Drawing 23 - Tender Body
1. Side Sheets - bending    
The tender tank is constructed almost entirely from 16swg brass sheet and I had the main pieces guillotined from a 4ft x 3ft sheet by a local fabricators. The two tank sides need to have bends put in them and this would normally be done on a large folding machine or pressed to shape in a hydraulic press. I have neither and didn't want to have to make a large piece of equipment just to form four bends so decided to make a wooden former and bend them around that. The material chosen was a short length of kitchen worktop with a half-inch radius to the front top edge. Since this is specific to the one item, I cut the worktop a 1/4" longer and 1" wider than the tank sides.
Then, using an offcut of the same material, I clamped it up and put a bend in to find out just where to set the tank side on the former and to see how close to a right-angle I could get. I reckoned that I needed to set the panel 7/8" from the back edge and about another ten degrees undercut to the front edge. To achieve this I planed an angle on the edge. Then I put my offcut back on the former and clouted it a bit more with my nice new rubber mallet. I was happy with this so now I loaded one of the tank sides to the former with a decent-sized piece of support material clamped on the top, put the whole lot in the workmate and gently tapped the edge over, working back and forth along the panel to keep the bend reasonably even throughout.
Eventually, I reached the point where it wouldn't take any further set (springing off the former) and just kept gently tapping until I had a decent straight line at the bottom. This was then repeated with the other tank side.

The upper bend in the panels is at twenty five degrees from the horizontal and to form this I first had to reduce the size of the former and chose to saw it down to five inches wide.
Then the offcut from earlier was clamped to the former with a half-inch spacer at the back and the metal tapped down until the twenty-five degrees was achieved. This was then offered up to the drawing to check the angle and to see what distance to set the rear spacing at. As you can see, it's about 7/16" short so the spacing needs to be 15/16". I put a couple of wood screws in the back of the former and set them to this depth, in such a position that they would not foul the bend on the material. I also set another piece of timber, which I planed to twenty five degrees, at the back of the fomer to act as a stop.
This time, I also clamped a support bar to the part of the tank side that I would be hammering over because this should help keep the sides straight. However, I found that this was too restrictive and I couldn't get the material to bend so I went back to the freehand method and slowly worked the material down till it hit the stop. And I now have two tank side sheets ready for the next operations which are to remove the surplus material from the top bent section and then drill all the holes - a heck of a lot of them.
2. Side Sheets - machining    
Once the bending work was finished, the tank sides were clamped to the mill table and had the lower section that connects to the sole plate reduced to the correct size. This was done with a small end mill but it chattered quite a bit until I balanced the cut at about thirty degrees. The next job was to drill all the 3/64" rivet holes, one hundred and thirty per side, and this was done on the milling machine using co-ordinate drilling with the DRO. It wasn't possible to do them all in a single setup so each tank side was placed as far to one end of the table as possible and holes drilled until past the half-way mark. The sides were then moved to the other end of the table, a previous hole picked up for reference, and the rest of them drilled.
This was followed by mounting against an angle plate and the 1/16" dia holes drilled in the lower section for riveting to the sole plate. Another job that required attention was to reduce the length of the upper tank section to the correct size for abutting the coal bunker. This was done by setting a slitting saw at the correct height above the mill table and the tank sides clamped down. As required, extra clamps were added to hold the centre section firmly to the table and dampen any vibration set up by the cutter.
Clamps were also added to the section that had been cut just in case the cutter grabbed and caused any probems. This was simply a case of taking my time and feeling my way along the cut. After the half-way stage, the work was moved to the other end of the table and continuing to the end. I stopped 1/8" short of the end and finished separating the parts with a hacksaw. There are more holes required for fixing the support angles to but I will drill these once the angle pieces are made. These will be joined with 8BA countersunk brass screws and sealed with soft solder.
