Wills T9 Repair and Upgrade in 4mm 00 Gauge

[Stevers]

This is a Wills T9 built by a friend's father using a large tinman's soldering iron and gas ring using a 'low melt' solder that doesn't respond to boiling water. I am assured that the father wasn't technical, but it is really quite neatly joined together using Rose's Metal or something with similar properties. I use a 75W Weller on a lighting dimmer switch, and it does respond to a turned up soldering iron, but the melted solder doesn't flow well so really needs removing for any repairs.


Found on the floor behind a book case it has clearly suffered at the school of hard knocks. Not all the bits were fully attached, but all parts are present. At the rear there was significant impact damage on one corner, and the fit of the back of the cab floor and drawbeam left much to be desired making the model width a trim 32-31-33 from front to its out of square rear. The SEF T9 chassis are out of stock at Squires, but my friend remembered that I happened to have a SEF chassis that wouldn't fit my Hornby T9 (wrong wheelbase). This chassis kit was part built in EM with twin beam compensation and was complete except for the lovely bogie which I was able to use. I ended up using a Perseverance chassis for my Hornby T9 conversion and that left me with the Persy bogie left over. Initially I offered to convert the SEF chassis and build the Persy bogie to 00 gauge, but this has escalated to me attempting the whole repair and upgrade. Also supplied is a Hornby six wheel tender in bare black plastic.

The completed model is to be painted in Southern wartime livery, numbered as 311 and is required by Christmas - no pressure then!

 

[AJC]

That looks like a *challenge*, Steve! Presumably the rust is down to split pin handrail knobs?

I'll look forward to seeing how it turns out (and at least the paint job is simple).

Adam

 

[Stevers]

Adam, absolutely right, steel split pins, pushed in to the very hilt, probably reacting to whatever hot glue and vicious flux was used to melt it together originally - can't rule out damp being a factor as well.

The battered corpse arrived a week ago and things have moved on somewhat on the rejuvenation front. First I removed most of the paint using Modelstrip, as it was spray primer it had successfully got everywhere so some remains in hard to reach places. I needed to disassemble the rear to straighten the sides and correct the out of squareness. Boiling water didn't touch it, so mechanical removal was attempted. The damaged roof was not fully attached, but bits of it were attached well enough for them to be left behind, so a replacement roof has been sourced from my own Wills kit to be replaced when SEF/Squires are up to speed. Smarting with that not too consequential failure, I decided to try to unsolder the rear assembly that sort of locks together, as before it wasn't fully attached and my trusty 75W Weller and dimmer combination set higher than for 70° soon had the measure of it, and with some difficulty I was able to separate the three parts involved. These parts were too wide as cast pushing the rear apart by about 1mm contributing to the noticeable bend on one side. So half a millimetre was removed from the centres of the splasher mouldings and the drawbeam adjusted to fit the reduced splashers. The sides were straightened so that the rear matched the front although the waist (so to speak) remains irredeemably at 31mm. I also modded the locating tongues so that the rear assembly could be fitted square to the sides. This was all quite mysterious as the sides were the same length and the buffer beam square. Anyway square to both sides the rear had to be. As three parts it could be coaxed into place between the sides and then soldered together.




The gap filling properties of Low Melt were utilised to the full in this process. With something to locate the chassis in the body the EM chassis was rebuilt using what are alleged to be the 00 spacers. I had built it compensated for EM using the SEF beams with bearings, and this could not be undone, so this 00 gauge engine will be compensated as it will be far less trouble than trying to fix the beams in the correct place. The simple Perseverance bogie is etched brass, and is intended to be built compensated even if used on a 'fixed' chassis, the complete opposite of the fancy SEF bogie which can only be built fixed. I added an extra strip on the sides to make them a closer approximation to the prototype's equalising beams.

So we have the SEF chassis using all SEF parts built as per instructions in the Wills body it was designed for. So it's bound to have the correct ride height - yes?


