"Your source for standard gauge modeling in 1:20.3"

George Konrad, who hails from Beaumont, California, is best known as an prolific author in the model railroad press, having authored over 60 articles on such diverse topics as 19th century mining and industrial operations to large scale kitbashed locomotives and rolling stock. In real life he is retired from IBM field service and spent 20 years in the US Air Force Auxiliary as an aerospace education officer, retiring as a lieutenant colonel; but he now facilitates product development for Accucraft Trains and has worked many years as a custom outdoor layout builder. But his singular claim to fame in F scale is the construction of the first 1:20.3 standard gauge mallet (see below). He may be reached at gkstudios@msn.com

At left: George and his wife Trudie (6-8-08).

D&SL 2-6-6-0:

The first standard gauge mallet type steam locomotive to be built in 1:20.3 is the work of George Konrad who was inspired by Barry Bogs' model of the same locomotive in 1:22.5 scale, Gauge 3. The prototype locomotive at right, built in 1916 by Alco's Schenectady works for the Denver & Salt Lake RR, a line absorbed into the Denver & Rio Grande Western in 1947, was a 2-6-6-0 compound with 55" drivers and exerted about 77,000 pounds tractive effort (hence the D&RGW's designation L-77). The L-77s were short lived on the D&RGW, all of the class being dismantled by mid 1952. As an interesting side note, some of the L-77 class began life in 1908 as 0-6-6-0 mallet compounds for use in both hump yard and helper service and were later equipped with a pony truck for better tracking. On the D&SL, or "Moffat Road," they were heavy freight power; but on the Rio Grande, they were midget mallets in comparison with the road's massive simple articulated L-131 2-8-8-2 and the two classes of 4-6-6-4 Challenger sported by the Grande. Like all Rio Grande standard gauge steam, save for a lowly C-28 class 2-8-0, none escaped the scrappers' torch.

Research & Resources:

George's D&SL 2-6-6-0 is his first effort in 1:20.3 standard gauge and is intended for outdoor operation around 370 feet of dual gauge garden railway with about 7'-6" minimum radius. Since there are no readily available prototype drawings for this particular locomotive, dimensions for the F scale model have been approximated from several different sources: (1) the D&RGW's own folio drawing, which provides rudimentary wheelbase measurements, (2) a set of elevation and section drawings for the predecessor 0-6-6-0 which some of the L-77s were based upon, and (3) the O scale builder's drawings that were made for Pacific Fast Mail's imported brass O scale model (kindly provided by the current owners of the non-defunct PFM product line). Several of these drawings are reproduced below:

George's model is a combination of scratch building and adaptation. The tender, tender trucks, boiler, cab, and lead truck are all scratchbuilt from either brass, styrene, or urethane castings based upon styrene masters. The chassis however is the result of regauging two brass and steel Accucraft 1:29 scale, Pennsy K4s Pacific mechanisms with new axles, frame spreaders, frame extensions, and cylinder halves. It is on this part of the project that I have had a prominent role, George having asked me (Dave, your host) to make the Accucraft mechanisms work and to look like a small mallet rather than a couple of of truckless 4-6-2s! But more on that below.

The Tender

The D&SL 2-6-6-0 tenders were modest rectangular affairs, carrying 22 tons of coal and 9,000 gallons of water, and riding upon 5'-11' wheelbase Andrews trucks. George's tender trucks are cast in urethane from styrene masters, and are reinforced with music wire embedded within the casting. They are not sprung, but do equalize, pivoting about the truck bolster. Wheelsets were originally of my offering, now available through my friend Don Niday. The tender itself is built up from 0.040" styrene sheet, the sides of which were marked for lines of rivets, predrilled, and then hundreds of small Peco brand track nails were inserted into place. Provision has been made for a Phoenix Sound System (of course, George is one of their dealers).  