3. Rear Panel    
As with the tank sides, I had the rear panel guillotined from the 1/16" brass sheet and was pleased to find that the sizes of all the cuts were within ten thou of my requested dimensions. The drawing doesn't specify whether the end panel should butt up on the outside or fit inside the tank sides. I have chosen to have the end panels outside and are sized accordingly. The rear panel was clamped onto a piece of MDF, trued up and all the holes drilled using co-ordinate drilling on the mill using the DRO. The drawing also calls for the panels to be riveted together with brass angles internally but I have chosen to use 8BA countersunk screws to fix the panels. These will be filled and rubbed down after assembly is complete. This is how the Modelworks tender appears to have been constructed and it means I can dismantle the tender at different times as needed. Riveting everything early on means any modifications become much more difficult to execute. Once assembled, the form of the tank sides was scribed onto the rear panel, the panel taken off and the waste removed with hacksaw and files.
4. Centre Bracing Panel    
In a departure from the drawing, I made a centre supporting panel to add a bit of strength to the tank sides and to create somewhere to add two more brass angle supports. It fits just behind the coal bunker and has a piece of brass angle to support the rear of the bunker and has another piece on the other side to support the rear top plate. It was made 1/16" shorter than the internal height of the tank and the top plate sits on it, butting up to the coal bunker. As before, all holes were drilled using co-ordinate drilling, and the centre sections were removed by drill 3/8" dia holes at the corners and milling out with a 1/8" end mill. Brass angles were made to suit which were riveted to the centre panel with 1/16" brass rivets. The panel is fixed to the tank sides and sole plate using 8BA brass countersunk screws. This panel is similar to the one that Modelworks used in their variation of this tender. Thanks are due to John Johnston for the photos of his 5" gauge Britannia that showed me how these parts were made and utilised.
5. Front Bulkhead    
The front bulkhead forms the forward extremity of the water tank leaving a short space between it and the front panel for the lockers in the centre and the water scoop and brake operating mechanisms on either side. This was made to be as close a fit as possible between the tank sides and sole plate to facilitate soldering and the fixing angles are riveted to the panel. As with the other panels, I have chosen to fix the panel to the tank sides and sole plate using 8BA screws. The locker section is removable to give the driver easier access to the footplate when firing and driving but is replaced when on static display.
6. Bunker Floor    
I have made the bunker floor and the rear panel from a single piece of material, this time it's an old brass kick-panel salvaged from a discarded office door. Apart from the rear section, which is dimensioned as per drawing, the rest is scaled as no dimensions are given, it's a case of bend-to-suit. To set the radius of the uppermost part, an indent was made on the centre-line at the correct point and a compass with scriber used to mark out. The waste was then removed with a hacksaw (with the blade mounted sideways) and the edges filed to clean up. The dark colouration in the photo comes from the varnish coating and this was removed with emery on a sanding block.
Before bending, I set the panel up on the mill and drilled the four rows of holes in the rear section. This was then followed by clamping the top section between two pieces of large flat steel and bending the floor section to approximately twenty two degrees and repeating with the rising section to about eighty degrees. These were then eased until a nice fit was obtained at the front bulkhead and the 2.3/16" reference dimension for the lowest point was attained. There is more to do on this section but I will wait until I have made the bunker sides and the lift-out section before finishing this.
7. Bunker Sides    
The bunker sides were marked out on another salvaged door kick-panel using a cardboard template that I made to ensure the developed form came out correctly. I managed to get these cut with the minimum of waste by using a common centreline between them. The panels include both the angled sections as well as the vertical sides and a bend was marked out by placing them into the framework, resting on the bunker floor, and marking each end point from the tank sides, then scribing a line between the points. These were then bent to approximately twenty degrees and offered up but a fair few sessions were needed before I had a good fit to the bunker floor and vertical sides that were actually vertical. Some adjustment was also needed on the ends because the bend changed the angle of the rear upright by a few degrees so I filed this back to square.
Once happy with the fit of this, I then removed the surplus that I had left on the top of the vertical sections to the specified 1.1/2" height by clamping on the mill table, clocking out the bending line and milling the waste with the side flutes of an end mill.
