Prepared by Tommy Day
Originator J. Snowdon
Thread and Forum Title Jim's workbench 3: GW 2000g Tender build
Thread start date Nov 16, 2017
For full Forum thread click here : - https://www.gaugeoguild.com/xenforo/index.php?threads/jims-workbench-3-gw-2000g-tender-build.983/
it's back to 7mm scale modelling again. A job that has been in the list for some time has been the provision of a new tender for a small GWR 2-4-0 that I have. This currently has a 3000 gallon tender that is nearly as large as the engine and which, since Warren Shephard introduced the kit, has been earmarked for replacement by the short 2000 gallon version. Starting with the chassis, this comes as a single piece etching that is great from the point of view of simplicity but isn't as easy to fold up as it looks. The fold lines are continuous from one end to the other and to complicate matters, there is a large opening in what will be the top of the chassis. Without thinking, I set to bending this up using the bending bars in the vice. The result was that I could only bend the frames down to about 80 degrees, well short of square. Looking at it again, the underlying problem was that the half-etched fold line was a bit narrow for the thickness of brass, with the result that it was closing up tight before the bend had reached 90 degrees. The length of the fold then made it difficult to persuade the metal to give any further without using brute force. What I should have done, before starting, was to cut along the length of the fold line in varoius places to make it a series of short bends in line. Doing that with the job half done is not so easy and the alternative, the frames were put in my small vertical mill and the fold opened up that way -
Thus relieved, the fold could be finished off square and the axle bushes inserted. Milling the chassis etch this way is a bit of a case of applying a sledgehammer to crack a nut, and is really the consequence of not thinking first.
The bushes for the centre axle are designed to have vertical float in the chassis, but no way of preventing them from rotating in the holes. The solution is the classic one of fixing a springy wire link between the bush and the chassis, in this instance using an L shaped arrangement with the tail coming vertically downwards to the bush.
With the wheels in place, the next step is to add the brakes. For these, Warren has provided brass castings for the double hangers, with separate moulded plastic brake blocks. The castings are substantial, in as much as they are designed to fit over 1.5mm rods threaded through the chassis. the bit that needs thinking about is with all that brass, how to attach them without the heat affecting the plastic parts.
I solved that by turning up a small brass disc to match the dimensions of the plastic brake block plus a small clearance to the wheel tread, and used that to set the hangers for the leading wheelset.
The remaining hangers are then set by using the etched pull rod , after which the plastic brake blocks are secured in place using short lengths of Evergreen rod.
Jim
Turnbull21603 - Nov 16, 2017 at 12:54 PM
Some year or so ago I came across this kit and thought it would be just the ticket to put with a Duke class 4-4-0, in fact the 2000 gallon tenders were built to be paired with the Dukes. Only 60 were built at Swindon between 1895 and 1899. They were built as lots A24, A25, A27 & A34 and subsequently passed on to 3521 and Stella class locomotives as the Dukes were rebuilt. The reason for these short wheelbase tenders was due to the short turntables in Cornwall, were the Dukes were intended to work
Now the big mistake. This kit is for a non existent tender and probably scaled down from a 3000 gallon tender. During a conversation with Warren, for replacement of missing castings, he mentioned that the water pick-up clashes with the rear axle. This prompted me to look up details of the Duke tender and resulted in the information above, which was taken from RCTS volume part 12 ( see pages M10 & M11). I did, at the time, send copies of the drawing from O S Nocks book for his information.
I subsequently scrapped the kit and made drawings for scratch building the required tender.
I am attaching a drawing of the tender (see forum thread for drawing), taken from Russell volume1 and a photograph of the completed tender paired with a Duke class in as built condition.
I still have my drawings and am prepared to send a set to you Jim, if you would like copies.
