Wednesday, 10 January 2018

Riverboats on the Murrumbidgee

I have been asked about the paddlesteamer positioned in the picture at the end of my latest Blog post.   
Paddle steamers were once a common sight on the Murrumbidgee River in Wagga, although that was generally before the railway arrived.  According to Keith Swan’s book “A History of Wagga Wagga”,  the steamers “Corrong” and “Wagga Wagga” were employed to take railway construction material downstream to Narranderra (where the railway there was being built), and cut redgum timber back to Wagga in 1879, and 1880.  The transport of redgum sleepers from the Narranderra sawmill via the river continued  until 1914 according to Keith Swan.  Upstream activity to Mundarlo for limestone occurred occassionly.

Riverboat on the Murrumbidgee, believed at Wagga dock in the 18870s

You can still ride a commercial sight seeing boat on the Murrumbidgee twice a week  Unfortunately, it is an oversided “Tinny” rather than a paddlesteamer. 

Paddlesteamers can still be found on the Murray River, where one can get an authentic feel.  And if Echuca is too far, then the PS Enterprise is in Canberra on Lake Burley Griffin.  So it is not too far fetched that a paddlesteamer may make it again to Wagga.

Pevensey at Echuca Dock. VR Railway shed on the Righthand side of the picture
Steaming on the Murray. Note the water colour

PS Adelaide at Echuca in the 2010 flood

I have had an interest in paddle steamers for many years. When LJ Models brought out their kits of PS Adelaide, and PS Pevensey in the 1990s, I bought both.  Instead of building the kits, I used the plans to construct versions in wood.   Both my models remain uncompleted.
My unfinished scratch built paddlesteamers, PS Pevensey, and PS Adelaide.

About 10 years later, came the sad news that Fred Gill MMR, had passed away.  Fred was very  influential in the hobby, and many of his scratch building articles and  tips in early AMRM magazines are still inspiring.  (hint: search for his name in the on-line AMRM Index)  The NMRA was asked to find homes for his models, and the lists included the PS Adelaide.  The start price was reasonable, and I was successful in my bid.  I expected it to be the LJ models kit version, so I was very surprised when it wasn’t.   The model is  is exquisite in the detail.  
Unfortunately, there was some postal damage to the wheelhouse, and the stay support post, but these were easy to fix.  

Fred Gill's, PS Adelaide - with some damage to the wheelhouse, and stay post

After repair, I thought I would see the differences.  My model is based on the LJ Models kit, and is close to the present  configuration, (which I photographed in 2010 - picture earlier in this blogpost)

After repair. There are some differences in the two models.  I do not know where Fred got his plans from, and I suspect this will remain a mystery.  The general size of the lower deck, and dimensions is quite similar.  From a technical viewpoint, how does the captain, and his dog  get into the wheelhouse, as there are no steps to get from the lower deck to the wheelhouse deck.?  

So, as my tribute to Fred, I will proudly display his model on my river.  I hope you will all allow me this indulgence

Saturday, 6 January 2018

Murrumbidgee River bridge – part 4 – Track

Compromises are something we all have to deal with .  On a model railway, almost every decision is a choice between alternatives,  and  there is a trade-off between speed of construction, the size of the layout,  materials used, money spent, and satisfaction of the finished product.

Murrumbidgee river at low flow levels. Note the weathering effect of the river on the piers.

The Murrumbidgee River bridge is a model where I have tried to push my own modelling standards, at the expense of money spent, and time.  I am fortunate that I am not up to my armpits on actually building the layout proper at the moment, as the goal  to get trains running would have me working on other tasks.  The layout room build is still some way off. 

Even so, I am glad that my bridge model is not full length, as the construction of the shortened version  has become  a marathon.  The end is in sight.

