Painted Kadee couplers

23-04-2017: This is a temporary post to illustrate another current thread on the Maine On2 Yahoo! group.

 

The above picture shows 4 standard HO Kadee couplers, as used as the “de-facto” standard for On2 by many modellers. Some other modellers do not like the appearance of the knuckle spring. The top and bottom couplers are unpainted. The middle two have been airbrushed with paints depicting a fresh orange rust colour (2nd from top) and an alternate darker brown rust (3rd from top).

Take a look to see if you think the knuckle spring is that obvious on the painted versions. Clicking on the picture will load a larger version into your browser window.

The Precision Scale ¾ MCB knuckle coupler.

Model couplers –  the Precision Scale ¾ MCB knuckle coupler.

A temporary post to illustrate a current thread on the Maine On2 Yahoo! group.

Precision Scale Models have a scale ¾ MCB knuckle coupler in their extensive product range of brass parts. Part number PSH-4032-2 is described as On3 coupler, regular shank, slotted knuckle.

Dimensions: the piece of wire is 1.197″ long: the knuckle on the left measures 0.186″ high: the left coupler body measures 0.254″ across its maximum width and the left coupler shank measures 0.095 – 0.097 wide.

The Precision Scale part is accurately scaled, with a working knuckle released by a working pin that can be operated by a working cut lever. It would make a nice front coupler for any On2 loco that only very occasionally needs a working front coupler, but it does require careful assembly.

Jerry Kitts of Foothill Model Works wrote of his experience with these couplers on his Westside logging models: “I have been using both the PSC and Back Shop 3/4 size MCB working couplers but they tend to uncouple at all the wrong times. I keep going back to using Kadee #26.”

 

Track standards- parallel track spacing for Maine On2

Parallel track spacing can be a problem for Maine On2 (and other scales) layout builders at the design stage, particularly if they need to shoehorn a design into restricted space, as there is not a lot of information around for guidance, but here’s what we have found.

The NMRA Standard for model railroad parallel track spacing is S-8 which can be viewed here. The general view of the Maine Two Foot modellers is that this standard is not particularly helpful as it depends upon interpretation and interpolation, but still does not give clear results for On2. This was so eloquently put by Yahoo! group member Hilary Smith very many moons ago:-

In my reading so far, I have either not come across spacing discussions or have overlooked them. Looking at NMRA S-8 Track Center Standards for Class II (small four-wheel truck diesels; small steam; old-time, logging, branch line rolling stocking stock) O scale standard gauge puts tangent track at 3.75″ (or 15 scale feet) apart, and for a curve of 41.5″ radius, 3.8125″ apart. For On3, tangents are 2.75″ and for 41.5″ radius, 3.375″, proportionally much wider spacing in curves relative to tangents than in standard gauge. Comparing On3 tangent spacing with standard gauge shows that the track centerlines are a full 1″ or four feet closer together, but on curves only 0.4375″ (or 1.75 feet) closer together. So, are On2 cars no skinnier or longer than their On3 counterparts meaning that On3 spacing standards are a good choice? Or should the centerlines be closer still? Thanks for any observations, experience, or advice.

Hilary Smith,  Northern Virginia

 

The post got two replies;-

Terry Smith: The Maine two foot passenger cars can be longer than was common on three foot gauges, meaning that models of Maine cars “cut the corners” more than the equivalent three foot lines. The controlling dimension is the truck centres. I think that three foot narrow gauge cars are slightly wider in general than two foot, but not sufficiently different to make a difference modelling wise for track centres. ie using the On3 data will be OK for On2.

Trevor Marshall:  
You asked about spacing between tracks on tangents and curves for On2 layouts. I think it’s safe to say, as Terry Smith suggested in his reply, that passenger cars are the critical equipment for curves. Therefore, if you have such a car (SR&RL – not, say, Monson, whose lone passenger car was much shorter), you could do your own testing:

1 – lay a minimum radius curve on a surface on which you can draw with a pencil. You can either spike down a piece of On2 flex, or just a pair of rails, directly onto, say, a 1×6 piece of pine.

2 – put the car on the track.

3 – holding a pencil at each end of the car on the outside of the curve, roll the car along the track so the pencil draws a line on your surface

4 – repeat holding a pencil at the midpoint of the car on the inside of the curve.

5 – measure the offset between the track centerline and each line you’ve drawn.

