Ian_C's workbench - P4 and S7 allsorts

Discussion in 'Workbenches, including workshop techniques.' started by Ian_C, 21 May 2017.

  1. Threadmark: AWS conduit clips

    Ian_C Western Thunderer

    AWS conduit clips. Tiny things, but very visible on the LH running plate valence and on the RH lower cab valence. No drawings for these but easy enough to estimate the size from photos. My guess is that the conduit was about 1" in diameter and 0.6mm wire is near enough. The clips are not too difficult to make and we need 10 of them. Goes like this...

    AWS clips story.jpg

    1. Brass shim annealed, for reasons that become clear later. No idea how thick, it was in the scraps box and seemed about right.
    2. Shim is cleaned up and tinned lightly.
    3. Strips 1.4mm wide are cut from the shim and flattened.
    4. The strips are folded over a length of 0.6mm brass wire, tinned side inwards...
    5. ...and clamped carefully hard up to the wire in a toolmaker's clamp. Annealing helps to eliminate the springiness of the shim stock and gives a very tight bend without much effort.
    6. You end up with a lot of these.
    7. The tinned surfaces of the clips are soldered together with a zap from the RSU. That stiffens them somewhat.
    8. The screw heads are represented by putting rivets in freehand with the trusty GW rivet press. Well they're very similar if not identical. The screw heads are surprisingly prominent on the prototype and riveting hard through two layers of annealed shim does squish plenty of material into the riveting anvil to form a decent sized impression.
    9. Cut to length with snips and tidied up.
    Started off with plenty to spare. Scrapped a few in the process. Ended up with only 11. Hoping therefore that more than one doesn't go to live on the floor, not much chance of finding that again.
    Last edited: 3 May 2019
  2. Threadmark: AWS done and fitted- mostly

    Ian_C Western Thunderer

    With the battery box fitted in front of the cab and the main reservoir positioned on the framing it it was clear that the reservoir was too fat and short when compared with photos. I eventually found a photo taken from almost dead side on from which I could scale it relative to known sizes. If anybody's interested I reckon it works out as 17.9mm long and 7.33mm in diameter. Serves me right for simply taking the dimensions straight from the Ragstone white metal cast reservoir without checking :(.

    AWS installed RH 1.jpg

    The conduit and clips beneath the cab were zapped on by RSU and the vacuum pipe from the main reservoir was made from 0.3mm wire. One thing that becomes apparent is that filling the rear sand box must have been a right pain as it's almost behind the reservoir and the pipe runs right above and behind it.

    AWS installed RH 2.jpg

    The small timing reservoir was soldered in place tucked away behind the ejector vacuum down pipe. Something I'd not thought about prior to tackling the AWS was that there are small pipes from the ejector vacuum pipe and the timing reservoir entering the cab through the cab front sheet. The holes for them were difficult to mark and drill in situ - one to note if I live long enough to do another AWS fitted 8F! Likewise there's a pipe fitting on the ejector vacuum pipe that would have been easier to fit before the ejector assembly was fitted to the loco. On many photos these small pipes were bent around all over the place - none of your 'ship shape and Swindon fashion' here!
    AWS installed LH 1.jpg

    Conduit was run from beneath the LH side of the cab along the valence to forward of the cylinders where it enters a fitting that takes it inside and out of sight. The wiring conduit seems to have been connected to the end fittings, and sections joined end to end on the long run down the LH side, by means of a flexible tube or hose clipped to the conduit. Those hose connections were represented by a short length of 0.8mm microbore brass tube (left over from the Silvertown lubricators - earlier post) drilled out to 0.6mm to accept the conduit wire. Here are the tiny clips doing their job...