Les
DavidL - Nov 16, 2017 at 1:57 PM Those plastic brake blocks look useful. I wonder if Warren could be persuaded to sell them separately. David
Turnbull21603 - Nov 16, 2017 at 3:00 PM David, Warren used to get the brake blocks from Springside Models. If you go to their website you will find that Springside supply these in three sizes. Les
Bob Alderman - Nov 16, 2017 at 3:07 PM Neat solution to the long fold line Jim. What would you suggest for those without a mill? Bob
DavidL - Nov 16, 2017 at 3:48 PM Use a skrawker to widen and deepen the etched groove. David
Jim Snowdon - Nov 16, 2017 at 5:24 PM Les, Thank you for your comments in regard to the tender. However, can I point you to Figure 597 on page 231 of Russell Volume 2? That is a copy of a Maskelyne drawing for very much the type of tender that Warren has modelled, and whilst I cannot vouch for the accuracy of the drawing, I believe Maskelyne was rather more on the ball than some of his contemporaries. Fortunately, in my case, plausibility matters more than absolute accuracy, for reasons that will become apparent once I post the before and after pictures of the loco and tender, which I will leave, of necessity, to the end of this topic. Bob, Had I not walked straight into a trap of my own making :mad:, the simple technique is to drill through the fold line at suitable points and then, using a piercing saw, cut along the fold line to create the gaps. It's what I should have done in the first place, so the exercise in using the mill is largely an example in the use of tools to get out of a corner. Using a skrawker would not be a solution of choice as whilst it will remove metal from the thickness of the fold line, it will do nothing to widen a fold that was already too narrow to start with. With the fold line gapped, getting the bend to the full 90 degrees will involve a certain amount of stretching the remaining metal and having the full thickness still there helps. Thinned, there is always the risk of inducing a fracture and then having to go along the lengt of the fold with solder just to add reinforcement. Jim
Bob Alderman - Nov 16, 2017 at 6:12 PM Jim I asked the question as a bit of a leader to get alternative methods from you and others. For myself I run a three square (triangular) file down inadequate half etches. This does widen them to a degree as well as deepening them. Cracking failure can be avoided by only bending once. On such bends I always run a fillet of solder into the corner. Bob
Jim Snowdon - Nov 16, 2017 at 6:16 PM Bob, Fair comment. As an aside, a triangular file is one thing I haven't got. Whether I would find one useful would have to be a moot point.
Arun Sharma - Nov 16, 2017 at 6:24 PM Jim - most, if not all, of the cheap sets of “needle files” have one or two with a triangular profile suitable for etch line widening. So whilst the old adage that you should buy the best quality tools that you can afford holds true, there is a lot to be said for spending a few pennies for utility purposes.
Turnbull21603 - Nov 16, 2017 at 6:42 PM Jim, The figure that you have referred to in Russell vol2 is part of the drawing in J.N.Maskelyne’s drawing of GWR 3230, a member of the 111 class. The full drawing and description appear on pages 62 and 63 and dates c1911, from the text, although possibly as late as 1922 when this locomotive was finally withdrawn. Scaling the drawing Maskelyne has accurately drawn the overall wheelbase as 11’ 0”. The tender is the same as my drawing of the original but modernised with coal side boards instead of the original coal rails. The problem with the model is the rear of the footplate. The kit does not have the apparent extended rear footplate as shown on either of the drawings as the actual length of the tank is incorrect. You will find this when you assemble the tank on the footplate and compare it to Maskelyne’s drawing. However as you have written accuracy does not matter as much as plausibility. That is your choice. Personally I prefer accuracy, Rule 1 comes into play here. Before I build a locomotive I research through my books in order to make a reliable facsimile of my chosen class and period. I look forward to seeing your photographs of the end product. Les
Jim Snowdon - Nov 19, 2017 at 12:02 AM The next step in completing the chassis was to work out how the brake shaft fitted across the front of the frames. As the kit is designed, the brake shaft is carried between vertical brackets that will sit behind the front footsteps, thus allowing the chassis and body to be separated. For whatever reasons, probably to do with space, these brackets are provided as separate pieces, rather than a fold down continuation of the chassis brackets. The snag is that the instructions give no clue as to whether they should be solder to the ends of the horizontal bracket, or under the end.