The trackbed of the real bridge consisted of 2  metal girders running the full length of each span.  Sleepers laid across the top of these 2  girders.  The Uneek kit supplies 2 thin stripwood lengths to substitute for the lengthwise girders.  The assumption,  although not stated in the Uneek instructions, is that one will use flextrack for the bridge track.  Simple and quick.  Well, this was a compromise that I wasn’t going to take.  It looks wrong

I lacked measurements of the lengthwise girders, although my own detail photos gave a hint of the size.  I also needed to maintain the “loading gauge” clearances for locos and rolling stock, so they couldn’t be too thick.

 I checked the Greg Edwards data sheets (sheet P13) to get some ideas of the measurements on unballasted bridge decks.  The sleepers are spaced about 11mm apart for HO (just under 3 scale feet), and the sleepers are longer than sleeper length on ballasted track   Searching my supplies, I re-discovered a Campbell’s packet of 500 bridge and trestle ties, and rather than make yet another pair of metal girders, I thought the 10” square North Eastern stripwood would work.

Eight lengths of 10” square stripwood were painted grimy black to match the colour of the bridge spans.  Calculations showed just over 200 sleepers were needed.  I painted these sleepers white in batches over a few days.
Some of the timber after painting. I have decided to go with a full white sleeper, rather than just painting the tips. This may be a mistake (or own goal), but it is my layout, staining the sleepers first would have added yet more time to the project, and I am building the bridge for me.

Adding the sleepers to the span.  I found it easier to measure 88 mm (7 sleepers worth), and flue down one sleeper.  When this was dry, I would then halve the gap, etc, until 8 sleeprers were in place.

After the first base made, the spacing could be transfered to the second, third and fourth.

Fitting the sleepers to the girders is as pictured.  Make sure that everything is as square as possible.  I am using PVA glue as the adhesive.   I am also building this on a sheet of glass, to stop any twists, and bumps

The rail is stripped from Peco SL100X flex track   Yes, Code 100 – which is yet another compromise.  My experience with code 100 in the past with “12th Street Yard”, and “Yendys”  has been fairly positive, provided the rail is painted.  Removal of the shine on the sides of the rail is the key to hiding the rail height.   On the forementioned layouts, I sprayed Floquil “Roof Brown” on the track – including the sleepers, cleaned paint off the railhead, and ballasted afterwards.  Spraying the rail once it was down on my bridge isn’t possible, so the rail has to be painted   first.

Having observed handlaid track on many layouts, both in Australia and in the USA, shows that it is spiked directly to the sleeper base.  But to my eyes,  it doesn’t look right.  Something is missing?  Have you noticed?  Every wooden sleeper should have a pair of tie plates, that hold the rail in position.

The tie-plate pictured is 20 cm x 30cm in size.  In HO, this is around 2.3mm x 3.5mm.  As the peco code 100 rail track base is marginally wider than scale, I thought I would go with 2mm x 4mm.  I thought about many methods to simulate this on the sleeper.  In the end, I painted  Floquil roof brown onto white paper, sliced off a 2mm strip, and chopped the strip into 4mm lengths.    I have placed no other detail on them.  A spot of glue on the sleeper, and place the paper tie plate in roughly the right position, adjust, and wait for the glue to dry.   Hint, use the tip of a sharp knife (I used a #11 surgical scalpel blade) to spear the paper, and then move it to the glue spot.  The glue will pull the paper off the knife tip, and the knife tip can then align the paper.

Paper tieplates added to the sleepers

The track to the right has the tieplate detail, to the right without for a direct comparison. . Make up your own mind to whether this detail is worth the effort?

I thought I would see if the effect is worth the time and effort.  It is subtle, but I do like it.  What is also apparent, is that the tie plates are not consistently positioned on the sleepers.  I will have to adjust my technique in the future, as the viaduct track will not be hidden by the bridge girders. 

Rather than spiking the rail to the sleepers, I used Techgrip Polyurethane  adhesive I bought from Bunnings, and smeared it on the rail underside. With  track gauges to hold the rail at the correct gauge,  apply weight.  The glue goes off slowly, allowing fine adjustments in the first 15 minutes .  Important to try and  get the new track centered exactly over the sleepers.