Pony Truck

Since no detailed plans were available for the lead truck on the D&SL 2-6-6-0, a compromise solution was to use scale drawings for a built-up lead truck of roughly the same design. One which I happened to have on hand came from a late 19th century Southern Railway G-Class 2-8-0. After the original Southern Railway linen blueprints were scanned, I imported their raster image into AutoCAD, traced them, and then generated separate part drawings for each part of the truck. Since this was not to be a casting pattern, but a one-off model, each part was cut at 100% exact scale size from either brass sheet or brass bar stock on my Japax Wire EDM. I also supplied the 30" wheels and axle, the former adapted from 7/8" scale steel wheels made by Gary Watkins of Sierra Valley Enterprises. These were spot on with respect to their OD and required only a bit of facing on the lathe to make them appropriate for F scale. The remaining assembly work George did in his own workshop, including the addition of the radius arm.

The Chassis

The biggest challenge with George's project has been how to come up with a plausible chassis without going to all the trouble of building everything from scratch. The solution that George arrived at was to use two 1:29 scale Accucraft Pennsylvania Railroad K4s mechanisms since their axle spacing between the drivers was very, very close to the L-77's 5'-0" in 1:20.32. This is the point at which I got involved in the project, and my job has been to regauge and modify them into the high and low pressure units of a mini-mallet.

I began by drawing both the Accucraft frames and frame spreaders (as is) in AutoCAD and then doing the same for the 2-6-6-0's frame, cylinders and crosshead guide spreaders in both elevation and section views (or at least as best as one could estimate these from the somewhat sketchy 0-6-6-0 drawings posted above). The challenge has been both to estimate the missing pieces from the D&SL engine and to figure out how the narrow (.080" thick) laser cut frame halves of the Accucraft model could be modified to look like the 4.5" thick bar frames of the prototype--which are both longer in some places, shorter in others, and generally ride higher than the Accucraft frames.

Another one of my tasks was to create new main rod blanks, crosshead guide hangars, and crossheads since nothing large enough for a standard gauge locomotive in F scale is even remotely available commercially. These, as with the frame spacers above, were CADed up and then cut from brass flat stock on my wire EDM. George originally intended to represent the fluting of the main rods but cutting slots down the center of the brass main rod blanks using his mini-mill (as well as detail the crossheads and slot their backside for attacking to the main rods), but after further reflection, decided it would be easier to make styrene main rods and crossheads of his own, rather than attempt to hold the irregularly shaped main rods in a milling vise. George's own side rods, seen several paragraphs below, are actually of a composite styrene and brass construction.

Conversion Axles

The main task which George asked me to do was to re-gauge the roughly 55" drivers which came with his two 1:29th scale, Gauge 1 Accucraft Pennsy K4 pacific chassis. This involved not only turning new 8mm 303 stainless axles, but also swapping out the crankpins on two pairs of the drivers since the D&SL 2-6-6-0's main rods connected with the #3 and #6 drivers whereas the Pennsy K4's connected to the #2 driver--and the #2 drivers on the Accucraft model are blind (flangeless). Since I have had very little success using a parting tool on my Chinese 7x10 mini-lathe, the axles were first saw cut to rough dimensions and then were faced to their final length using the mini-lathe, the axles being held in a 5/16" collet. The collet and drawbar, I should add, are aftermarket accessories sold by The Little Machine Shop which markets a substantial line of products designed to improve, adjust and add further capabilities to these dinky Chinese lathes--of course the old adage remains: You get what you pay for, particularly when it comes to inexpensive Chinese machine tools.

Two of the original Accucraft axles had been knurled on their ends using a straight knurling tool. So I decided to duplicate these axles first in F Gauge, using another Little Machine Shop aftermarket product (a "scissors" style knurling tool holder mounted in the lathe's tool post), but found the knurls actually to be unnecessary, and really something of a problem, when it came time to press the drivers onto their axles. The straight knurls, which raised the OD on the ends of the axles by almost .005", were a bear to press into each driver, causing so much interference that the 1/4" aluminum inner plates of my quartering jig were bent (hence the two additional brass spacers shown in the jig below). In the end, press fitting the 8mm axles into the Accucraft drivers and some red Loctite proved sufficient to hold everything together (so tight a fit that in a test with a spare pair of drivers, I could not twist that set of drivers about their axles with the substantial strength of my bare hands).