The brake shaft is, on GW tenders, supported in the middle by a vertical stay, which is provided in the kit. Added to that, the brake shaft casting, in brass, is provided with a boss at either end that determines its overall length, once the casting runner has been trimmed off. Trying it against the chassis indicated that the vertical brackets should go underneath, so they were tacked in place and the brake shaft casting offered up. It is one of those design details where a half etched rebate on the appropriate component would make the assembly self-evident, as much easier to keep everything in line whilst making the joint.
Then, with the shaft in place, the front section of the brake pull rods can be attached. The brake cylinder can wait until later, particularly as it may conflict with the drawbar.
That completes the chassis, bar minor details, so it is on to the body.
As might be expected for anything that is basically a box, construction is fairly straightforward. The sides are each provided with half the end, with a cautionary note in the instructions that the end may need trimming to allow a clean joint in the middle of the back panel of the tender. To my surprise, they fitted perfectly, which must be down to a fluke on my part, as the fit depends on getting the corner bend in just the right place.
Ordinarily, the sides of GW tenders are liberally provided with rivets, which are noticeably absent from the etches, both on the platework and the angle that runs round the bottom edge of the tank. Since the tender sides are half-etched, I am inclined to wonder if something has gone awry in the preparation of the etching tools. Something I can take up with Warren separately.
The sides of the coal space fold up from a single piece and are attached to the underside of the tank top plate. The “U” is a little wider than the opening in the tank top, which is better than the other way round. The top plate is supported on the coal space sides and to fold up spacers that I presume, as they were not referenced in the instructions, fit in the back corners of the tender. With them in place, the top can be soldered in place.
Jim
paul copsey - Nov 19, 2017 at 7:06 AM Just like your other builds Jim , clear photos with very descriptive text . Excellent . The jar with T208C written on the side is your flux ? Cheers Paul
Jim Snowdon - Nov 19, 2017 at 9:12 AM Paul, Well spotted - that's my flux, in a small jar as a precaution against calamities. It's an industrial no-clean flux, obtained from a company by the name of Solder Connection, in Monmouth. It's only snag is that they like to sell it in rather industrial quantities. I have enough to last me into eternity. Jim
Jim Snowdon - Nov 20, 2017 at 7:09 PM
this is what the top of the tender looks like now -
Jim
Jim Snowdon - Nov 21, 2017 at 11:55 PM
Getting back to the subject, with the basic box of the tender tank complete, the next stage is to make a start on the detailing, starting with the water filler. In contrast to the usual separate dome and the water filler, these early tenders had a flat D shaped box that covered both functions. The kit reproduces this by way of three etches - base, sidewall and lid. Assembly is straightforward enough, but I did spend a bit of time wondering why the inside of the sidewall incorporated an etched groove, into which the D shaped lid fitted deceptively easily. In the end, after looking at several of the tender GA drawings in Russell, I concluded that the groove was a red herring and that the sidewall fitted flange downwards, mating up with the flange on the base etch, and the lid placed on top. Two etched hinges, with a piece of 0.5mm wire added, are provided. Logic suggests there ought to be a handle for lifting the lid and this will get dealt with later.
Before the filler can be fitted to the tank top, the transverse bulkhead at the rear of the coal space needs to be fitted, then the filler butted up against it and soldered in place.
With that done, the rest of the coal space is built up by adding the side extensions -
Next will be adding the beading along the outside edges of the tank sides, which will require digging out the supply of fuse wire, but that can be for tomorrow.
Jim
Jim Snowdon - Dec 2, 2017 at 11:40 PM
Well, usually I would use 15A fuse wire for this, as it is a suitable size, easy to straighten by stretching slightly, soft enough to bend easily and, usefully, pre-tinned. I duly obtained some from our local branch of Wilkinsons and ended up finding that any attempt to straighten it resulted in the wire simply breaking.
Back to what I had in stock - 0.7mm brass wire…. At least it starts out straight, but requires a lot more by way of pre-bending to fit.
That is, to me, probably the most irritating part of building Dean-era GW tenders, and once done, finishing things off is largely a downhill exercise.