Note that the pencil line will be slightly offset from the edge of the car – that’s built-in clearance for you. If you don’t yet have a passenger car – or don’t want to use that nicely painted model as a marking device – you could always cut a passenger-car-sized piece from styrene or wood (doesn’t have to be exact, but should be at least as large as the car is, corner to corner), add some bolsters from styrene strip under it at the proper locations, and screw some On2 trucks to it. You could even add pencil holders at each corner and in the middle if you like. Build two of them and you can check clearance on the layout before running your real models through. Heck, even add vertical pieces at the ends and in the center to turn one into a clearance car for bridges, buildings, holes in the backdrop, etc.

That’s curves for you. For straights, I have two standards: For transfer tracks, freight platforms, or other places where cars will be spotted adjacent to each other or adjacent to a structure with the intent that plates be dropped between them for the transfer of goods via hand-trucks or strong backs, I use my widest pieces of equipment to lay the tracks as close together as possible without sideswiping. For other places, I’ve spaced tangent track on 3″ centers. This is way more than the minimum spacing required, but it looks right to me – narrow gauge yards tend to have a spare, open feeling to them, and 3″ spacing allows for an O scale figure to stand between cars on adjacent tracks without getting sideswiped. Remember, the minimum is not always the best.

Hope this helps. – Trevor in Toronto

Some time later, the question was asked again, and this time the replies included references to the Maine Two Foot prototypes and other modellers’ experience.  

sizecomp01a

Bob Schlechter: Scaling from the prototype photo above, taken at Bridgton Junction, gives approximately 8′ 7″ to 8′ 9″ as the center to center distance of the two foot trackage to the standard gauge trackage. The photo was taken from the roof of a standard gauge box car looking downward at the trains and trackage giving a good vantage point.

bob02pt01

Bob Schlechter: on my Kennebec Central diorama (above) I used 8 1/2 scale feet center to center. Looks good for prototype spacing but if one is going to do lots of operation and switching you may want it wider to allow for full size fingers to be able to reach in between cars!

 

tcspt01a

Terry Smith: “I used 2½” spacing for straight parallel tracks widening to 2¾” around the curved sections (36” radius) on the Reading iteration of my P&SR (pictured above). I added the extra bit of clearance around the curves because I had a passing loop cum staging tracks that were bent round a 90 degree corner in parallel.
I do not recall having problems with access for manual uncoupling (using a Rix wand for my Kadee’s) or with interfering overhangs/underhangs with the longer rolling stock (coaching stock and Bridgton 34 foot freight stock) or locomotives like the large Forney’s, (B&SR/B&HR #7 & #8), or SR&RL #24 using these values.

 

Bob Harper: I use 3″ spacing on my On3 tracks and 2½” on my On2,  which reflects the difference in width of the real rolling stock – 8′ on the 3′ and 6′ on the 2′ lines. Looking at Franklin on the FAQ’s will show how this looks.

Editor: see the picture above of Bob’s Franklin module where the slight out of focus plus lighting highlight the tracks. To view more pictures of Bob’s Franklin module on another browser page, click here.

Bob Harper continues: Obviously the throw-over of the long On2 coaches is important, but I have found that the critical dimension is the cab on SR&RL #23. This is far wider than anything else (apart from a snow plow?), and my mistake of having the driver leaning out of the side window made clearances even worse! He has since been repositioned.

But another factor to bear in mind is the sense of proportion of the whole scene. Unless we are very fortunate with the space available, we inevitably have to fore-shorten out scenes, and trying to keep a prototypical width while losing the prototypical length can lead to the eye perceiving the track spacing to be too wide. In order to keep the overall scene in proportion, it may be necessary to slightly reduce the track spacing from prototypical standards.

 

Trevor Marshall: I like the generously spaced look of the Maine Two Foot yards, like in the picture below of Bridgton on the B&SR/B&HR;-

tpmpt02a

Trevor continues: I’ve used 3″ centers on my On2 layouts. I find it gives that appropriately “NG” look.

tpmpt10a

Above is a picture showing Enoch Pond station area under construction in the first iteration of Trevor’s S&PCRR, which captures the spaced out look between the tracks.

tpmpt05b

This is a rare overhead shot of the Hebron Pond Slate Company sheds and passenger depot on the second iteration of Trevor’s S&PCRR.

tpmtp12a

Trevor continues: – I built a test track board that included both 2′ and standard gauge tracks, side by side, like a transfer yard (illustrated above by the interchange yard at Monson Junction). I lined things up visually – used a standard gauge and 2′ boxcar and double checked my spacing with an engine. I think I left a scale 6″ between the equipment, and raised the 2′ gauge line so that the floors of the cars were level.