Well no! This is the Hornby tender that has been allocated to the build, it comes with pickups and DCC decoder plug, but rolls really badly with skiddy wheels, possibly because it has no weight in it. It's a really rubbish picture and in fact the difference was less on the other side because the chassis wasn't quite level in the body - I still don't know why this was, but can't go back to reproduce it. So actually it's not far out probably a bit less than a millimetre overall. Adjusting the ride height was not entirely straightforward, and I'll cover that in my next post.

 

[Stevers]

First job was to add a curved top to the bogie (cut from some square section brass) to spread the considerable weight of the front of the loco evenly between the wheels, two rubbing strips are needed to keep the bogie level since both sides rock. This also set the final height of the bogie for what came next.

Adjusting the ride height on the SEF chassis was easy enough at the rear where the notch in the frame could be widened so that the chassis sat higher. At the front was a tab that was more resistant to change. After much head scratching I elected to cut off most of the SEF tab leaving about 0.5mm to act as a locator for my own tab. I made this up in stout nickel silver as an L of correct width and with hole in the correct place. To set it at what I hoped would be the correct height I added a strip of .020" nickel silver between the remains of the old tab and my new tab to set it lower by about 1mm and soldered it in. I then packed the space between bogie and frames with some plastic shims and adjusted the rear notch and shims until the chassis was at the correct height with chassis level in the body, mercifully my new tab tuned out to be just right!


Two .040" shims of styrene or about 2mm seemed to about right to for the bogie packing, so a U section was made up 2mm deep 8mm wide and about 12.7mm wide to suit the width of the chassis, the ride height was checked with this in place. Cosmetic front frames were cut to match the shape of the Wills frame with 2mm added to conceal my spacer and the assembly soldered together. Lastly a clearance drilled nut was soldered over the hole in the spacer to stop it being squished by the bogie fastening screw. The bogie securing screw is bushed to allow free movement of the bogie.


The finished assembly. The SEF 00 chassis spacing makes things extremely tight, but if there is to be any cab floor at all it needs to be that way. It has meant that the flanges on the bushes of the equalising beams had to be thinned down to a wafer thin .005" before there was sufficient clearance for the Markits wheels to turn freely. The loco now at the same ride height as the Hornby tender with the drive axles just kissing the valance:

 

[Mike Garwood]

What gearbox and motor combination are you going to use?

Mike

 

[Stevers]

In the box of goodies that came with the battered body was a Roadrunner gearbox (45:1) and a 1320C coreless motor, both from High Level Kits. I'm driving the front axle and everything fits nicely in the firebox with hardly any intrusion into the boiler. With the demise of Mashima these motors are I suppose the future, and it will be interesting to see how well this combo performs. Based on my experience of other HLK products I'm expecting great things.

 

[AJC]

Oh, very posh (and powerful, going on the spec's on the High Level website). I have one of the iron core ones and a suitable gearbox awaiting me finishing a couple of projects and, from talking to Chris about it, he reckons that they're the equal of the next size up (or more) from the Mashimas they replace. The 1020FE, I note, he suggests for things like Collett Goods or Jintys (which he previously recommended 14 or 16 series Mashimas for).

The larger coreless motors are presumably a step or two up again (if a bit pricey).

Adam

 

[Stevers]

Quite, I popped down to the workshop to measure the motor to confirm which one had been supplied, and it actually feels powerful without electrickery anywhere near it. I'm looking forward to trying this little drive unit out!

 

[Tim Hale]

Almost the same arrangement, HLK 1320 but using a Roadrunner + in a Bec D15 which has a tad more headroom.