The Quartering Jig

To maintain accurate quartering of the six pairs of drivers, and since the 8mm axles which Accucraft uses are too small for cutting keyways into, it became necessary to construct a tool which I have been needing for quite some time: A quartering jig. Since no commercially available quartering jig comes even remotely close to being able to handle standard gauge drivers in F scale (not even Northwest Short Line's Quarterer III), I had to build my own from aluminum plate, steel rod, and brass bar stock (more on that project will be posted on the CME website at a later date). Each axle is first gripped in an R8 collet on my mill-drill (in order to maintain accurate perpendicularity) and then pressed into one driver (which already has its crankpin pressed into place and a few drops of Loctite placed inside the axle bore). The assembled driver and axle are then placed into the jig, being supported by their bearing boxes on the custom-made inner plates (see pic at far right). The crankpin of the mounted driver is inserted into the appropriate slot on the stationary outer jig plate (using right-hand lead on all non-Pennsy engines). The bearing box clamps are then inserted into the inner plates and screwed down onto the bearing boxes. At this point  I apply a few drops of red Loctite onto the remaining driver's axle bore and add a disc or two of thin scrap paper to the axle end in order to absorb any excess Loctite). The remaining unmounted driver and sliding main plate are then loosely slid down the jig's steel alignment rods; and then the whole assembly is taken over to the mill-drill for pressing. I suppose this is all a bit involved; but where accuracy and repeatability count, nothing else will substitute for the right tool(s).

Modifying The Accucraft Gearbox

George wanted to keep the drive-train that had come with the Accucraft K4s, but this presented a couple of problems. First, Accucraft uses a 1:1 ratio bevel gearbox and then couples it with brass universals to a gear head Pittman motor. Unless one can conceal an entire second drive on the pivoting front engine of an articulated locomotive, then there is no way to power both engines on a mallet using bevel gear boxes. George chose to make the lead engine of his 2-6-6-0 free-wheeling, but at one point he considered whether it might be viable to leave at least one gearbox on one of the lead engine's axles, should he choose to power it in the future (probably by mounting the motor somewhere up in the boiler), but such that the bevel gears would not be engaged. Towards that end, I sought to dismantle one of the Accucraft gearboxes (see pics above), yank out the input shaft and its bevel gear, and then reassemble the gear box. The first difficulty I encountered was that the manufacturer apparently uses Loctite, or some other equivalent glue, to permanently secure the set screws of each bevel gear to their respective shafts. These simply would not break loose and had either to be drilled out or just flat broken by pressing the bevel gear while still on its shaft with an arbor press. When a drill bit broke off in the set screw of the input shaft's bevel gear, while still in the gearbox, no pressing option was available; and so it became necessary to cut through the input shaft ahead of its bevel gear using a Dremel tool. And this necessitated cutting through a 10mm OD brass bushing as well--sigh. The second difficulty was that each F Gauge replacement axle required spot drilling a depression in order to seat the tiny 3mm set screws which secure each of these pesky bevel gears. And so Mr. Dave had to buy yet another tool--a precision Vee Block--and set up another operation on the mill-drill (far right pic above).

Assembling The Regauged Drivers

The first two drivers were regauged and assembled in my shop during the fall of 2008. The last four were completed during the first week of February 2009. The last photo (bottom right) compares the size of the regauged 55" drivers to a 73" driver used on my Southern Railway Ps-4 pacific. The two photos which precede it show a spare set of blind drivers which I regauged for my own use (along with an Accucraft K4 sound cam).

Assembling the Chassis

As of October 2008, George had made good progress on getting his locomotive together. He had scratchbuilt his own wood and brass pilot and assembled the various bits and pieces of his regauged frames (first four pics below). By January of 2009, the chassis and its composite styrene and brass valve gear were together, the unique (and as close to prototypical as possible) articulated connection between the high and low pressure engines had been made, and the whole chassis readied for mounting to the boiler and cab assembly. All that remains is mounting the re-gauged drivers into the frames, finishing the main and side rods, and mounting the Pittman motor to the rear high pressure engine.