What I did NOT do was check the spacing with a 2′ gauge plow. ……..Guess what? They’re a lot wider than other equipment. So, if you can get a plow or mock up the width of one (a block of balsa would do), it’s a good idea for checking clearances.

I think I’d still build the transfer tracks close like I did. Crews will just have to shovel the tracks by hand, or plow when there are no standard gauge cars in the yard.

Editor (18-12-2016): I’ve just found my plans of the Bridgton snow plow #2, which measures 8ft 10ins wide at its maximum – a lot different from the normal 6ft 6in to 6ft 9ins width of Bridgton freight cars and coaches. The model snow plow that Trevor acquired was built from a SRCS kit for SR&RL #513, which measures 8ft 4ins wide at its maximum.

Editor: Here’s a picture of the interchange yard at Snowdon on the second iteration of Trevor’s S&PCRR.

tpmpt03a

For more pictures of Trevor’s layouts click here. Well worth a browse!

If you have different suggestions and/or experience and would care to share it with us then please contact us.

 

 

Track gauges – On2

In this post we show some of the commercially produced gauges available at various times.

Simpson roller track gauge.

Russ Simpson produced these gauges to suit variety of rail sizes (eg codes 55, 70, 83 & 100). They were first announced in July 1977 and were available direct and from suppliers such as Coronado Scale Models, Caboose Hobbies and BK Enterprises.

tgtcs02

tghmm01

Portland Products NMRA style gauge.

Bob Werner produced this style gauge back in the mid 1990’s, and examples may turn up on eBay. One was offered as recently as mid-June 2016.

tgtcs01

tghmm02

 

We know that SRE produced a cast brass three point gauge way back in 1970’s but we don’t have a picture or two – can you help?

On2 Yahoo group member John Rogers also produced a three point gauge in the late 2000’s to early 2010’s, but again we don’t have a picture or two – can you help?

Revised 02-07-2016.

On2 Mounted Wheel Standards

tsws01

The picture above shows a selection of Maine Two Foot prototype wheelsets at the Sandy River Museum in the mid 1990’s.

In the following table, we show three different wheel standards compared to prototype dimensions derived from measurements of photographs and also a sample of measurements taken from commercial wheelsets. The ¼AAR for Two Foot Gauge – 1966 and the NMRA – 1974 Standards are included for historical purposes, as they reflect the standards in place when many of the brass locomotives and rolling stock were first produced.

Experience has shown that the back to back measurement of the wheels is most critical single dimension that affects the running of On2 trains through pointwork, and paradoxically this is also the dimension which is most likely to vary outside the control limits of the various specifications.

The table can be enlarged by clicking on it.

On2 wheel standards 2016-01

Notes

1/. Scaled from known object placed on central plane of wheelset.

2/. Scaled from track gauge with allowance for perspective.

3/. Measurements based on set of 4 wheelsets bought in the mid 1990’s. Measurement accuracy estimated at +/- .002 inches.

5/. By calculation.

7/. Back to back variation in a wheelset: three wheelsets had no measurable variation, one had .006” variation. This wheelset showed wobble on the insulated wheel.

8/. Back to back variation in a wheelset: all four wheelsets showed variations ranging from .002” to .005”. Three wheelsets showed wobble on the insulated wheel, one on the un-insulated wheelset.

9/. Back to back variation in a wheelset: one wheelset had no measurable variation, three wheelsets showed variations ranging from .002” to .003” and wobbled. Larger variations in measurement between set of four wheelsets, as expected by use of injection moulded axle with double shoulders. (Range of .018” versus .010” for NWSL and Portland Products).

10/. Back to back variation in a wheelset: wheelsets showed variations ranging from .003” to .024” and wobbled. Larger variations in measurement between set of four wheelsets, as expected by use of injection moulded axle with double shoulders. (Range of .027” versus .010” for NWSL and Portland Products).

Grandt Line Wheelsets.

The original Grandt Line On2 wheelsets consisted of metal rims with cast plastic centres featuring ribs on the back of the wheel and a moulded half axle outer profile with a steel functioning axle.