​

 

[Stevers]

A lot of time spent - particularly on the buffer beam, and a little more progress. Quite early on I drew up the rivet pattern for the wide splasher T9 (this drawing should be attached as a pdf file), and used my GW Models rivet press to create a buffer beam.
The rivets aren't at all obvious on the smokebox, but they do show on the buffer beam. These were made using the 2mm scale anvil and are still a little overscale. I find it essential to pencil in the wheel settings (to 0.1mm) for each latitude and longitude so that I can keep track of where I am and need to go next as I crisscross round a workpiece as complicated as this. The slot for the hook was cut too wide, but I'll be soldering the hook in anyway as there's a big chunk of footplate structure in the way of having it sprung. The buffer holes still need opening out and moved across slightly as they drifted when drilled.


The Markit's buffers were disassembled, and the housings soldered into place together with an overscale but functional Markit's hook. This assembly was then epoxied into the rebate that I spent far too long filing out.

The housing bolts are represented by holes, and only the deranged would give that a second thought, let alone spend even more time doing anything about it! Next we moved to the arse end and made a combined cab floor and splasher side from .010" nickel silver, For best fit I found this worked best made in three parts, I then trimmed the Caley Models backhead slightly to fit the completely none negotiable space between my new splasher sides.

Also visible above are the repaired rear steps and corner. Three of the steps needed repair which was done by either slotting or rebating little rectangles of tinned nickel silver into the Wills casting and soldering up with 100° solder for strength, this was a little fraught as the melting point seemed a bit close to some, but not all castings in the kit, particularly any sort of extremity. [Edit] I faithfully put the missing rear middle step where it was before, not noticing that the middle steps at the rear were at different heights - also true of my own kit. [/Edit]

Following my friends' dad the front footsteps were fitted flush to the valance where they were very poorly located. Only then did I look at a photo of the real thing and realise that they should be inset. Once unsoldered and cleaned up for refitting, they turned out to be just poorly located in their new positions and very difficult to get square. In securing plenty of low melt solder was used - not pretty. In washing down afterwards it seems our hot water may be a little hotter than 70° (a very bad thing anyway), this became apparent when one of the front steps started to move, requiring it to be re-visited with soldering iron. One of the original front steps was deemed beyond repair and was replaced from my own kit.

That is a very long driven wheelbase, yet 1mm short of where it should be. Why that should be is a mystery to me, as I don't believe the kit was ever intended to take a proprietary chassis.

 

[Stevers]

After a short break to prep and paint my railway era fancy bargeboards, I'm back. First up was building the RoadRunner gearbox which I found a tad fiddly, but as anticipated it really does run superbly. The coreless motor is almost silent and has plenty of ooomph. Continuing a theme it's tight in there and some bearing material was removed from the gearbox before it all fitted in the chassis and ran freely.


To make the gearbox work I needed to add an extra spacer at the rear to anchor the motor and attach the PCB for the wiring loom. To reinforce the chassis I used 1/16" brass tapped for the PCB securing screw. This was chosen because the cab floor is deliberately a tight fit between the frames, and we want it to stay that way. The brackets to mount the injectors were added these being pre-drilled to match the associated pipe positions on the body.
To fit the coupling rods I used standard Romford crankpins shortened to just clear the splashers with the rods held on with the barest sliver of purest green insulating sleeve. With the crankpin holes getting little more than a polish the chassis ran without binding.



Next job was pickups. The SEF kit comes with a nice little etch that represents the ashpan, but that covers up the two captive nuts intended to mount the pickup PCB. It was decided that my pickup PCB (actually some sort of fancy copper covered fibreglass) would need to be wedge shaped to match the SEF ashpan, so I made a couple of wedges to pack the PCB to the right shape. After mulling it over, then trying a few things I've fitted backscratcher pickups mounted from underneath - does this make them 'bottomcratchers'? Anyway there's nothing dangling down below, and just a corner of the pickups visible - very neat. In the absence of PB strip I've used brass, but will upgrade to PB next week after I've visited Signals in Midsomer Norton. I thought it preferable that the pickups would be accessible from underneath.

With continuity established across the pickups, they were wired into the 4-Way loom PCB. Then the same with the motor with one hard wire acting as the motor restraint. In the absence of the tender loom I soldered jumpers to connect motor to pickups. The Interweb informed me that the drivers side (right) should be red, so naturally I connected this to the positive terminal of the motor. So that's another of my engines that will need a quick sex change!