Last stage in completing the basic tender structure is closing off the front end. The tank fronts are simple enough, as plates soldered across the front.
followed by the front coal plates
The bottom of the coal space is a little trickier, as although the height is set at the front by the folded down flange, there is no other means for keeping it level whilst it is tacked in place. Fortunately, it is a good fit (not tight, but not rattling around in the hole either) and the back end can be prodded from underneath through the various openings in the tender floor plate. That is enough to get it held in place whilst applying finger pressure to the front end; the back edge can then be tacked and the rest of the joints round the edges completed.
There are, just inside the coal space three boxes for which small fold up etches are provided. Two support and protect the handles and control rods for the water feed cocks, the third is a sand box. The first two are square and simple enough to fold up, wit the only catch being that the side and top want to occupy the same space when folded up, instead of one being inside the other. It is down to a minor error in setting out the etching drawing and requires a little attention from a file before folding. The sandbox, in almost typical Swindon fashion, has rounded corners and although etched lines are provided on the inside to help with the bends, requires care if the back edges are to come out square to each other and the bends in the right place to match the top. Being a very small component, and difficult to get hold of, I made the bends using a length of rod and the flatnose pliers to grip the brass whilst pulling it round the rod. Suffice it to say it didn't come out right first time and took a bit of persuasion to adjust matters.
The front corners of the coal space bottom plate are cut away to accommodate the two boxes and the sandbox, which would be fine had I fitted the latter in place first and then adjusted the coal plate to fit. A case of RTFM, as it does indicate that the cut outs should be checked once the boxes have been formed. .
The result is that the boxes on my tender are sitting on, rather than in, the coal plate and are actually too tall, although in a practical sense it is noticeable only to those who know. .
The fire iron box is the last of the etched items to be fitted. The side and end are easy enough to fold up, although the flange along the top edge needs to be cut away at the end and the fold line deepened with the skrawker as it is a very narrow flange. Fitting is simple enough, as it attaches behind the front coal plate and butts up against the inside of the coal rails.There is no positive location so it is a case of holding it in place with one finger whilst getting the front end tacked in place. In theory, there are a couple of plates to go under the the fire iron space, and which are shown on the instructions but absent from the etches. Even if they were, they would need to be fitted first, otherwise they become inaccessible.
With all that done, it is time to start adding the detail to the underside of the tender, starting with the frames. Although there is no positive location for either the frames or the front and rear headstocks, their placing is largely intuitive.
Trying the chassis in place revealed that it needed to be shortened by about 1mm in order for the rear headstock to fit against the ends of the frames.
The last items to be fitted are the two valances and steps, the ends of which lie flush with the ends of the headstocks. The instructions advise fitting the steps first, and whilst I can see some virtue in doing that whilst the etch is flat on the work bench, I can also see there being a risk of the upper steps becoming unsoldered when making the joints to the headstocks. I chose to leave their fitment until later.
Having fitted that platework, the tender can be stood on its chassis and is beginning to look the part….
Next, the detailing.
Jim
Buckley10339 - Dec 3, 2017 at 9:01 AM It looks to have been quite a fiddly job Jim but it has turned out okay in the end.The problem with a lot of older kits is the fact that fold up parts can be awkward to do if they have not drawn out quite correctly in the first place.I have fond memories of struggling with a Jim Harris kit about 25 years ago when i was just starting to get serious about 0 gauge as nothing seemed to be exactly correct ! Pat.