A later version was produced using the metal rims and cast centre only with a moulded plastic functioning axle. This combination can give problems with true running of the wheels, and most serious operators at the time chose to upgrade to NWSL wheelsets.

Grandt followed suit and supplied their freight truck with NWSL wheelsets. More recently Grandt has chosen to supply their freight truck without wheels.

Sample locomotive back to back wheel dimensions.

The following data was provided by Terry Smith from his own locomotives;-

Portland Products, F&M #1, from the front;-   .422 – .426 – .430 – .432

Putnam & Stowe, SR #2,     .417″

The Car Works, WW&FR #7, from the front;-   .417 – .4145 – .4205 – .4105 – .4105

The Car Works, B&SR #6, from the front;-   .410 – .412 – .4165 – .4115 – .4115

The Car Works, B&SR #7, from the front;-   .4025 – .4175 – .4135 – .408 – .4135

The Car Works, B&SR #8, from the front;-   .415 – .4165 – .422 – .413 – .413

Custom Brass, SR&RL #6 (1) , from the front;-   .4195 – .425 – .418 – .4165

Custom Brass, SR&RL #6 (2) , from the front;-   .420 – .421 – .421 – .419

Custom Brass, SR&RL #6 (3) , from the front;-   .424 – .421 – .415 – .4155

Custom Brass, SR&RL #6 (4) , from the front;-   .421 – .426 – .416 – .4155

Custom Brass, SR&RL #24,   .413″

 

Is “On2” P48, Proto 48 or Finescale? No, not according to current NMRA definitions or Standards. Click here for more details.

Updated 22 – May – 2016, 28 – May – 2016 and 20 – August – 2016.

Precision Scale Flex track

tstr10Precision Scale introduced its On2 flex track in one yard lengths in mid-1983.

The track consists of an injection moulded tie (sleeper) base strip of 21 ties of random irregular lengths with moulded spike heads holding nickel silver code 70 rails. The track gauge typically ranges between .508″ to .512″ on measured samples.

tstr14b

Typical rail dimensions.

The tie base material is believed to be a Delrin type polymer. The dimensions of a typical tie are: 1.279″ long x .153″ wide x .103″ thick on .406″ centres (scale : 5′ 4″ long x 7.3″ wide x 4.9″ thick on 19½” centres).

tstr11

This deliberately foreshortened view of the track clearly shows the irregular tie ends. The yellow arrows show the length of one of the tie strips. The dimension from the inside edge of the bottom of the rail to the outside of the tie varies from .377″ to .419″ on one side and from .374″ to .422″ on the other.

tstr12

tstr13

This underneath view shows the single bars connecting adjacent ties. Experience has shown that care needs to be taken when laying this track to ensure that the ties are as perpendicular to the rail as possible, and not angled, to avoid reducing the gauge of the rails. This is especially important on curves, and when running the larger Forney type locomotives.

 

 

Couplers: On2 models.

Model On2 couplers added from MaineOn2FAQ-Gen:        updated 04-Jan-2014

Way back in November 2000, Bill Kerr wrote, in response to an enquiry on the Yahoo! group: traditionally, Kadee #5 couplers are the standard choice for Maine On2, but I suspect there will be a move to the Sergent “S” scale coupler.  See Trevor Marshal’s evaluation below for using Sergent couplers.

Coupler height should be set for coupler center to be 16-17″ above rail height.  There is a  coupler height gauge made by AMB’s John Hitzeman.

In response to a Yahoo! group enquiry asking for clarification about coupler heights in 2009, Terry Smith wrote:

On2;- I build my own On2 cars to the prototype derived dimension of  18″ from the underside of the underframe to the rail head………and then I bung an HO Kadee coupler on each end……it works for me and  seems to match the brass loco’s that I have. I don’t refer to the  NMRA Standard for this, no need to.

Maine style On30;- based on my past experience as a small scale  manufacturer,  then all those On30 customers who had real votes (ie dollar bills) chose to build their On30 cars at a scale 21″ underside of the underframe to the rail head dimension and floor mount their couplings which then
matched with the rest of their On30 stock.

Definitions; I find it far simpler to define an underfloor height as a coupler mounting surface and then mount the coupler than work to a rather nebulous and complicated centre of knuckle height.