A couple of videos of it being controlled via my rather crude mini-drill controller - outputting some sort of chopped 12V.

On a very limited test track it seems very smooth, very quiet with a reasonable top speed. The eagle-eyed may spot that in a spare moment I've repaired that gap in the top edge of the running plate using 100° low melt solder.

 

[Stevers]

Loose ends tidied up and a quandary to end with.

First job was to dismantle everything and apply Loctite to the intermediate axle and worm of the motor. I then changed the sex of the engine to match my test track and South Junction, and that just involved me turning the motor through 180° and resoldering the connections. It seems the holey loom PCB couldn't take the torque from the motor support wire, so that was replaced with some of the fibreglass PCB copper clad that was used for the pickups - much stronger. With it all back together again the brake hangers were fitted needing me to reinforce the two fragile ears at the front of the chassis, and I added a captive nut at the rear for the tender drawbar. The tender will be coupled to the engine using the Hornby prong. As part of the drawbar work I did a bit of padding of the slot in the back of the chassis to made sure that the rear of the chassis was positively located in the body at the correct height. Looking at the photo I can see that the captive nut is pushing the chassis forward slightly. Also visible in ghostly green fibreglass is the sloping part of the ashpan/pickup module.



There is also now enough sideways movement in the bogie for it to squeeze round a second radius curve so I don't think the 24" minimum radius of my friends layout will frighten it. It doesn't have the Phosphor Bronze pickups yet, but I did tweak the brass pickups and give them a polish and if anything it's even smoother and more responsive both as a 'power pod' or loco.

Views of both sides showing the engine nice and level with the tender, well, oh so nearly. What isn't level, and really should be, is the base of the cab cutout with the top of the tender handrails. Somebody has got something wrong, and it's Messrs. Wills. The cutout looks a bit narrow when compared with narrow splasher T9s, but according to the GA drawings in Bradley, is correct for the wide splasher version, which has a narrow cutout more like an L12. I had intended to file away a section of the beading so that I could solder in a length of nickel silver strip with a hole to take the cab handrails. That will no longer be enough to assuage my residual OCD, so I made a template of what I thought the cutout ought to look like to see how much needed to change.

As can be seen there's quite a chunk to come off, but it does look like I could blend my new strip into the existing curve In addition on one side part of the top of the cutout is missing, so I need to do something clever up there too to reconstruct it. Given this is one of those extremities that are likely to shrink from 100° solder, tinning and 70° low melt will be the only option, although I could probably attach the handrails with 100° and get away with it. Process would be: Remove material using template, leaving step in bead. Put hole in strip. Form curve to put hole in correct location for handrail. Cut strip to exact length, and solder in. Though definitely part of the T9/wider LSWR look, and not really a question to be aired in the context of model railways - is it worth it and is this really what my near future consists of? :-/

 

[Stevers]

Well that went more or less to plan. Not in the original plan were the essential two pieces of card used to set the offset of the bead, one cut to the new shape about 0.3mm thick and the other thicker to solder against. Another job where an extra set of fingers would have been jolly handy!


That I'm going to count that as some sort of success. The handrails are tapered, if not perhaps exactly like the prototype which step impractically thin above the middle bracket. Hat tip to Martin Finney for pointing me in the direction of nickel tapered clock pins.
Nickel Taper Tapered Clock Pins Assortment (100) Ø0.30 to Ø1.50 (40mm) - CX126 | eBay
Quite a range of sizes included, the ones shown are I suspect the finest. The larger sizes in the pack were sold on to a couple of 7mm scale colleagues at YMRG, as being more suitable for Big-Big Trains. Am so tempted to replace those oh so puny Hornby handrails, but really must resist...