    AWS installed LH 2.jpg

    There's a bit more AWS left to do; the receiver on the pony truck and the bash plate fitted to the buffer beam to protect the receiver from swinging coupling links.
  3. Threadmark: AWS bash plate

    Ian_C Western Thunderer

    And here's the bash plate. Kept me out of mischief after tea.
    AWS bash guard.jpg
    3 LINK, Mark F, Dan Randall and 11 others like this.
  4. Nick Rogers

    Nick Rogers Active Member

    Very fine work Ian.
  5. Podartist79

    Podartist79 Member

    Absolutely wonderful Ian!!
  6. Threadmark: AWS receiver mounting

    Ian_C Western Thunderer

    Last chapter on the AWS. Honest. In the gloom under the pony truck there's a bunch of bracketsmithery that holds the AWS receiver 6 inches or so above track level. It's 1960's thing and unsurprisingly drawings for this are not included in the Wild Swan book, so it's another estimate and approximate job. I imagined that photos showing the AWS receiver installation would be hard to find but 30 minutes of browsing 'The Book of the Stanier 8F 2-8-0s - Part one: Pre-War Engines 48000-48125", Irwell Press, ISBN978-911262-15-2 (splendid book if you're a Stanier fanboy or modelling an 8F) has some half decent views.
    ASW rcvr compilation.jpg

    There's enough here to work out roughly how it's done. (Note to self - must add the diagonal stays to the front footsteps - only just noticed them). Measured up the pony truck and modelled the parts in CAD to make sure it all ends up in the right place before cutting metal.
    The green receiver is the brass casting from Ragstone Models, the rest fabricated from odds'n'sods. Easy enough, and I didn't bother to add much detail because it's not that visible. Silver soldered together so it can be soldered to the pony truck as a sub assembly without it all coming unstuck.

    AWS rcvr sub assy.jpg

    And just to put it in context here it is compared with a glass of beer. Beer helps to put a lot of things in context.
    AWS rcvr sub assy beer.jpg

    Straightforward to fix it in position on the pony truck. The flexible conduit that carries the receiver wiring onto the loco was made by winding some very fine copper wire round a length of 0.5mm brass wire and tinning the lot with 145 degree solder. I used 145 because it's very fluid and doesn't clog up the detail too much. One end soldered into the receiver outlet and the other bent around to join the receiver mounting plate to make it robust. In reality I think the flexible conduit ended up fixed to the back of the bash plate on the buffer beam, clearly not going to work on the model!
    AWS rcvr mounted.jpg

    And here's the finished article in the Colin Gifford style...
    AWS rcvr complete BW.jpg

    Here's something I've never seen discussed - AWS battery maintenance. There was no means of battery charging provided on the locos so logically the batteries would have to be removed and recharged periodically. How long would a set of batteries last? Days? Weeks? There must also have been battery charging and storage facilities at some steam sheds mustn't there? You have to wonder how well looked after this electrical gubbins was on a typical steam shed.
  7. Threadmark: The smokebox and a load of difficult stuff - part 1

    Ian_C Western Thunderer

    One entry on the finishing off list - "speaker mounting and means of removal". I've kicked this one down the road for a long time, now it needs to be tackled.

    The original plan described many WT posts ago was to insert the DCC circuit board and speaker on a subframe through the underside of the firebox and push them forward inside the boiler with the speaker ending up in the smokebox. It became clear that the DCC module wasn't going to fit that way so it's been relocated to the tender (i.e. kicked a bit further down the road) , where there's enough room. On trying to accomplish the same with the speaker it became clear that that wasn't a practical approach either. Since I definitely want the speaker at the noisy end of the locomotive (human hearing is astonishingly accurate at determining the direction of a sound) the only other way to fit and remove the speaker is through the smokebox door. Fortunately I had not fitted the smoke box door at that point. You've no idea how much work this is going to create!

