Time Out

Written by Philip Light

This has been a crazy last few weeks!  No time for modeling In a while now due to my work load and most recently the need to move.  My wife just accepted a job that requires us to relocate and there just wasn't enough time to buy a house before the school year starts.  So we're renting a condo for the next twelve months.  As such, I won't have a workshop for at least a year.  So as of now (8/2/15) light scale models is officially on standby.  I will update when I finally get a shop back up and running.


Tender Update

Written by Philip Light

Work has been slowly progressing on the Class 56.  I've been focused on building the tender to get it out of the way.  I've also started machining the driver centers and fitting the tires.

For riveting on this project I'm using a technique/tooling that I haven't seen before.  I turned a rivet profile tool out of tool steel on the lathe and chucked it in the cnc mill.  I mount the part to be cut out to an MDF carrier and create a toolpath in my cam software to punch the rivets, drill holes and finally cut out the part.  I'm doing the tender wrapper in .005" thick brass sheet and the punch code in CAM does a -.0025" travel in the z axis to punch the rivets half way.  This process works extremely well with accurate rivet spacing and much better alignment than I could achieve by hand.

Here's a shot of the cnc mill punching the rivets for the tender top.  Notice how closely spaced the rivets are!  This would be near impossible without a special jig and a TON of time to do this by hand/manually.

IMG 6536

Once the "punch" operation is complete, I next fit the appropriate drill in the mill to drill the holes for the mounting brackets, hand grabs, tender fill etc.  Finally, a 1mm diameter end mill is chucked and the mill cuts out the part needed.  I like the results better than photo etching the parts.  For one, you get a proper profile to the rivet rather than an etched cylinder look from etching.  Second, you don't get the undercut/sharp edges from the photo etching process.  Third, you have a clean brass surface rather than the matte etched surface.  Fourth, you don't have to wait weeks to get the parts!

Here's a shot of all the major tender wrapper components after being processed on the mill.

IMG 6544

The tender base (second from the left) was cut out of .010" thick brass to provide more structural support to the base and provide a good sturdy substrate to solder the sides to.

One of the most challenging pieces of the tender is the flare at the top.  To model this, I made a 3D model of the tender and modeled the flare at the correct angle and profile.  I exported this 3d surface to a program used to make paper models from 3D models called Pepakura.  The software flattens a 3d model and creates a pattern to cut out.  Here's a shot of the paper pattern before being cut from brass.

IMG 6543

The software adds tabs and a fold/bend line where the tabs can be folded.  Figuring this pattern out by hand or even in cad alone would be a very difficult task!  By modeling it in 3D and using the paper model software, I was able to create this pattern in about 20 minutes!

The tender frame was fabricated using brass "C" channel with the end beams cut from .060" thick brass on the cnc mill.  

IMG 6562

The center beam is a brass "I" beam or H profile beam.  The plates on the ends were cut from .005" thick brass sheet with tiny stainless steel rivets welded in.  The side "C" channels had the open side facing the center of the frame and I cut .005" thick strips to enclose the open side of the channel.  This makes the sides look like they're solid when in fact they're hollow.  Precision Scale NBW's were added along the side frame pieces in cnc drilled holes.

On the under-side of the frame, I turned truck bolster bearing plates on the lathe and soldered them on the bolster straps.

IMG 6573

In the above shot, you can see the floor boards for the tender.  I cut each one from brass strips which are extremely close to the thickness they should be on the prototype.  They were grained with the edge of a razor saw to give the effect of wood grain.

The following shots show the progress of soldering up the tender wrapper components.

IMG 6550

The bends on the engine side were carefully rolled to match the curve of the base plate and top.

IMG 6555

The above shot shows the major pieces fo the tender in place and soldered up.

IMG 6567

Above is the flare in place.  Note that beading was added around the top of the flare!  This is thin gauge brass wire used in jewelry making that I picked up at the local craft store.  Soldering this thin wire around the compound curves of the flare was a challenge to say the least.  This process would not be possible (without a lot of swearing) without a resistance soldering unit with nice tweezer style tips!

IMG 6566

Above is a close up shot of the compound curves in the flare.  You can see that it flares out in both directions while wrapping around the curve!

The tender floor boards were soldered on after being grained on both sides.  I spaced them as I've seen in some photos of these engines.  The spacing didn't appear to be consistent on the prototype and so I replicated that on the model.

IMG 6700

Shot from below:

IMG 6701

In the bottom left you can see 2 round discs with NBW's on them.  These are the ends of the tender air tank.  Rather than buy a casting, I decided to make these per the prototype photos I have.