Checkrails.  I have always thought that raill bridges should have checkrails, but the end view of the cut-up bridge at Wagga only shows the main running rails.  However, I was sent a picture of the Whitten truss bridge at Woolbrook, and it has checkrails.  

Woolbrook Whitton Truss bridge - single span.  the checkrails positioning does not follow the Greg Edwards Data sheet, so this pattern won't be followed

Interestingly, the rail appears to be spiked directly onto the sleepers, with no tie-plates.  This is good, as that means that the check rail will be slightly lower than my running rails, so they may not lose the paint during track cleaning.. The check rails were painted, bent to shape, and glued  between the main running rails

Attach the track base girders to the Whitton truss with adhesive.  In this case I used PVA glue.  The glue takes a few minutes to go off, which allows for adjustment, but also if I ever need to remove the bridge deck, then the glue can be pulled apart without undue damage to the bridge deck or the bridge. Again, make sure the rail is centered. This is the last chance to get this right  

The bridge deck extends onto the future viaduct.  

Fishplates are there as a aligning point, where the bridge spans meet.  I have another sleeper (not shown) to fit under the fishplate when the bridge is finally installed  on the layout

I have fitted  fishplates to the end of one bridge span, as I want to be able to separate both spans for storage.    My goal with this series of Blog Posts was  to complete the 2 main spans, and lay track.  Actual installation on the layout is in the future, as the process to create my train room out of half my garage is ongoing.  Decisions on the river depth will then be made, installation and weathering of the plaster piers, as well as the construction of some of the connected viaduct.

It was 43 degrees C outside, and a wind blowing in Junee, so I took these pictures inside. The plaster piers have been painted since the initial part 1 blogpost, and they are just holding the bridge up by gravity.  The "Adelaide" paddlesteamer was built by the late Fred Gill MMR, and whilst well out of time period, it seemed too nice a model to not use.

Some colour at the end.  Loco and carriages all fit under the overhead girder sections.  Whew!!,

Until next time,  build a model or two.  Happy modelling

Wednesday, 27 December 2017

Murrumbidgee River Bridge - Part 3. 2nd span, and painting

Modelling can be very tedious at times.  Having finished the overhead beams on the first span, the second span was just a long slog on the workbench.  As I was now confident on my technique, the remaining girders were mass produced, and then fitted to the second truss span over a period of 3 weeks 

The Murrumbidgee River flooded last year - this is a picture taken of one of the billabongs during the flood, Note the gum trees, as this is some of the effect I hope to have around the bridge model.

Girders ready to be installed.  The above represents around 12 hours of workbench time

The instructions of the kit recommended dunking the completed bridge span into paint, and allowing it to drip dry,  Whilst  this had the advantage of speed, I have doubts over this method, as well as having sufficient quantity of paint.  In any case, the brass had to be cleaned of flux, fingerprints, cobwebs, and anything else that shouldn’t be there.

Rinsing under the tap was all I had done so far, but evidence of green verdigris in certain areas, showed that the tap  does not get rid of all the flux.    So the search started for a suitable container where I could scrub with a toothbrush, and use chemicals if needed.    Each of the spans is around 50cm in length.  To cut a long story short, (and no, the house bathtub is not suitable),  I made my own out of timber, and plastic film.  My inspiration is thanks to a display of R/C boats that I saw at the Newcastle model exhibition many years ago.  They had made a large shallow pond  on the floor of the exhibition hall using black plastic and bricks

The timber is scrap I happened to have lying around  

The plastic is double skinned to prevent leaks in case the upper plastic film is damaged.

Scrubbing was with cream cleanser, and toothbrush.  Followed by a rinse.  Then repeat.  Check for any verdigris, and if present, break out the ammonia and salt solution.  Wear gloves.  Rinse again.

After cleaning, the next stage was to attach the rivet shim to the upper, and lower chords of each span.  The shim needs to be carefully cut to size.  Checked for fit, I then glue the shim to the span.  Why glue?  Applying heat in the form of a soldering iron will cause the shim brass to expand, and then contract whilst it  cools.  I am not skilled enough to be able to control this, and any distortions in the rivet overlay will be difficult to correct.