    There's a casting supplied in the kit that represents both the smoke box front ring and the door. Many posts ago I did some work on it to add the smokebox liner rivets. I didn't think too hard about it at the time. Since the smoke box door now needs to be removable the way this casting fits to the front of the smokebox becomes critical. And here's the first problem. On the prototype the smokebox front ring is a steel pressing that sits inside the smokebox wrapper plate and it riveted to it around the outer flange. What you see is the rounded corner of the ring pressing and the edge of the wrapper plate outside it. The plate is 3/8" thick on the prototype, equals 0.22mm on the model. The way the smokebox is made up on the kit presents double thickness at that edge: the base etch plus a half etched overlay. The resulting edge is way too fat and it inevitably shows the imperfect lamination of etch and half etch. Also the smokebox front casting doesn't quite replicate the prototype in the way it fits the smokebox. It doesn't quite slip inside the smokebox wrapper plate and it doesn't quite cover the front edge of it either, it's sort of undecided. There's one more thing that doesn't look quite right, the door seems a bit flat. More GNR Gresley than LMS Stanier. Hmmm...

    For me the smokebox front if the 'face' of the loco (although not in a Rev W. Awdry sort of way, and in passing it has to be noted that although Henry the Green Engine was clearly a black five the Fat Controller never saw fit to acquire an 8F) and it's easy to lose the character of it if something's not quite right. So it'll have to be rebuilt from scratch. Hadn't planned on this and it looks a bit daunting.

    The first job is the thin the smokebox wrapper by carefully (very carefully - one slip and it's a big mess!) undercutting the inside etch with a cutting disc and peeling out that thickness of material. Nerve wracking stuff. smokebox edge thinning.jpg

    After cleaning up the inside diameter can be measured and a smokebox front ring drawn up. I just copied the prototype using drawings in the Wild Swan book. The ring is easily, if rather wastefully, machined from a lump of brass, and the rivet holes drilled diameter 0.5mm on the milling machine.The inside of the ring is chamfered as per the prototype (the pressing was machined here to provide a seat for the door to seal on) so that there will be a small recess visible around the door when it's fitted. It's there on the prototype and almost impossible to replicate on the model unless you make the ring and door as separate parts like this.
    smokebox ring machine 1.jpg

    Back in the lathe to part off and you get one of these...
    smokebox ring machine 2.jpg
    The original casting is on the left. It's a decent casting, shame I can't use it. The MOK smoke box is formed around etched circular formers and as a result it comes out almost perfectly round. The machined ring is a gentle push fit into the wrapper with hardly any gap around the edge. Looks much better and worth the effort.

    Took a bit of thinking to work out how to make a new smokebox door. I measured the casting and compared it with the profile taken from the Wild Swan book. The cast door is definitely too flat. The dome of the door is a spherical surface apart from the flange at the outside edge. I imagine the traditional way of doing this (unless you have a CNC lathe!) is to mount a length of bar in the chuck and go at it watchmaker style with a toolrest and gravers and a profile template. Never tried that approach to turning and I don't have the tools for it. I came up with another approach that guarantees the geometrical accuracy of the profile and relies much less on hand work.

    First step is to draw the door profile in CAD. Then from the top centre of the curve step off parallel lines at 0.1mm intervals. You can use bigger intervals if you want to spend less time doing numbers on the lathe, but you'll have more hand finishing work to do and there's less control over the profile. Divide the door profile into what is effectively a stack of discs each 0.1mm thick and of a known diameter.
    smokebox door CAD.JPG

    That lot is tabulated in a spreadsheet so I don't lose track of things when I'm on the lathe. The lump of brass bar is turned down to the outside diameter of the door and faced off. A diameter 2.5mm hole is drilled through the centre for the dart boss. Then it's the best part of an hour twirling handles with industrial efficiency and trying not to be hypnotised by the DRO numbers - one cut too far and you have to start again.
    smokebox door machine 1.jpg

    You're sort of making a contour map of the door.
    smokebox door machine 2.jpg

    The surface is then taken down with file (yes - file on lathe = heresy) carefully until all the rings have just disappeared. A scrub with wet & dry (more heresy) and a Garryflex block finishes the door profile. It took a while but it was easier than I thought it would be.
    smokebox door machine 3.jpg