Here are a few tips;

  • You'll note that I have the tender sitting on a piece of shiny metal.  This is a slab of 1/4" thick T6 aluminum.  When soldering on a major assembly like this, I like to keep the model on a slab of aluminum.  The aluminum wicks the heat away from the model quickly.  I also use this slab of aluminum when soldering in general and especially on tiny pieces.  I'll solder on a piece of wood and then quickly move the part to the aluminum to cool it down.

Here are a few shots with some tips on building an air tank.

Soldering up a wrapper in a cylinder form can be a challenge.  I calculated the wrapper dimensions in cad by checking the circumference of the tank diameter.  I punched the rivets using the same cnc technique as above on the major tender sides.  I turned a solid brass cylinder the inside diameter of the tank.  Next I wrapped the rivetted wrapper around the cylinder and used steel floral wire to tightly secure the wrapper to the cylinder.  

IMG 6578

This is an extremely useful trick!  Twist the steel wire down to tighten the wrapper to the inner cylinder.  This same technique is useful for boiler wrappers as well.  Rather than use a solid brass cylinder inside, you can use rings turned on the lathe or heavier gauge brass tube or use a ferrous metal like steel or non-ferrous aluminum (basically anything you can't solder to) as an inner form while soldering the wrapper.

Once the part is soldered, you can remove the wire and clean up the part.

IMG 6580 

IMG 6585

Finally, a quick shot of one of the driver centers chucked in the lathe for processing.  I machined a fixture to hold the center while machining.  These centers were bought from Bill Meredith from the Leadsville Shops.  Beautiful castings!

IMG 6561

The fixture is basically a brass rod turned down slightly smaller than the inside diameter of the tire with 2 holes drilled and tapped for .5mm bolts which correspond with the exact location needed to hold the center tight on the fixture through the gaps in the spokes.  There are actually 2 of these fixtures.  One has the center drilled out so that the axle hole can be drilled/reamed to the proper diameter and the other has a piece of slightly under axle stock friction fit in it so that the axle hole can slide over it perfectly without forcing the hole open any more.  This 2nd fixture (shown) is for turning the outside diameter down to the proper diameter for the tire.  The fixture can also be flipped for facing off the back of the center so that it can fit flush and true on the fixture for turning the outside diameter.  It's quite a lot of work to process these precisely but well worth it to prevent any kind of wobble in the mechanism.

I've still got quite a bit to do on the tender.  Mostly piping the air lines, installing the brake hangers and hardware, the fill hatch, rear coal bulkhead, front coal bulkhead, fabricate the toolboxes (will have working hinges and have tools inside), add the valves and drain cocks, coupler, engine link and the small strap that covers the edge of the floor boards.

I intend to finish the tender completely before moving back on to the locomotive.  While the list seems long, the tender shouldn't take a whole lot more time.  I'm anxious to get the locomotive frame finished and a rolling chassis built.  The engine will take quite a bit more time.  I'm focusing on fabricating as much as possible for the engine so that it has a consistent, hand-built look.

Thanks for looking!




Intermission for Etching

Written by Philip Light

Just a quick post today about my adventures in building an Sn3 BLW Class 56.  It's been about 2 years since I photo etched brass in my shop.  I used to have a ferric chloride setup with an aquarium bubbler and heater in a plastic tank.  I never really had a lot of luck with it.  I got some usable etchings out of it, but it usually took hours and getting the resist just right was a real pain.

I'm trying to build as much as possible on this engine in house without using commerical castings.  That said, I did decide to buy some PSC castings for the project.  While I wait for them I've got a big list of things to do that I can proceed with in house.  One of those things was etching the builder plates.  I was going to send these out to be etched but I came across an article about using muriatic acid and hydrogen peroxide to etch circuit boards and I thought I'd give it a try.  I didn't set out to etch through the sheet, only etch enough to get a good relief to represent the plates.  I mixed up a batch of acid and put it in a small tupperware container suspended in hot water.  I used press-n-peel for the resist and designed the pattern from an 1883 Baldwin builder plate.  The locomotive I'm modeling is Baldwin serial number 5185 and was built in 1880.  I assume the plate design is pretty much the same between 1880-1883.

Here's a shot of an 1883 builder plate:

builder plate

Here's the design from AutoCAD


I realize the font isn't right and it's a little "off" here an there.  Like I said above, I haven't etched for about 2 years.  In that time I forgot a few things.  Mainly that the scale pretty much renders most of the stuff illegible.  I didn't really forget this as much as I wasn't sure how legible it would be.  I did, however, forget that you have to make the resist mirror image of what you want!  Thankfully, my first "oversight" countered the second.  The resulting plates are backwards, but they're so small that I think they're ok.  You can almost read them.  I'll try not to forget on the next etching job.