Underside of the span, showing the rivet overlay.  Each of the main uprights are secured to the span with a pair of triangular gusset pieces made from scrap brass.

The glue I am using is 6 minute, 2 part epoxy from Great Planes in the USA.  Mix 2 parts equally, smear it over the bridge girder, and attach the rivet overlay.  Weight it down.  After about 10 minutes, remove the weights, and repeat on the next rivet section

Prior to painting brass, I like to provide a key for the paint with a light vinegar etch   The term used is “pickling”.  The vinegar is a mild acid, and the process gives a microscopic roughness that allows better adhesion of the paint.  My homebuilt container was half filled with fresh vinegar – about 1200ml.    As the bridge span is not covered by the vinegar, I am turning the model every 15 minutes to a new side.  You may notice a slight “pink” tinge on the brass after the pickling process, and this is a good sign.  After each side has been treated,   rinse the span and allow to dry

I am hoping that the vinegar treatment would allow me to skip the “self etch black” primer coat, and go straight to the final paint coats.  The self-etch black is not the easiest of paints to spray, and the bridge is rather a big model  The paint I am using is Floquil “Grimy Black”, thinned with turpentine, , and sprayed with a Badger 200 airbrush.   Pretty nasty stuff, so make sure that things are well ventilated.   Note. I know I should be using Xylene as the floquil thinner, but my experience with that product, besides expense, is that it dries far too flat. The turpentine , if anything, takes a bit longer to dry, and dries with a slight satin sheen.  And the other question is why Floquil?  Well, I have accumulated a supply of this excellent but now discontinued product, and to not use it would be rather sad, particularly as it sprays well onto brass, and gives a nice finish.

The main problem (other than the size of the span) with painting the bridge is the space between the 2 lattice etches on each side.  One had to try and avoid getting a shadow mask effect, so the airbrush angle of attack needs to be varied with each pass over the bridge.  My suggestion is that you do not attempt to complete the full span in one spray sitting – after you think you have painted the lattice, then allow the span to dry, and inspect for imperfections in a bright light.  Once you are happy with the lattice, then paint the remaining beams. 

One span painted.  I am happy with the colour, as it is close to the colour I remember on the real bridge.  The second span is still to get the pickling treatment, and then the paint. Note too, how the painted span only has 3 overhead support trusses. .  This is because the real bridge "shared" a support at the end of each span. .

The second  span should be a bit easier, as I had previously painted the insides of the lattice black (see part 1).  I will have to give the first painted span an additional coat, because, as expected in good light, there some areas where the brass colour  is showing through.  However, the effect of even this initial paint coat  is magic.  It is good to get a few wins on the way

The next part will be the wooden track support, and track.    

Until then, happy modelling

Saturday, 9 December 2017

Murrumbidgee River bridge – part 2 – the overhead support beams

I am not sure when the Whitton truss bridge across the Murrumbidgee River at Wagga had its overhead arch girders replaced with steel beams, but it was before 1970.  And I do not know the reason, as the double track Whitton arch bridge at Albury over the Murray River still has its arches

Shortly after being built in 1879, this NSW Archives picture shows the Wagga Wagga bridge, and the timber trestle

In any case, I needed to model the Murrumbidgee bridge as it appeared in 1970, and that meant scratch building the overhead “U” support beams.   The first part of the Murrumbidgee bridge blog post is way back on the 26th June, 2017.  I finished that blog post saying that I was waiting for ABS styrene from China, or a source of brass “C” channel or the correct size.

The bridge as it appeared in my June 2017 blog. 

To cut a long story short, the styrene “C” channels from China were substandard, and I was unable to locate the correctly sized brass “C” channel.  However, I don’t like to leave projects sitting around for too long, and I wanted a distraction from the pain of shingles (yeah, that is another story), so a start was made. 
The styrene “C” channels, whilst poor, could be modified.  I cut out 7.5 cm sections that were more or less straight from the 50cm lengths supplied; and reduced the height down to approx 1.5mm.  These I put into a jig to hold them at the correct distance apart, whist I glued the styrene spacer plates on.  Unfortunately, removal of the jig stressed the joints out, and the whole lot disintegrated.  Better to find that out now, than after I had spent any more time on them.