    The diameter is reduced behind the depth of the door to be a close fit in the previously made ring, and the whole thing is parted off.
    smokebox door machine 4.jpg

    The handrail knob holes are drilled on the milling machine and a small part is turned to represent the boss for the dart locking handles.
    smokebox door 5.jpg

    There's at least one more episode of this...
    3 LINK, eastsidepilot, P A D and 10 others like this.
  8. adrian

    adrian Flying Squad

    That does make a marked difference to the appearance - although I suspect that when I get round to my build I'll just cut a circular ring out of a thick sheet of nickel-silver sheet with a piercing saw and tart it up with a file rather than machining it. :rolleyes:

    I think you're being far too modest - you can make some good gravers from grinding up some old needle files so I'm sure a man of your calibre wouldn't have a problem at all. :thumbs:
  9. Threadmark: Smokebox door hinges - emotionally damaging and now a bit stuck

    Ian_C Western Thunderer

    I wasn't looking forward to making these parts, and they have turned out to be a pain. First attempts unsuccessful.

    I'd originally thought of soldering a length of tube to a strip and working from there. I actually made some parts to try before changing my mind and deciding to make them from solid. I reasoned that you can't actually hold and work on parts this size as individual items so I fixed the blanks to a sacrificial surface and worked from there.

    A couple of brass lumps are cut roughly to size and soldered to another piece of brass that's large enough to hold in a milling vice. I made a first attempt with soft solder but that failed on final machining when the milling cutter started to peel the workpiece off the base. For the second attempt the blanks were fixed with silver solder, and there were no machining problems with that.
    soldered blanks.jpg

    The lumps are machined away to leave the basic shape.
    machined 1.jpg

    Hinge cut outs are roughly machined (they get cleaned up later with an escapement file) and the hinge pin holes drilled through.
    machined 2.jpg

    The hinges are cut from the main base material to make them easier to finish by hand. Some careful filing to add the radii, and the hinge pockets cleaned up. Looking quite good I thought.

    It went horribly wrong at the next stage...

    The heat needed to re-melt the silver solder was sufficient to destroy the hinges. I tried to heat the base lump rather than let the flame contact the hinge itself. Even so. On one of them it seems like the brass has delaminated before the solder melted. I might have been better leaving them on the original base and heating from the rear. That's a few hours work heading for the bin:headbang:.

    I need to find another way of doing this, but I'm not in the right frame of mind to have another go at the moment. Any suggestions?
  10. Brian McKenzie

    Brian McKenzie Western Thunderer

    A shame you lost those parts, Ian. I suggest machining the hinge parts again, as you have done, but directly into the end of an upturned brass bar (e.g. 25 x 3mm or suitable size standing vertically). Then separate the completed part off with a flycutter in the shape of a lathe style parting tool.
    Slitting tool.jpg

    The third photo in your message above, shows the cutter is leaving coarse marks that suggests some process unhappiness. I suspect the cutter may not be running truly concentrically - or it's getting a bit dull. Either would cause undue pressure on a plain soldered joint.

    If you have a finger type d.t.i. (dial test indicator), it would be worth checking for runout, especially when using your smallest diameter cutters. ER collet chucks, or rather the collets themselves, are a bit fickle. Sometimes, just by tapping the closer nut at the side is enough to correct runout.
    Rob Pulham likes this.
  11. Ian_C

    Ian_C Western Thunderer

    Hmmm...never had a go at fly cutting of any kind. Not something I would have thought of. Something for the 'to do' list. The hinge strap is 0.25mm thick, do you reckon that could be cut off OK? Thanks for the suggestion.

    Re milling cutter - it's virtually new HSS and seems typically sharp. I think the problem is that I have a few small slot drills that have a geometry optimised for clearing chips on aluminium and other non-ferrous metals. The helix angle is quite shallow compared with normal and I guess that makes for a higher vertical component of the cutting force; more 'lifty' if you like. I'll use a different cutter if I try it again, and probably a higher melting point soft solder.