8. Removable Front Section    
The front panel needs to be made in conjunction with the removable section but I will cover this as a single entity following this part. I have no intention of making any of the locker doors to be opening, nor am I bothering with the actuating gear for the scoop and the tender brakes. The handles will be fixed dummies, and the tool tunnel will be modelled at the front but only for the depth of the removable section. When in use, a drop-in panel will replace the lift-out section to retain the coal but when on display, a dummy tray depicting a full load of coal will be used. Therefore, I also feel that to model the full length of the tool tunnel is a waste of time and material. Here are the various parts that make up the removable section, all made from 16swg brass and some more home-made angle. The extra holes in the right-hand side panel have no purpose, I've just used up a previously scrapped item. .
All the parts were bolted together and, as can be seen, the angle extends down past the wider section and acts as a locating face to bear against the front panel. The side panels and the tops of the angles are set down by 1/16" to allow the roof to rest on them. The assembly was then offered up to the tender for fitting and I could see that the sides of the coal opening were too angular, projecting past the coal plate, and needed taking back a little. Also, it didn't sit too square in the opening and was pulling backwards slightly at the top right-hand corner.
I decided to make the roof section next to see if that would pull the whole thing square and a section of brass was cut a lttle longer than needed and the same width as the side panels so that it would slot in to the assembly. I don't have any bending rolls, and I was too impatient to wait until my next visit to the club, so opted to form the radius using the nearest available piece of kit. Using the topslide to trap the lower edge, I formed the radius around the chuck by hand pressure alone and then eased it back freehand to fit nicely in the lift-out section.
The ends were then trimmed to length and the assembly offered back into the frame but it still didn't fit as nicely as I would have liked. Laying it flat on the surface table, there was a noticeable rock along the 2pm - 8pm axis so I decided to dismantle everything and skim the sides of the angles again, remembering that they were now riveted to the side panels. I set up the small angle plates on the mill and skimmed the edges of the inner panels to about a ninety percent cleanup, reducing both parts by the same amount. Approximately fifteen thou in total was removed.
Of course, this meant that the roof section also had to be reduced by fifteen thou and this prompted an interesting but unusual setup to enable the milling of the edge. Winding both handles of the mill at the same time to keep the cutter working all the time takes a little getting used to, and I found that it chattered slightly in conventional-milling mode but performed much better when I moved the cutter over and worked in climb-milling mode. Being a heavy table and making a fairly light cut, this presented no problem for the mill. Upon re-assembly, the whole thing was offered back into the frame and is now a much better fit so I will leave any further tweaking until after I have soldered the various parts together.
10. Front Panel    
The front panel has been made in a similar fashion to the other panels and has been seen in the previous pictures. After offering up and scribing round the tank sides to get an outline, I next cut out the hole for the water-level gauge, being careful to work right-side up because of the countersunk screw holes. I then cut out the widened section of the front plate. I have opted to remove the section just above the water scoop and brake gear panels and I set the width based on the dimensions for the edge of the tool tunnel. However, photos of Britannia show that the drawing is wrong and the tool tunnel is much wider than drawn. I really should toss these drawings in the rubbish, where they belong!
The insert for the removable section of the front panel was clamped to the face and marked out prior to cutting. On the front plate itself, I have now produced the cutaways on the lower section that reveals the brake linkages and scoop operating shafts respectively. I milled a 1/8" slot along the top first, then hacksawed the rest out.I find this much more satisfactory than chain-drilling, bending out and filing to size.
I also clamped a couple of the previously-made window frames (next page) to the panel, 1/8" in from the edge, and spotted through to create the mounting holes. Rivets were then dropped through and the internal shape scribed onto the front plate. I was lucky to find that a 16mm hole saw passed through the window opening with a room to spare so the centres of the windows were marked on the panel and then drilled through, followed by filing out to size.
Finally, I have made the tool tunnel cutaway in both the front panel and the lift-out section, marking out by scaling off the drawing for part of it and freehand with reference to a photo. Here, I have offered it up to the tender to see how it matches with the bunker sides. Quite a few of the tenders that I have seen have made the tool tunnel too narrow. It should extend to be in line with the edge of the brake panel. A little more filing was needed when that picture was taken. To the right is a photo of the panel with the locker doors cut from brass shimstock and laid on the panel.