The failed girder attempt with styrene

As no suitable brass “C’ channels were available. I had little choice than to fabricate my own from strip brass.  In this case I was very lucky.  I had in my supply of North Yard brass strip, enough 1.5 x 0.25 x 150mm, and 3.0 x 0.25 x 150mm strip pieces  to fabricate the 3.15 metres of the “C”  channels of the correct size for the bridge.  Yes, that is a lot of brass strip

Parts needed to make one 150mm strip.  This completed strip will be cut into two 7.5cm sections. This is a fortunate measurement, as it eliminates waste of the brass.

Once the “C Channels” were made, I fitted them into the distinctive “U” shape of the prototype, and matched a pair back to back, and soldered spacers to hold them a scale foot apart. 

Three 7.5cm sections make one half of the "U" beam

One completed "U" beam.  The spacer plates are 3mm strip brass.

  All these measurements have been guesstimated from the prototype photos I took during the bridge replacement over a decade ago

New concrete support piers were formed whilst the Whitton Truss bridge was still operational. The actual replacement of the bridge took around 3 days. The spans of the truss were cut into sections, and removed by crane. The old piers were then removed, and new concrete beams were laid onto the new concrete supports, and the track was then re-instated.

After being gas axed, the bridge truss sections were craned to one side awaiting removal. This view shows the interior construction of the girder quite well.

To match the prototype, I will have to fabricate 7 “U” beam supports, 4 on the first girder, and 3 on the second. 
 “U” beams were secured to the bridge using triangle gusset shapes.  These were made from sheet brass, but soldered to the bridge with 144 degree C solder, rather than the 60/40 solder I had used earlier.  I did this for 2 reasons – a)  I did not want to desolder the assembled “U” girder upright, and b) 144 degrree solder can flow with my small 25 watt iron on the large heatsink of brass that is the Whitton truss bridge side

The "U" beam fitted to the end of the bridge. The bottom had been secured to the bridge via a brass plate, although the triangle gusset plates on top are still to be fitted.  I was also concerned that I may have reduced the clearance too low, but I tested it with a USA outline diesel, and things should fit, provided I use lower profile sleepers.  But, I won't be able to run double-stacks through the bridge, 

One span, with two "U" beams fitted, and braced, and after removal of the temporary brace

The result was very strong, and I felt confident to remove the temporary brace, designed to stop the 2 bridge sides from collapsing inwards with handling.
I made a minor change to the assembly order of the subsequent “U” girders.  Basically, the 2 uprights were installed, soldered, and braced, before the overhead beam installed.  This was a lot simpler than my earlier method.

Detail soldering of one of the venter uprights

Angle braces were fabricated, and soldered to all 4 “U” beam assemblies.  These are purely cosmetic on my model.

I have been using fluxes to aid the soldering, and these need to be removed.  I have scrubbed the girder with kitchen cream cleanser, and washed the lot in water.  Further treatment will be needed prior to painting, but at least the model is no longer a clothing stain hazard. 

I have not yet finished the first girder.  The distinctive rivet overlay supplied in the kit will need to be cut to size, and glued [soldering heat would cause distortion of the thin rivet overlay]   to the flat brass on either side of the center uprights.  In some ways these uprights actually correct a problem in the kit, as the rivet overlay is around 6 mm too short to cover the area on the upper, and lower bridge chords where it was expected to go. I do not know if this was corrected with the newly re-released kit

As you can see, I still have to fit "U" girder supports to the second span.  Followed by painting, and then the sleepers and track can be added.  