As of late February 2009, George had mated the re-gauged Accucraft K-4 drivers to the frames and attached the original K-4 side rods to them as well as his own scratchbuilt styrene main rods and valve gear. Since one of the Accucraft K-4 drive trains was to be reused, only the rear set of drivers are powered; and at that, the gearhead motor is mounted at a sharp angle in order to hide it in the firebox area. George writes about this part of the project:

     Here are some specs. on the shaft cutting, etc: The overall length of the gear box with shaft is now 3 1/4" Of this, the gear box itself is 1 9/32" long, the end collar is 3/8" and the rest is the shaft, some 1 3/32". Also, the shaft on the end of the motor was cut down by 1/4" and a new flat was cut onto the shorter shaft. That leaves 1/2' in the center of the coupling (U-joint). I had to alter the frame spacer that sits directly under the motor to allow the motor to pivot upright enough to be mounted at this angle. With the motor mounted as it is, there is about 1/4" of clearance between the motor and the inside of the boiler top, and the motor and the inside of the rear of the firebox.
     One might think that this model with its large boiler would allow one to slap the ole motor in there most any way, but the shafts needed to be shortened as I did and there was just one spot that allows the motor to mount with that 1/4" clearance on top and back.
     So now it is downhill from here. The wiring in the loco is done, and it does run by itself without the tender (I'm having fun watching all the action running back and forth on the test track). Connecting the power leads from the tender is next, and then finishing up the rest of the boiler and cab details, and paint. Zowee!

George has built his engine to include "working" lubricators (and if he could find scale grease, George writes that he would find a way to pump it!). Most of George's work on the locomotive between February and May of 2009 has been on detail parts, the little bits and pieces that transform a spartan toy into a scale model. Among some of these parts are scratchbuilt leaf springs, brake hangars, brake shoes, ash pans, steam pipes, exhaust steam injector, and more!
 

Boiler & Cab

In 2007 George created patterns for the domes and stack on his D&SL mallet. These he copied in urethane. The boiler, as per the same techniques that Barry Bogs uses, is a styrene tube, likewise the cab, with Peco track nails to simulate rivets.

During April 2009 George was working on the numerous boiler and cab details. He writes:

I am working on the confounded heater pipe for the passenger cars, which wends its way all over the right side of the loco from front to back, BUT before one does the pipe, one needs to do the ash pans, BUT before one does the ash pans one should place rivets on the bottom of the firebox, BUT before one places the rivets, one should build the attachment points for the ash pans, BUT before doing the attachments, one should place the "cheek" weights that are in the lower sides of the firebox, BUT..............I think you are getting the picture. All these things should be done today, I also then wish to build the rest of the cab windows (sliders, etc) armrests, and the front doors.

Of particular interest are the Magnus and Nathan non-lifting injectors which George made from scratch. The early, large generator was made by adding a scratchbuilt turbine to a modified Accucraft casting.

By May of 2009, George was nearing completion of all the locomotive's electronics. He writes:

The model now has ALL the wiring necessary installed. That includes the sound system, cuff gating, back-up light, markers (scratch built using a jewelry bead) and front headlight. The circuit board for the lights is in the front part of the boiler, so wiring had to be run all the way back to the motor, and then forward through both engines to the light! Fussy work.
     The sound board is in the traditional tender location, BUT the speaker is in the boiler. I figured that sound had better be "up front," for with a loco this long, about three to four feet would go past an observer before the tender speaker would come to pass. I have not vented the bottom of the boiler as yet, for the sound just "BOOMS" from the sounding box that the boiler makes. I have gated the chuff from the drivers, and that reports back by wire to the tender. All lights work, and the rear light has a diode system that turns it off going forward.
   There is not a whole left to do but some final details. There is the whole cab interior to do, but I am going to get all exterior details complete and the loco painted before that.
 

The Completed Locomotive

Over the weekend of June 6-7, 2009, George exhibited the completed D&SL mallet #212, sans paint, at the Big Train Show in Southern California:

As of October 2009, George has the D&SL mallet painted and lettered. George has also built some 1/20th scale freight cars to go with his mallet:

Last update: 29 October 2009