 

[Stevers]

The promised trip to Signals at Midsomer Norton yielded two strips of the finest phosphor bronze, some of which was used to replace the brass pickups. As designed the brake pull rods would end up on the outside of the wheels like on an M7 or 700, so some rework was needed to narrow the rear brake cross member. The front cross member is 0.45mm N/S in brass micro bore tube to beef it up a bit. I'll add micro bore tube sleeves on the rear cross member to space that better too. As can be seen there's a bit of a failure to plan ahead by SEF and the rear cross member is directly over the rear screw for the pickups. Still. SEF's real plan was to cover both attachment points with an etched ashpan! All this means that I'll need to make the brake rigging removable.


The injector feed pipe inlets(?) were drilled and springy phosphor bronze wire was epoxied in, This was difficult to form and took a couple of attempts with one of the rejects visible above. Holes were drilled in running plate and cab roof for those tapered cab handrails, fully penetrating the running plate so that the handrails can be fitted (or replaced if required) from underneath with the cab roof still attached. A pleasant late hour was spent milliputtting the body and fitting (not fixing) the dome and smokebox door. The capuchon on the chimney has since been removed and because the snifting valves were taken off when the BR number was applied, blanking plates will need to be applied to the smokebox.


The smoke box door had some fairly deep file marks across the top and these were filled with Milliput. This mix didn't set or adhere as well as usual, so may have needed more mixing. If the epoxy repair visible isn't satisfactory it will need redoing. The dog catches were then gently centre popped for the little handles, and drilled out with a 0.35mm drill. Iain Rice warns of the tribulations to be had in drilling whitemetal with tiny drill bits, but I've never experienced any such issues, and these drilled out easily. The little handles were then formed with 0.3mm N/S wire, and having been epoxied, were posed from a prototype picture. The cruel close-up proves that the push fitted Jim Smellie dart has drifted round and requires further work!

An unfeasibly large Lenz 'LE1000' DCC chip was fitted in the cab, and the little loco taken to its future home for a quick test run. Although the loco ran impeccably on DC when tested in this form, this was not the case on its new home layout with DCC control. The motor had a disturbing 'growl', and it seems these very old chips are not compatible with coreless motors. On the plus side a better 'Southern' lettered Hornby tender with Zimo chip already fitted has been provided, and other parts for the build were collected at the same time.

 

[Stevers]

First job of the day was to put the step under the smokebox. This step is quite near the edge of the smokebox barrel, so mindful that this will not be my last Wills T9 I made a little jig in styrene to accurately place the holes. Underneath the smokebox door is the upside down step showing the wire bent to form a bracket, soldered on and then thinned. The dart has had further cleaning up and is now countersunk into the door. The lamp irons are from a Mainly Trains (now Wizard Models) fret. They were soldered on with 100° low melt solder, but in practising my technique for this I found it absolutely essential to tin the brass with normal (electricians) solder first. Don't remember the divot taken out of the door and more clean up clearly required. It's so easy to make this stuff just ever so slightly wonky!
The smokebox door casting dry fitted in the barrel and still not quite straight! For LSWR smokebox lamp irons it's very hard to get the bends in all the right places on something that is a) impossibly small, and b) quite complicated. The further away one gets the better it looks though.


Next an attempt was the made to fit the D handles in the frames just behind the front step. In looking at exactly where each should be in relation to the clack valves and step, I finally twigged that the front steps were still in the wrong place! My friend's dad had fitted them just ahead of the front driving wheels directly under the valance. First of all I had replaced them there, then I recessed them there. In both cases they were poorly located like it was trying to tell me something! Anyway it turns out that the front steps should be in front of the rear bogie wheel where they would have fitted nicely if I hadn't ground the area flat to improve clearances for the bogie wheels. The only drawing that I had to hand was for the saturated version and that shows the front step on the centreline of the rear bogie wheel, however when these engines were superheated the bogie was moved back 3" to improve weight distribution, so that the footstep should (in 4mm scale) be 1mm ahead of the rear bogie wheel centre line and that's pretty much where they've ended up. (Ed. Not true - there was no change to the wheelbase on rebuilding!)
Not very obvious is the tiny D handle that caused all the trouble! Next jobs are to drill the cab front for the rods and pipes for injector feed and whistle, then go over the barrel with some 1200 grit wet and dry to smooth it all over. Pleased with chimney and dome, but the fit of the injector feed pipes continue to leave much to be desired!