SN3 builder plates

I will say that the Muriatic acid/hydrogen peroxide solution is better than ferric chloride.  These plates etched to an acceptable level in about 15 minutes.  These are .010" thick brass.  The resist is really still the major hurdle.  I did a sheet of 16 plates and I got about 6 usable etches out of it.  That's been my experience with the stuff anyway.  I've tried (in the past) all kinds of variables to try to get a better resist application, but ultimately, it's just too difficult to get consistent results.

Regardless, these pass my QC test and will go on the model.  As you can see, they can't really be read.  They actually might be legible if they weren't mirrored.



Building an Sn3 Baldwin Class 56 from scratch

Written by Philip Light

I've decided to document this build in a blog series.  Hopefully I can keep up with posting as I build.  This project came to life in an unusal way.  As I was clearing out a bunch of parts, I noticed I had a set of 8 Sn3 drivers which would build an early C class or Class 56/60 locomotive.  I had actually bought these to see if they would work as 48" in Hon3, but the flange profile is for 1:64 and is just too big to be correct in 1:87.  I've always liked the early C-16's and Class 56/60 engines and decided I would build this engine as an "as delivered" Class 56.  As I started researching the engine, or which engine to build, I discovered some photos and plans for a Class 56 that was sold to the Silverton Gladstone & Northerly RR from the D&RG.  The engine spent the first few years of it's life on the D&RG, was shopped and sold to the SG&N to haul ore out of the mountains.  Named "Gold King", I couldn't resist making this my subject.

As with most of my current projects, the first step is to layout the model in AutoCAD to the correct scale and start modeling the components in 3d.  This obviously starts with the frame rails.  Below is a shot from the CAM software used to design the toolpaths to cut the frame on my cnc mill.

frame model

Next the frames were cut from a piece of .125" stock.  The entire frame was cut using a 1mm 4 flute endmill.

cutting out frame


With the frames cut out, I milled out the journals from the same stock.

cutting journals

Shot of the finished frame pieces and one of the journals.

Fresh frames

Next I modeled some of the detail parts.  Sometimes it's easy to get caught up in the 3d modeling/cnc milling process and do too much where traditional scratch building would be sufficient.  These are the parts I decided would be best to be cnc cut.

frame details

I like milling thin stock on a sacrificial piece of MDF with spray adhesive (the tacky repositionable kind) then clamps on the sides.  I've found if you don't use some kind of adhesive, the center of the stock can bow and/or the parts can become dislodged during cutting.  Note the one link in the center only has a support on one side.  The spray adhesive was enough to keep it stuck to the mdf during cutting.  A lot of this stuff would be good candidates for photo etching, but with a cnc mill in house, it's sometimes just faster to cut them this way.

Next I machined the frame to hold 2mm NWSL springs to give the engine real suspension.  After drilling the frame for the springs I soldered on the journal binders, drilled through the binders and frame w/ a .68mm drill to a depth that would allow me to tap the holes, then drilled the binders with a .83mm clearance drill so the .8mm bolts would pass through the binders and screw into the frame.  Then I unsoldered the binders and tapped the frame for .8mm hardware from Scale Hardware http://www.scalehardware.com/.

micro fasteners

I highly recommend all of scale hardware's micro hardware!  Especially the wrenches!  Do yourself a favor, if you're going to order micro fasteners, don't assume you can screw them in without the wrench.  It "IS" possible, but you will likely loose some and probably loose more hair!

micro wrench

With the machining/drilling/tapping done on the frames, it was time to start applying fiddly little bits to the frame.

frame with details

Mostly done with the top frame and I still have to start on the bottom.

So that's where I'm at currently.  The next step is to machine out the cylinder saddle/steam chest and machine the cylinders.  Then (or at the same time) the rear frame/cab support.  This will set the frame spacing and allow me to move on to the motor mount, gearbox and start working on getting the chassis rolling.

Stay tuned...




Gem/Seiko EBT #18 Complete

Written by Philip Light

EBT 18 black and white

The EBT #18 project is complete!  The process of fixing up this less than stellar model was good experience.  While I did fix some of the issues, ultimately, there are just too many issues to overcome.  Hopefully the finished model is at least an improvement of this particular specimen and it results in a longer life for the model.  I did acquire a Hallmark model of #18, which rendered this model somewhat redundant.  Finding and acquiring the Hallmark model took the wind out of my sails for this one.  It was fun to paint and fix it up, but if it were mine to keep I would have gone all out and rebuilt it all correctly.  At the end of the day, this became a model to sell, and as such, got enough to help someone else along and not much more.