    Run out is good. One thing I've learned is that the smaller the ER chuck the lower the runout (usually). It is possible to mount these cutters with a 6mm shank in an ER32 6mm collet, but run out is often poor. These days I do all the small machining in ER11 or sometimes ER16 collets. I get a run out of about 0.1mm or less (plus I can see what I'm doing beneath the collet nut!). It's possible to buy higher precision collets with run outs of 0.05mm or less, but they're disproportionatlely expensive. My ER 11 collets are just 'standard', but maybe I'm blessed with a decent set. I can run 1mm diameter cutters in ER11 at around 2,00o rpm without much trouble now, though it took a bit of trial and error to get there.
    Rob Pulham and P A D like this.
  12. Threadmark: Reasonably close to a happy ending - and seeing what you're doing!

    Ian_C Western Thunderer

    Not sure where the line is drawn between determination and obsession, but I had to have another go. This time I went back to the plan A : making the hinge up from tube and strip.

    The strips are cut and filed from 0.25mm N/S, and the tubes are small brass turnings 1.28mm OD and drilled through 0.6mm for the hinge pin. Set up carefully and silver soldered with enough solder to form a fillet between tube and strip. The tube was deliberately slightly too long in order to form the fillet right out to the edges of the strip. It gets filed back to the correct length later. Contains adverts - join the Scale 7 Group!
    tube and strip 1.jpg

    The tube length is corrected, the hinge gap is cut out with a piercing saw and cleaned up with a small file. They're not quite as lovely as the machined from solid ones, but they're OK I think.
    tube and strip finished.jpg
    I get to sleep at night now!

    Here's something I discovered recently - watchmaker's eyepieces. They're really cheap (less than £3) and available in a variety of magnifications. Bought a few recently; x3, x4 and x10 magnification. I find myself using them quite a bit on small work - to see if I've filed up to a marked line or started a drill on centre etc. They're a bit cheap, they have plastic lenses, like something out of a Christmas cracker but they help you to see what you're doing and that must improve your work. At that price why wouldn't you have a couple on the bench?


    Rob Pulham, chrisb, Dog Star and 3 others like this.
  13. Brian McKenzie

    Brian McKenzie Western Thunderer

    The production of your new hinges (looking good) has just come ahead of my now redundant reply to your machining problem. :)

    Yes, parting off a 0.25mm thickness will present challenges. The amount the part will want to curl away from the cutter will be dependent on the crispness and (positive) geometry of the fly cutter.
    It might be preferable to flycut the hinge parts off at a greater thickness, and then solder them down face flat and mill off the back side. The soldering could be 'flooded' in this instance to provide plenty of support.

    > Hmmm...never had a go at fly cutting of any kind.

    Flycutting is a useful technique, especially where odd shapes or curvature is needed.
  14. Ian_C

    Ian_C Western Thunderer

    Sorry Brian! It drives me nuts when I can't do something. I just have to keep at it until it's done.
    Brian McKenzie likes this.
  15. JimG

    JimG Western Thunderer


    Looking at your problem, I might want to try orientating the part by 90 degrees along its longitudinal centre and cutting it out of a piece of sheet or strip which is thick enough to accommodate the height of the hinge. I have to admit that I'm looking at it from the point of view of a CNC operator so getting the shaping round the hinge would be easy for me, but not so easy when done on a manual mill. You might be able to achieve it with a bit of jiggery-pokery with a rotary table. I would also tend to take several cuts round the parts and not cut right through, but leave a few thou at the bottom of the cut. This would probably stop the thin hinge strap bending away from the cutter if you tried to do it in one cut right through. You can also drill the hinge hole while you are at it. Once the piece has been cut out of the sheet or strip, you can finish of with filing.