This is the replacement tender which runs very nicely and already has a Zimo DCC chip on board. The Southern transfer is shaded black, and I have no idea if this is correct for a wartime paint job. The other side of the loco showing that the other front footstep isn't quite straight yet.

 

[Tim Hale]

I rather like the tender’s SOUTHERN, no idea of the true origins as it was a gift but as the loco will invariably be finished in the hybrid SR/BR livery, it suits the loco.

The T9 will spend its days on pick-up goods on Beaminster Road, they tended to be used for just about anything and the Southern certainly got its money’s worth from Drummond’s ‘Greyhounds’.

30285 climbs Parkstone Bank - Dr.Ben Brooksbank creative licensing ​

BR

 

[Daddyman]

You're making a nice job of this.

However, I do think the smokebox dart is letting the side down. I read the following from Malcolm Mitchell in the MRJ last night: "Have a go at improving or replacing any casting or component that you feel doesn't quite come up to the mark , or may be missing altogether." You're dealing with plenty of missing parts; have you considered making your own dart?

Here's the principle: 0.9 tube (probably - see below), nicked with a triangular mousetail file to aid the location of the 0.4 drill, which will then take 0.4 wire to represent the handles proper. For obvious reasons, the hole for each handle will need to be at a different angle on the tube, and inwards of the other hole. In the photo both holes have been drilled and the wire for one handle inserted and soldered in; it awaits the excess removing. In removing the excess 0.4 wire and dressing the solder afterwards, the triangular nick disappears.

It may be that you need to change the dimensions of the components to suit your prototype; I took my dimensions off the J36 at Bo'ness.

All my locos have homemade darts (but they're all LNER, which would sully your thread, so I haven't included a photo of the finished article!). I've never found a turning that's not grossly overscale and/or crudely rendered. The only exception is Dave Bradwell's, but as it's part of a sprue with other fittings on, it's quite expensive (£7 last time I bought one).

 

[Stevers]

Thank you Daddyman for the wise words. Not quite so obvious in the flesh, but yes, in such a cruel close up the dart still leaves something to be desired - and that's after an awful lot of whittling. Originally I was going to make a dart, but didn't think the twist of wire method would cut it. The Jim Smellie dart was provided, so it seemed politic to use it. The beautifully made Markits dart seen on a Westward T9 build also seemed too chunky for a T9.

A photo of 311 in the Derry T9 book shows the handles to be almost vanishingly thin and the top one at 10:30. T9 darts also need a sort of plate behind them, but that could be a Markits crankpin washer or AGW crankpin nut. It is pure genius to notch the tube to aid drilling - that idea had never occurred to me! Will try your technique this evening, and feel confident that it will not only be nearer scale, but allow me to model the handles positioned as they were in the Derry picture.

 

[Daddyman]

That's it, yes - it allows you to position them as you like, and to choose the dimensions you want. I've often used 0.3 for the handles on smaller locos.

The triangular notch is not really genius - just a work-around for not knowing how to use my bench drill! If you have one (a bench drill), the tube could be wedged in something or other and then a 0.5 centre drill used (or smaller if they exist) - but it's the wedging that's always seemed too much trouble with the bench drill, so I rarely use it.

 

[Stevers]

Chapeau to Daddyman! Exactly as suggested this is 0.9mm microbore tube with 0.45mm N/S wire handles. If I hadn't been up on my roof today stripping out rotten slates, I would have gone straight down to the workshop to have a go, but as it was it had to wait for tea. Not difficult to do, if a tad fiddly due to size. The ring is an AGW crankpin bush. Just how cool is that?