These models really aren't that terrible compared to the Hallmark version.  They could be fixed up to be just as nice, but with a LOT of work.  You can't beat them for the price though!  Finished model photos below.

IMG 5398

IMG 5399

IMG 5407

IMG 5411

IMG 5417

IMG 5426


Continuation of GEM improvement project

Written by Philip Light

I've finally got some time to post an update on the GEM Large Mikado improvement project.  I'm happy that I'm finally close to paint on the model and I'm much happier with the model now than when I first bought it.  The next piece that I worked on after the piping on the fireman's side was the backhead.  As shown in the last post of the original model, it came with a giant open frame motor hanging out of the back of the cab.  This is, of course, unacceptable by today's standards and needed to be corrected.  I fabricated the new backhead to fit the space required which was quite a challenge considering that there is effectively no way to remove the cab on the model.  I seriously considered unsoldering the cab back to fit the backhead, but decided against it in the end due to the amount of solder used and the difficulty in accessing the areas that needed to be unsoldered.

EBT 18 backhead


I wasn't looking for a perfect representation of the backhead, but something would be better than nothing which is what I was left with after removing the bomber of a motor.  The above parts are from PSC and the backhead itself was simply cut from .010" brass sheet stock to fit the back of the boiler.  It was at this point that I decided I needed a new cab back for the model as the one supplied wasn't correct and really wasn't acceptable.  I didn't want to simply photo etch just the cab back, so I arranged a few parts to improve the model.  Below is the resulting sheet, which is available for purchase in the shop HERE if anyone needs one.  

IMG 4204

The sheet includes a new cab back which is drawn from plans of #18, 2 new windows for the front of the cab since the original model didn't include them and the remainder of the parts are to add detail to the rear of the frame under the cab.  The frame under the cab on the Gem/Seiko model is completely void of detail and is in fact incorrect.  The prototype has a widened frame at the rear and while the photo etches don't accurately correct this error in the model, they do imply the detail, which is enough for me.  Below are photos of the frame components in place on the model.

IMG 4200

Note that the motor mount (if it can even be called that) is only temporary to test the running gear.  It will be replaced by a proper mount later.  With the frame pieces added I moved on to completing the piping on the engineer's side of the model.  This was a much more complex set of piping and I couldn't find very many photos of the prototype to confirm the actual arrangement.  Since 18 is the only engine that I don't own a Hallmark model of, I had to go by the few photos found online and a bit of "artistic license".

IMG 4202


Finally, the backhead and cab back were fitted to the model.

IMG 4207


The before and after:

before after

All that's left on the back of the model is to fabricate new grab irons for the cab.  The wiring is from the headlight bulb which is part of the final bit of work on this model.  In addition to lighting, I'll be installing a sound decoder in the tender and wiring everything up.  Prior to the DCC/Sound install, the model will get one last shot of aluminum oxide to blend the newly added brass with the original model and get rid of any dirt or grease that may have been picked up by handling.  The DCC/Sound install will then be carried out with nitrile gloves to prevent oils from  collecting on the model prior to paint.  The next update should carry through to the finish line and include the photos of the finished model.

While this post isn't about #14, I thought I would throw in a photo of the paint job I just finished up on her.  I received the model with flaking old paint that was wearing off very easily.  The model was stripped to bare brass and re-painted/weathered as shown below.  The same treatment will be done to #18 so that the fleet looks somewhat uniform in appearance.

EBT 14 hon3

In addtion to the work on the motive power, I have also been building rolling stock kits and built the benchwork for my EBT Robertsdale module.  I'll be posting photo updates as the work progresses.  Enjoy.


Detailing a Gem/Seiko EBT large mikado in HOn3

Written by Philip Light

I thought I would start a series of blog articles on improving the vintage Gem/Seiko EBT large mikado model.  Our choice of motive power for modeling the EBT in Hon3 is quite limited.  The best locomotive models available for modeling the EBT in HOn3 are the covented Hallmark models.  While these models are accurate, beautifully detailed and excellent runners, they are a bit on the pricey side.  For the large mikados, there is one alternative which is the Gem/Seiko model.  These can be found for quite a bit less than the Hallmark versions.  The model isn't specifically labeled as any one of the 3 large mikados.  The label on the box simply says "EBT Hon3 Articulated 2-8-2 Mikado".  Most people consider this model to be a model of #16.  I decided to detail the model to be #18 for several reasons listed below.

EBT 18 box

Below are some of the reasons for using the Gem model as 18 rather than 16 or 17.