This process is very time consuming.  I have not added up the hours I have spent – but my last “U” beam assembly took just over 5 hours to fabricate, and add.   I am sure that talented modellers might have used 3D printing to make the “U” beams - I am not in that category

Happy to report that after 3 weeks, the shingles pain has much reduced.  Whoever said Model Railways is therapeutic must have had me in mind.  I will continue with fitting the “U” beams, and with luck, will have painted the bridge for my next bridge blog post.

Happy modelling 

Thursday, 23 November 2017

A Yard Crane for Tumbarumba

Way back in 1993, Stephen Johnson Models were offering a NSWGR 5 Ton yard crane kit for $55.  “Just what I need” when I bought the kit, although in 1993, I had little idea that it would take me another 24 years to actually build it.
Photo of the yard crane at Tarcutta (photo supplied by Pauil F)

The kit consists of 2 small etch brass frets, some lost wax castings, wire, rail, angle brass, 4 resin blocks, a set of instructions, and 3 resin jigs to assist in the construction.

  A review of the kit was in issue 179 (April 1993) of AMRM. (Thanks James).  This kit has a reputation of being difficult to build. The reputation is well earned.

As is my normal approach, I try and find as much information as possible.  Unfortunately, I had failed to photograph the full crane at Tumbarumba when I was there in 1980.  Greg Edwards Data sheets do have a HO diagram of the crane, but the details of the winch, and gearing was unclear. All is not lost.  These cranes can still be found in various locations, so  made a 55km trip out to Ganmain a few months ago, and photographed the 5 ton crane there.

Ganmain 5 ton Yard crane

After I started the build, Paul F sent me a good picture of the Tarcutta 5 ton crane pictured above.  As Tarcutta is on the Tumbarumba line, it might have been similar to the Tumbarumba crane. Disappointingly, the picture shows the “other” style of 5 ton crane used in NSWGR, which is a hint that the Tumbarumba Crane might also be this style, and not the kit version. But, these are the risks one takes.

Bob S (see the SCR blog link), also sent me some scanned  diagrams of the NSWGR crane from the railway archives.  These diagrams showed the winch and gearing in large scale.   These diagrams proved invaluable. Thanks Bob

Use of the jigs to build the “A” frames

Lower winch attached to one of the A frames

Anticipating a future step, I made winch spool rotatable.   However, I ran into a lot of problems in getting the top beams, and A frame assembled.  In retrospect, I should have made a jig to hold both A frames vertical, and not skewed.  Once the A frames are firmly held in correct alignment, then solder the pair of beams on top, and then add the cross bracing.  I won’t go into the method I used, but it almost resulted in the whole lot being launched into orbit
Even the bracket for the upper gearing was a significant exercise in alignment of the parts. 

Crane basically assembled.  Still have the set of steps to add to one of the A frame pillars, and the attachment of the ropes, and chains

Attaching the ropes, and chains was at best frustrating.  The instructions advise twisting the fine copper wire supplied together to simulate the chain.  This was straight forward, and from a distance, doesn’t look too bad.  Looking at Bob’s supplied plan, the upper chain goes from the trolley, and winds above the pulley, then go over the trolley to the other end, when it returns to the trolley.  (The instructions suggest the chain should go around the pulleys in the other direction, and the chain is hidden behind the top beams).  The chain descending to the bottom was also just a matter of getting the length right.  However, the brass wire supplied for the winch to the hook gave me lots of problems.  Basically the wire supplied was too hard, and fractured easily.   Solution straight forward.  I substituted some softer “tinned”copper  wire –taken from stripped insulated hookup wire.  This wire bent easily, and allowed me to wind it around the winch spool

Finished, painted with sellf-etch black, and weathered with my pastel rust powders.  Although the chains might have to be reworked if I find chains fine enough

Now that I have this crane built, I discover that there was also a crane at Ladysmith.  So unless another SJM kit appears, I might just have to throw money at the problem, and acquire the 5 ton crane model being sold by “The Model Railroad Craftsman” at Blacktown (even though this has an error in the bracket for the winch being on the inside of the “A frame”, and not the outside.  Who is going to be this nit picking?).  

Picture of the MRRC crane