In 2013 Terry noted that the NMRA correct coupler height will result from using the proper car floor to rail dimension of 18″ (prototype) or .375″ (model) and normal Kadee HO couplers (#5 and similar) in their normal draft gear/boxes mounted with their topsides against the floor.

Mark Hall wrote a good article of representing prototypical coupler boxes when using Kadee couplers in M2FM 1989 Vol 21.

In May 2013 Terry Smith posted to the MaineOn2 board; – In the many discussions over the years about what couplers were used on the Maine Two Footers and their sizes and what coupler to use with On2 models, the one thing that mildly concerned me was that I had never thought that the Kadee HO couplers I use actually look out of place on my own stock in comparison with the full size Maine Two Foot cars that I had seen at Edaville, Portland and other places, or with various published historical photo’s in standard texts.

I have now finally managed to print out an end-on photograph of one of my own boxcar models with approximately the same perspective as some pictures of the full size car taken at Edaville and the earlier historical photographs from the B&SR. The ratios of boxcar width to coupler knuckle heights in the photographs are;-

Edaville 11.2

Model 11.2

Bridgton 10.7

It seems to me that the standard HO Kadee coupler is a pretty good match sizewise for those fitted to the B&SR Boxcars,

For what it’s worth,

Terry

Shown below are Terry’s photos of (L) one of his models with a Kadee coupler and (R) prototype car at Edaville.

coupler11  coupler12

The Kadee #5 knuckle measures 0.146″ top to bottom, this is equivalent to 7″ full size in US O scale. The knuckle may be oversize for HO scale, but seems to be the right size for On2. Update 02-01-2014: note that a recent search through the past messages of the Yahoo! group has shown that the height of the knuckle of the 3/4 MCB coupler most likely to have been used on the prototype is 6 3/4″ or 0.141″ in US O scale, so the Kadee #5 is pretty darn close!

Trevor Marshall’s evaluation of Sergent Couplers.

tpmcoup01

This picture shows Trevor’s model (based on a SR&RL prototype) fitted with the scale  HO Sergent coupler.

tpmcoup03

This picture (same model as above) shows how neatly the scale HO Sergent coupler fits between the wheels of the Grandt Line freight truck.

tpmcoup02

This picture shows one of Trevor’s caboose’s fitted with “S” scale Sergent couplers.

Trevor wrote (back in 2004 0r 2005);- “I’m very pleased with the appearance. The coupler head scales out to about 10” square, which looks about right. I even add a small piece of chain between the coupler and the cut lever – there’s space near the back of the top of the coupler head to drill a small hole and anchor the chain with an HO scale eyebolt. In doing tests with these, I did note that using Sergents seriously affects layout design – it will force you to keep everything within easy reach because you need to be able to reach equipment for coupling as well as uncoupling. There are benefits to this, however:

1 – If everything’s within coupling reach, it’s also within reach for building and maintaining track, cleaning track, etc. 2 – The need to open a knuckle before coupling increases the amount of work the crew does, which takes time and therefore makes the run seem longer (something from which all layouts can benefit). 3 – The Sergents are just more realistic. They’d be incredibly frustrating on a large, operations oriented layout (like an HO club layout with several hundred cars to move in a session)…. but who in On2 is doing that?

As for height, I use a little laser cut coupler height gauge made by AMB’s John Hitzeman. I’ve attached a Sergent coupler to the top of it (without the ball bearing), and use this to set the height on my equipment. Usually, this requires a spacer between the locomotive frame and the coupler box. I’ve been building my own coupler boxes, as I’ve yet to find one that these couplers fit into. The hole in the coupler shaft is smaller than that in a Kadee #5, so the plastic tube in a #5 coupler box is too big to fit through the Sergent coupler. Some good coupler boxes (laser cut acrylic, perhaps, John?) would be a great benefit for these.  Trevor in Toronto.

(Update: Feb 2005) “Having tried my own coupler boxes for several months now, I decided to try some Kadee coupler boxes. The boxes for Kadee’s S std gauge/On3 couplers (same box) fit the Sergent’s perfectly. By cutting them to discard everything behind the two mounting holes, they’ll fit between the wheels of an On2 car and not interfere with the outboard axle. The Kadee S boxes provide better coupler swing and are easier to install. I was able to bulk order 100 boxes directly from Kadee. (www.kadee.com)  Trevor in Toronto

More thoughts on Couplers from Trevor Marshall, including a comparison between Kadees and Sergents, originally posted the MaineOn2 group 25th April 2006

Hi Michael:

Those interested in a scale size coupler as operating on the Maine lines during their heyday should probably engage in further research before jumping on the Sergent S scale bandwagon (as attractive as that seems).

