251 as well as motor/gearbox/axle assemblies.