Jim Snowdon - Dec 3, 2017 at 9:39 AM Pat, Actually, its quite a new kit that Warren introduced only about two years ago, as part of his WD Deans Goods package, and to be fair to him, the CAD drafting is up to “modern” standards. If anything, it's just that the fold lines are a little in the fine side for some of the metal thicknesses and a couple of minor glitches (which even the best make occasionally) in parts where folds bend towards each other. I was talking to Warren at Reading yesterday and looking at the extent to which he has taken to 3D CAD for his 15xx kit, both in respect of design and its use in making 3D printed masters for brass castings. In terms of fold lines, as a rule I make them the same thickness as the metal from which the part is being etched. What I also do, in 2D CAD, is draw out the folded part in section as part of the GA drawing for the kit as it will be built, including the bend, which then allows me to work out the correct length for each part allowing for the metal thicknesses. The problem with many of the early kits lies in plain simple poor design, exacerbated by a tendency not to go back and correct the etching artwork once errors have been found (if they were ever found on a proper test build). Being early in the field doesn't explain such errors as not getting the axle holes in line (a Harris kit), axle holes there were in line - on an Atlantic, where the trailing wheels are smaller than the drivers (Ace), parts simply the wrong size, especially relative to other mating parts in the same kit (see my comments when building the Modellers World LNW Coal Engine). The list is pretty well endless. What shows that it could be done properly is that people like Jim McGeown, whose kits go back to the same era, could do it properly. Jim
Arun Sharma - Dec 3, 2017 at 12:03 PM Jim - I wasn't sure what you meant by straightening the 15A wire for the beading. Did you mean flattening it or just pulling it into a straight line. I only ask because my experience has been that squeezing copper wire/fuse wire and pulling it between fingers is a pretty effective way of straightening it. If you want to flatten copper wire/fusewire, David Smith at Metalsmiths sell a rather interesting hand tool which looks a bit like a pair of pliers with a small wheel in the jaws which flattens wire for beading very well. A picture is hopefully attached:
{{:forum:forumlocos:gw2000_26_.jpg?400|}}
——————————————————————————————————-
Jim Snowdon - Dec 3, 2017 at 5:38 PM Arun, Simply pulling it to get it straight. It would probably be easier if I got myself a reel of the stuff rather than using the little cards that you get in DIY and similar stores. Until now, the process for straightening was simple enough - put one end in the vice, grab the other end in a pair of pliers and pull. Normally, there is a perceptible point at which the wire goes into plastic deformation, stretching in the process. With the Wilkinsons' wire there was no “give” - one moment you could be applying strain and the next it would break. Jim
Jim Snowdon - Dec 5, 2017 at 6:55 PM
And so to the detailing, starting with the rear end.
The kit comes with Warren's usual pre-assembled buffers, complete with cast brass bodies that fit neatly into the holes in the headstock and can be soldered in place, although with their mass, this can sometimes be a taxing job for the soldering iron. Quite often, including in this instance, I find it useful to give the buffers a final waft with the gas torch to ensure that the solder flows properly under the base on the buffer body.
Fitting the plumbing calls for a little ingenuity, if only because many kit designers forget to think how the castings will be attached. The vacuum pipe is not too bad as it has a mounting bracket that sits in contact with the headstock, although before it can be fitted it is necessary to file a half-round recess in the back edge of the tender footplate. Fitting the steam heating pipe is another matter, as it lies below the headstock. It really needs a small bracket behind the headstock into which the pipe can be fitted. In the past I have sometimes made this from a length of small wire bent into a U, but in this instance, it occurred to me that I had something suitable in the form of the tail from the Premier screw coupling that I had fitted earlier. (I normally solder the drawhook in place, not least as nothing irritates me more than seeing it being pulled well of its stops when hauling a train. It also allows the amount protruding behind the headstock to be reduced, which helps with fitting the chassis.) All I needed to do was open out the split pin hole slightly, fit the tail behind the headstock and solder the steam heat pipe into it.
Fitting the lamp irons turned out to be one of those jobs where a resistance soldering unit would have come in useful, largely because holding the things in place is tricky and the amount of brass in the vicinity takes a good bit of the heat away. I have one, but still haven't got round to setting it up - it means clearing some more bench space.
A problem that virtually all kit designers face is that the typical spacing of the frames on tenders is close to that of the buffer centres. On full size locomotives, the use of self-contained buffers solves the problem, but these are tricky in 7mm scale. Inevitably, the buffer head has a tail that projects through the headstock, and either a tweak is required to the dimensions, or, as Warren has done with this kit, a cut-out is needed in the frame plate. With this kit, the cut-out is singularly generous in its proportions, and even allowing for the rear foot steps covering some of it, is still very visible.