One thing that I haven’t seen mentioned in this coupler debate is the giant spring on the side of every Kadee coupler, that controls the knuckle. I find it amusing that people are getting twisted up about a scale inch here or there, but are apparently quite happy to ignore the big spring. And that brings me to a larger issue that I haven’t seen addressed, and that’s this: Before picking a coupler, decide what it is you want to do with the model when it’s done. Each style of coupler has pros and cons… I’ll list a few, based on my experience with both.

Assembly/Installation: – Kadee wins here, hands down. They’re easy and quick to assemble and install. This may be important to those planning to build a lot of rolling stock. Sergents take about 45 minutes per pair to build – and most people should plan on rebuilding their first half-dozen pairs because you’ll get better at doing them, and the first ones won’t operate as reliably as the rest.

Appearance: – Sergent wins, hands down. As I noted above, knuckle couplers – whether 1/2 MCB, 3/4 MCB, Type D, Type E, or what have you – don’t have large springs on their sides. Even if the Sergents are a little too large, or a little too small, not having a spring on the side more than makes up for it, appearance-wise.

Operation of the coupler itself: – Kadee gets a slight nod here, as they are pretty damn reliable right out of the box. But if you take your time and work carefully on building the Sergents, they can operate very, very well. Your success depends on your own skill.

Operation on the layout: – This depends on what your design goals are…

KADEE LAYOUTS – If you are building an operations-intensive layout – let’s say you’re doing Farmington to Phillips, with staging north of Phillips and staging northeast of Strong – you will be better off with Kadees. You’ll be shuffling a lot of cars, and the need to open knuckles on the Sergents can become tiring pretty quickly.

– If you are building a layout that has car-spotting locations that are more than 18″ from the aisle, or that are inside buildings, you’ll want to use Kadees. Obviously, this depends on a number of factors, including layout height, the length of your arm, and how good your eyes are. But in general, with Sergents you need to be able to get your head directly above the couplers you want to couple or uncouple, every time. Sergents have a smaller gathering range than Kadees, so you must open the knuckle(s) then carefully check the alignment of the couplers before trying to couple. If you can’t reach and can’t see to do this, you’ll experience no end of frustration. Kadees have a better gathering range, couple automatically, and can be uncoupled using under-track magnets so one doesn’t have to reach into the layout to make them work.

SERGENT LAYOUTS – If you are building a smaller, more relaxed layout – for instance, the KCRR – then the extra work that Sergents require (opening the knuckles, aligning the couplers) actually becomes more play value. Prototype railroads have to open knuckles, manually align couplers, etc., so it’s more realistic, too. Sergents slow down your operations. It’s impossible to bang the cars about with Sergents, whereas with Kadees one can do jackrabbit-like changes in direction over a magnet and kick cars all afternoon. Sergents, manually writing switch lists, train orders and other paperwork, and installing and learning to properly use sound systems (DCC or analog) help stretch out an operating session and turn it from a puzzle to a re-enactment of real life. For a small layout, the extra work involved with all of this will make a 15 minute operating session take an hour. Some people may like that, others will not. Decide which type of person you are before picking couplers.

Cheers!

– Trevor (using Sergent S scale couplers on his On2 equipment) in Toronto

Editors note dated 04 January 2014: Trevor Marshall is currently modelling a CN branch line is S scale, standard gauge, and using Kadee couplers for reliability. This may well change if Sergents produce a more “robust” S scale coupler for home assembly or factory assembled couplers. Please note that “robust” is used here in its manufacturing engineering (Statistical Process Control) context, rather than the more common physical meaning of hefty or over-large.

Update December 2013:  there is now a Sergent couplers Yahoo! group;-

http://groups.yahoo.com/neo/groups/SergentEng/info

and Sergent now have a more accurately moulded S scale coupler on the market.

(Trevor Marshall had reported having difficulties with reliably assembling the earlier version of this coupling, so much so that he opted for the pre-assembled scale HO version (visibly undersized) for his MaineOn2 layout). See http://themodelrailwayshow.com/cn1950s/?p=4122 for pictures, or http://www.sergentengineering.com/ for more information.