  • The model has a riveted tender just like 18 does.
  • The headlight is correct for 18 (16 also)
  • 18 is slightly larger than 16 & 17 and the Gem model is slightly larger than the Hallmark 16 & 17 models
  • The complete lack of piping on the air reservoir tanks let's you model it however you want

With the decision to use the Gem model as 18, the next step was to identify everything that needed corrected or modified on the model.  While studying the model compared to photos of 18 (as well as 16 & 17) I identified an alarming number of errors.  Listed below.

  • Missing grab iron on smokebox front
  • Only one cleanout on the side of the smokebox
  • Missing grab iron on the side of the smokebox
  • Incorrect and poorly detailed bell
  • Missing air cooling lines on both sides over the air reservoirs
  • Missing all the air lines from compressors
  • Missing sanding control lines
  • Pop valves mounted on top of steam dome instead of on the side
  • Whistle on wrong side of steam dome
  • Generator mounted backwards
  • Missing cab windows on the front of the cab
  • Second driver is flanged rather than blind
  • Frame is completely wrong under the cab in the rear
  • Cab back lacks all detail
  • No backhead (giant open frame motor)
  • Wrong trucks on tender

The list could go on and on.  It's no wonder the model brings half the money of the Hallmark versions!  The first order of business was to remove the old open fram motor hanging out of the back of the cab.

EBT 18 rear

While I had the motor out I had to verfiy the rolling ability of the chassis without the gearbox.  The model ran fairly well right out of the box but there seemed to be some minor binding in the mechanism.  While I've got the model apart, it's best to correct any potential causes for unnecessary motor stress.  The old gearbox wouldn't cut it anyway.  It was bulky and didn't provide much in terms of gear reduction.  I removed the running gear and the drivers from the model along with the gearbox.  With the gearbox off, I could test the rolling of the chassis without the impedence of the gearbox.  I quickly identified that the 2nd driver had a bit of a bind. The drivers were removed from the axle and the bearing block was very slightly reamed.  The axle bearing was also not slotted enough to allow the compensation to work correctly so it was enlarged slightly.  With these modifications done, the chassis rolls very very smoothly with no bind at all.  You can push it along a track with just your pinky finger and very very light pressure.  If you tilt the track it rolls very smoothly.  Finally, a new NWSL 28:1 gearbox and NWSL 16x30 can motor were installed and tested.  The new motor and gearbox provide absolutely perfect operating characteristics.

With the drive/chassis issues corrected, it is on to fixing some of the glaring errors on the model.  I began by replacing the bell with the correct air ringer type.  Next the steam dome was removed from the model and rotated 180 degrees so the whistle is now on the correct side.  With the steam dome off, the pop valves were unsoldered from the top of the dome and moved to the side.  The holes in the top were filled with styrene rod and filed smooth with the dome.  Next came the air tanks.  The air tank on the fireman side was far too long.  On the prototype, the tank has air cooling lines that wrap around the end.  On the model as it came, there was no room between the end of the tank and the ladder to the running boards on the front!  The tank was removed and a new tank turned on the lathe.

IMG 4093

As you can see in the above photo, the original tank was too long by several feet!  The new lathed tank is in the front and is based on photographs and drawings of the large mikados.  Also note the bell and pop valves on the steam dome.  With the new tank off the model, the air cooling lines could be fabricated.  In the next photo you see the new lines which were bent up from brass wire.  Hangers were fabricated out of thin brass.

IMG 4106

Finally, the new air reservoir and air cooling line assembly was soldered to the body.  Note, the model was micropolished prior to the start of soldering work on the body.  Most, if not all, brass models were sealed with either a clear coat or actually painted with a brass paint.  In order to solder parts on or remove them from the model, it is necessary to remove the old finish.  I don't like to dip the models in stripper as you can damage any plastic parts that may be on the model.  Instead of dipping, I disassemble the entire model piece by piece and micro polish/media blast the model to remove all dirt, grime, excess solder, clear coats etc.  The micropolishing process also leaves an absolutely perfect surface finish for painting.  The model will get two media blasting treatments prior to paint.  One to remove the nasty stuff listed above that prevents good paint adhesion and one after new parts are soldered on to ensure all parts have a consistent finish and accept paint well.

IMG 4140

You can also see in the above photo that piping for the air compressors was added and the turret valve block on top of the boiler in front of the cab was added.

Work is continuing to progress on the project.  The engineer's side air reservoir is almost finished.  New photo etched cab windows, a new cab back and frame detail has been ordered.  As a result of the photo etching designs being needed for this project, I've added a detail kit for this model to the store.  You can buy the cab back, cab windows and frame extensions as an upgrade kit!

Stay tuned for the next update.


Login Form

Shopping cart

Cart empty

Featured products

Paypal Verified Seller

Official PayPal Seal