The last work to have been done on the tender front panel was cutting out the triangular cutouts below the scoop and handbrake handles. I have now "ironed on" the two brass shimstock panels just above these and have been set with some 3/64" rivets to mimic the retaining screws seen in the photo. The holes for the brake and scoop handles were also drilled. The bosses were made from 3/8" dia brass bar turned to about 1/4" dia to better represent the full-size item and with a 6BA thread on the other end for fixing to the front plate. The handles are from 2.5mm brass rod.
These handles are dummies and will never be turned so the shoulder represents the boss on the bulkhead. The only thing that was less than straightforward was drilling the handle holes at twenty degrees and this was achieved with a simple drill jig to steady the drill. The box section below the coal plate has been made from a piece of 16 swg brass sheet soldered to a pair of 3/8" wide x 1/8" thick brass strips and held on the front panel with 6BA screws from behind. The 6BA holes were drilled and tapped before soldering together and the mounting holes in the front plate drilled to suit.
I decided to have a try at forming the water gauge using a punch and die arrangement. The drawing shows the radius of the gauge face as 1" and set into the front panel with a raised rim. A photo, however, shows the gauge set behind the front panel so this is what I have tried to create. I started by cleaning up the end of a piece of 1.1/4" square bar on the mill and then cutting a recess into the end using a slitting saw, making multiple entries at 1mm height intervals. To make the punch, I turned a piece of flat bar to 1/16" less than the diameter of the slitting saw. Then I cut a piece of 26 swg brass sheet about an inch square, put it on top of the die and gave the punch a whack. As can be seen, the first effort was less than successful, practically chopping the middle out. I then rounded off the edges of the punch and cut a new piece of material from 24 swg instead.
This was more successful and a little tapping of edges soon had it looking reasonable. I cut a slot in the front using a junior hacksaw and made a tiny indicator from some six thou shimstock, folding it round, soldering together and cutting the shape with a tiny chisel. All these parts were then fixed to the front panel. The coal dust deflection plate has also been made and fitted but is as per drawing rather than prototype.
11. Bunker Assembly    
It's getting near the time to solder the bunker panels together so I have been adding any extra features while I can still lay the panels flat. The first of these are some guide rails behind the removable section to help hold this in place. To help with keeping the bunker rigid and in shape while soldering, I have also made a range of brackets to enable me to screw the panels together. The brackets were all made from offcuts of the 16swg, with one hole drilled and tapped 8BA and then the bracket bent to the correct angle. The floor panel was drilled for 8BA clearance and countersunk to allow the screws to be recessed slightly. The bunker was reassembled inside the tank and the mid position of each bracket marked onto the side panels.
The panels were then removed, an 8BA clearance hole drilled 1/4" in from the edge and the panel replaced in the tank. The other leg of each bracket was spotted through the respective panel holes and then removed, one at a time, and drilled and tapped 8BA. Because I will be making a dummy coal tray for the top, I also made internal corner brackets for the rear of the bunker. These serve a dual purpose - holding the bunker sides in the correct position, and acting as resting points for the underside of the coal tray. They have been made with greater accuracy than the other brackets and are precisely 1/2" below the top. Other resting points have been added at the opposite corners and half-way along, making use of the rows of rivets visible on the outside. All the rivets in the end and sides have been fixed, as well.
12. Water Gauge Hatch    
The water gauge access panel is not often modelled on BR standard tenders, although I have found photos of a few examples, and I have decided to incorporate this feature on my own. There is no mention of this panel on my drawings but reference to a photo of the real Britannia and her tender gave me enough to get started. This is only on the left-hand side of the tender. One of the problems with making this as a cutaway and fixing a panel behind the tank skin is the position of the fixings. Looking at the photo, it shows the screws, bolts, whatever they are, on the inner section of the panel, which requires something further below to fix to. The other problem is one of scale. My tank sides are from 16swg material which would represent the tender being originally constructed from 1/2" plate!. I decided to mill away a recess and incorporate dummy rivets which should look reasonably authentic after a coat or two of paint. I started by mounting the side panel on one of my general-purpose milling fixtures and supported the outer end with packing to dampen any vibration.