Something needed to be done about that, the solution being to solder a section of scrap etch across the opening to close off the visible section.
Next is to start on the front end of the tender. The kit does provide two L-shaped brackets to support the handrails, however I failed to see how they were meant to fit. It is also an area where there is considerable variation in the full-size design, with some tenders having the handrail continue up and along the top of the tank side. I chose to keep things simple and made up two straight brackets from some etched balance weight levers that I had left over from build the signals on our club's layout.
At this point, the tender needs to be married up to the loco to determine what, if any, adjustment is needed to the footplate height. It will now become clear why I am building a tender on its own.
The locomotive to which this tender will be attached is a small 2-4-0, sufficiently so that the standard 3000 gallon tender was nearly as big as it was -
At the time, the smaller tender was available only from the Home of 0 Gauge Mega kits range, and then only to order. Consequently, no. 13 had had to acquire a larger tender from Mr. Finney's range. Then Warren Shephard came up with the diminutive 2000 gallon tender as a result of his doing the World War 1 era WD adaptation of the Dean Goods. The result is a tender that is better suited to the proportions of the locomotive.
Jim
====
SimonD - Dec 6, 2017 at 7:44 AM Jim, Apologies that I did not post earlier. You, mentioned rivets, afaik, these tenders were flush riveted, as the kit, and your (and my) build. Unfortunately, the water filler was not flush riveted. I ended up remaking mine. Best Simon
Shephard5287 - Dec 7, 2017 at 12:05 AM Just sticking my head above the parapet, I mention in my instruction to use a square file, each corner being 90 degrees whereas a triangular file each corner is 60 degrees so when the bend is made only 60 degrees of bend can be achieved, whereas with the square file the groove made is 90 degrees so the bend will form 90. Now ducking, Warren
Jim Snowdon - Dec 7, 2017 at 12:16 AM Good evening and welcome, Warren. With the etched groove set at the correct width, the use of files is not required at all as the remaining metal will bend to the full 90 degrees, with the edges of the groove meeting. The rule I use, and was taught by another respected designer of etched carriage kits long ago, was to make the width of the etched groove the same as the metal thickness. I have used it in all the kits I have designed without any problems becoming apparent, either when I have undertaken the test build or subsequently. Jim
Jim Snowdon - Dec 12, 2017 at 9:46 AM
And so to the end. With the main brass work all done, it's down to the details. Unlike many kits, Warren provides the two tool boxes as etchings, which saves the inevitable chore of cleaning up the mould lines across the ends. Even better, the hinges and locking hasp are designed to fold into L shapes that locate in the top edge of the box before the lid is added. The only snags are that the lid is quite a bit too wide (which is better than being the other way) and needs trimming to size once attached, and that the boxes have no base, which left me not quite certain how they were intended to be attached to the top of the tank. That was resolved by soldering each of them onto a section of scrap etch so as to provide a solid base, when then allowed me to glue them in place.
The rest of the front end is completed by the addition of the handles for the feed water cocks and for the sand box, together with the hand brake column. I haven't fitted a water scoop to this tender, partly because although the bits should have been in the box, they weren't - I suspect that they had been borrowed before I bought the kit (and Warren would have more than happily replaced them if I had asked). Consequently, there is no operating handle fitted either (although I do have the casting).
All that was left then (or so I thought) was to fix the axlebox castings in place. As if often the case, the ends of the springs are a tight fit behind the valance, and in this instance had to be filed at the ends. With hindsight, if the frames had been located a little bit further in, they would have fitted.
Fitting the tender body to the chassis revealed one tiny problem, in that the flanges of the leading wheels were rubbing on the bottom edge of the plate that forms the coal space, necessitating a quick session with a grinding bit to cut the exposed edge back a bit.
The result -
and just for comparison, the new tender against the original 3000 gallon version -
It still has to be painted and weathered to match the loco, but at least now the proportions of loco and tender are more in keeping with each other.
Was it a good kit to build? The answer has to be yes - there were a few issues of my own making, but it went together well with no fundamental snags, and nothing that could not easily be dealt with by the builder.
Jim