    Brian McKenzie and Dog Star like this.
  16. adrian

    adrian Flying Squad

    I don't know - they look pretty damn good to me. Sometimes I think we should consider how the original full size component was made. In this case I doubt they were machined in the first place but rather a made by a blacksmith and a forge so whilst silver-soldering a tube to the strip is not quite there I think it is more in keeping with the prototype than a machined item.
    daifly, Rob Pulham and Dog Star like this.
  17. mickoo

    mickoo Western Thunderer

    Nice work, all through.

    I find many solutions to tricky problems in the bottom of beer glasses, the really difficult ones often take three or four attempts! ;)
  18. Threadmark: 48142 has a face at last

    Ian_C Western Thunderer

    Having made some half decent hinges I'll be damned if I don't make an opening door. So, proper hinge pivots next.

    Starts with some small turned parts. The secret is to drill the cross hole first in the bar stock and then do the turning in the lathe so that it ends up in the right place on the finished part.
    door hinge turnings.jpg

    A bit of hand work next to make them square and round off the top. They're held in a an old Eclipse pin vice which in turn is held in the bench vice. A length of wire through the cross drilling helps align the part with the jaws of the pin vice which can be used as visual guides for filing square.
    door hinge blocks.jpg

    Working hinges without too much effort.
    door hinges work.jpg

    Takes a lot of parts to replace the single casting. That's 12 parts not including the lengths of wire for the rivets. I can see why a decent casting has some appeal in a kit. There's more to this than meets the eye though! The hinge pivot blocks had to be made twice. I miscalculated the hole position first time round and as a result the hinge axis was a bit too far from the smokebox front ring. Might have got away with it but decided not to. I recall bit of Tom Mallard advice from an old MRJ article - in essence 'one way to improve your work is to re-make anything that you're not happy with'. Something like that. Painful at first, but true. Also the handrail knobs supplied with the kit looked a bit lumpy compared with the prototype, and the handrail wire hole was a bit small. Like turnips strung on a washing line. The handrail actually scales to around 0.9mm diameter, bigger than you think. In 4mm the struggle is to make everything fine enough. In 7mm some things are chunkier than you think. Could have been stuck there but fortunately I found some splendid cast handrail knobs in an old MMP kit. (Diverging a bit - I bought a 7mm 9F MMP kit years ago when I really couldn't afford it. I thought about trying 7mm back then but apart from a couple of Slaters coal wagons but I never really got into it. I still have the MMP kit untouched. It's not state of the art now so it'll probably go on Ebay - minus handrail knobs! ) . All put together by a combination of silver soldering followed by soft soldering. Soldering the details to the lump of brass that is the smoke box door was a real trial. Not quite as hard as the firebox backhead - you'd have thought I'd learn from that, but apparently not. The dart handle and door clamping handle were cannibalised from a Laurie Griffin casting.
    smokebox door all parts.jpg

    Eventually it all comes together and 48142 has a face at last. I think it captures the Stanier look quite well. The smokebox front ring isn't yet soldered into the smokebox wrapper yet, just a push fit. That'll happen when I've made up my mind about speaker mounting and fitting the lead weights (earlier post - much earlier!). And the door does open on the hinges. Imagine the scenes when Mr Fireman swings open the door to shovel out the char only to be be confronted by a couple of tonnes of lead and a giant speaker.
    loco has a face.jpg

    Postscript - amazing what you see on a cruel photograph. The RH handrail end (button, stopper, finial - whatever) is slightly fatter than the LH one. Needs filing down a bit. Tomorrow...
  19. P A D

    P A D Western Thunderer

    Absolutely stunning work Ian!

    When Mr Fireman opens that wonderful door you have made, surely he will be confronted by the blast pipe, tube plate and all the other gubbins that you will be making to put in front of the lead and the speaker.

  20. Ian_C

    Ian_C Western Thunderer

    Ain't gonna be any space in front of the lead weights and the speaker! You'll see...