Using a 1/8" slot drill, I milled away the area to a depth of 15 thou with multiple back-and-forth passes and a final series of cuts around the periphery. The section was then given a quick rub-over with a fibreglass pencil to smooth out the tram lines. This was followed by drilling the fixing holes, which I worked out by scaling from the photo. It's not particulary accurate but close enough for this purpose. To finish, twenty 3/64" copper rivets have been soldered in the holes, and the two eyes were appropriated from the dolls house accessories, part of our hook-and-eye door closers, also soldered in from below.
13. Soldering the tank    
There were only the two long seams to join and one side was a close fit requiring no extra work to solder together. The other side, however, had a gap of about 20 thou for a fair distance. I didn't want to start pulling things about at this stage as I now have the bunker sitting nicely in the tank. To fill the gap, I flattened some copper wire and tucked this in along the section before fluxing up and soldering. Once it had all cooled, I gave it a good scrub in the sink with hot, soapy water and then inspected it for leaks. There was one area in one of the corners where the gap was just too large for the solder to hold so a snippet of copper wire was soldered in to seal. It's not the prettiest soldering job around, but I filled the bunker with water, levelled it off and sat the assembly on some kitchen roll. Eight hours later, the kitchen roll was still dry so the job was a success. Next, I soldered all the rivets on the rear panel, followed by the brass angles fixed to the sides and bottom to ensure no weepage around the fixing screws. I also soldered all the rivets on the tank sides while I had unfettered access. The hold-down screws and overflow pipe fitting were also soldered to the sole plate.
The position of the two bulkheads were checked and where the rivets fouled, these were dressed back flush and a touch more solder applied. The tank sides were now fixed to the sole plate with 1/16" brass rivets.One thing I noticed on completion was a few places where some daylight was visible through the joint between rivets, possibly because the underside edges of the tank sides were not perfectly flat or because I was too heavy-handed when fixing the rivets. Anyway, something to watch out for. A small amount of light hammering with a drift helped close the gaps and the solder will complete the job but there is a limit to how large a gap solder will fill, prefering to flow out rather than solidify in situ. The next stage was to fix the rear panel in place with the side fixing screws and solder the vertical joins to the tank sides. There was no mechanical fixing of the rear panel to the sole plate at this time so I fixed small clamps on each of the hold-down angles on the outside and used a length of copper wire as a fillet along the inside bottom of the tank. Once complete, the rear panel will have screws fixing it through the sole plate to the underframe to provide some mechanical strength.
I soldered the rivets and the bottom joins next, working from the back panel forwards one side at a time. Again, I checked the position of the two bulkheads and dressed back any protruding rivets. This was followed by fixing the middle bulkhead and soldering the side screws from the inside, then upending the tank and soldering the bottom screwheads from the outside. The final seesion was to fix the front water bulkhead in place and solder all round from the outside, although this becomes inside once the front plate is fixed in place. This was a little more awkward because the shaping wasn't perfect and I swapped over to an old-fashioned stick of "Genuine Virgin Solder" that dates back to the 1950's because this seems a little less runny than what I was using before.
At this point, I performed a leak test and found a couple of small weeps from the sole plate and one from the rear bottom corner. The ones from the sole plate closed after a reheat and more solder but the one in the bottom corner was more stubborn, requiring a couple of sessions with new flux as well as extra solder. Once sealed, I filled the tank with water, as far as was possible, and left it overnight on a towel in a growbag tray. Next morning, towel and tray were both dry and this just leaves soldering the bunker into place but I shall hold this over for a while and complete the front panel first. The solder work is very scruffy-looking but it will never be on view, so unimportant.
14. Next Item