Model trackwork, grades…..!

16th July 2018:  This post has been updated at intervals over the last week.


Note that larger versions of all photo’s can be viewed by clicking on them.

Terry Smith’s modified Custom Brass SR&RL #18 hauling freight cars uphill while on test. Note the packing on top of the left hand trestle to raise one end of the temporary board.

At the start of August 2005 on the MaineOn2 Yahoo Group, a thread started about grades. The discussions between Walter Orloff, Terry Smith, Peter Tuttle and Trevor Marshall (and others) are summarised here for the insights they provide.

This is one of those classic questions for which the answer is “it depends”.

It depends a lot on what locomotive we’re talking about, how well your particular model of it was built (brass is hand-built, so each one has the potential for its own unique blend of operational problems), how well your particular model has been tuned, the quality of your trackwork, etc.

Other factors also have to be considered, the rail material you are using, and material your loco drivers are made of,  this determines the coefficient of friction between the two materials, hence slippage  The rail cleaner you use can increase slippage. So the short answer is, the best way to determine this is to do tests with your own models. If you plan to pull any brass cars (cabooses or passenger cars come to mind) be warned that they could severely limit your grades.

For a very technical description of grades for the model layout visit

FAQ authors note, no track grade does not necessarily lead to a flat layout.

Trevor’s approach – I’d like to be able to pull passenger trains with 3-4 cars for the occasional railfan trip, camp special, or running of the S&PCRR’s premier passenger train, “The Rusticator”, so I’m building my layout with a planned grade of ZERO. I’m sure that’s not the answer you’re looking for, but it at least explains why I’ve decided to not introduce grades on my layout. My advice – build some track – straight and at your minimum radius – on a portable board, prop it up to various grades (which you can calculate by using rise over run), and then pull some equipment with your locomotives and figure out what works for you.

This picture shows a temporary set up assembled by Terry Smith in his garage in order to demonstrate the PFM Sound System to a friend. As will be seen, it is very easy to convert to grade test rig. The board is a 3″ thick slab of foam insulation board supported on two folding trestle supports.

Peter Tuttle – Allan Hanson, one of the pioneers in HOn2 and then On2, recommended a maximum grade of 0%, for all the reasons already mentioned. And Al’s scratchbuilt On2 engines were engineered to track and pull. (His SRR #15 had working leaf springs.)

Terry Smith‘s experience – I tested each of my locomotives for haulage capacity on the flat, round curves and up and down grades. The first tests were done with freight cars. Most loco’s could pull 6 or 7 boxcars built from SRCS kits with Grandt Line trucks and NWSL wheels. The bigger engines like the Sandy River Prairies and the Bridgton 2-4-4’s #7 & #8 could handle 10 boxcars (all that I had running at that time). The track was Precision Scale flex track with code 70 nickel silver rail. The tests originally showed me that the units tested could still pull a reasonable train up a 2% grade without too much slipping or other problems and my first iteration of my P&SR layout was designed with these figures in mind.

…………………………………….And then it all changed!.

I started using a PFM sound system, and had bought a set (#15/#16/#31) of The Car Works Bridgton coaches/baggage cars. Because of the sound system, I found that I needed to clean the track to avoid spurious sounds/signals etc. I ended up using Goo Gone, which is a fabulous cleaner, but the trains could no longer manage the 2% grade, so the P&SR was re-jigged to become as perfectly flat as realistically possible. 

It is only through discussions on another board that I have recently learned that rails should not be cleaned with Goo Gone as it leaves a deposit which is slippery. A pal of mine uses Rail Zap, and I have a sample to try, maybe more later.

I would also add that details on another site (see below) which deals with a number of technical issues for model railroads had an interesting section on haulage capacity and grades. It reminded me that different wheel materials will have different performances, and if I remember it correctly it more or less stated that nickel plated wheels on nickel silver rail was nearly the worse combination (lowest coefficient of friction, therefore lowest haulage capacity) and that turned steel drivers were better and powder metal iron drivers were the best. Interesting that the CB loco’s tested had different degrees of nickel plating, the Portland Products loco had turned steel driver tyres and I suspect that the Car Works Loco’s have nickel plated driver tyres or do they have stainless steel tyres? Technical site mentioned above,

On2 Maximum grade tests done for P&SR ;-

F&M #1 hauled two coaches up 5/8” rise in 48” (1.3%)

B&SR #7 hauled two coaches up 1 1/8” rise in 48” (2.3%)

F&M #1 hauled three freight cars up 1 1/8” rise in 48” (2.3%)

B&SR #7 hauled three freight cars up 1 1/2” rise in 48” (3.1%)

The F&M #1 weighs 370 grams, so in total it was lifting around 715 grams up a 2.3% slope (based on average freight cars) and 1074 grams of passenger cars and loco up a 1.3% slope.

The B&SR #7 weighs 548 grams, so in total it was lifting around 893 grams up a 3.1% slope (based on average freight cars) and 1252 grams of passenger cars and loco up a 2.7% slope.

F&M #1 model by Portland Products, no modifications at test, except well run in.

B&SR #7 model by Car Works, no modifications at test, very little running.

Two coaches would be two from The Car Works B&SR #15, #16, #31. Three freight cars would be SRCS kits equipped with Grandt Line Trucks and NWSL wheelsets. Freight cars in P&SR layout running condition weigh between 75 and 155 grams. The calculated total weight of the three most likely freight cars based on average weight is 345 grams (12 ounces). Track used was Precision Scale flextrack with code 70 nickel silver rail.


Additional tests done by Terry Smith in July 2018

SR&RL #6 hauled a coach and combine up a 3.6% grade.

SR&RL #18 hauled two coaches up a 3.6% grade.

SR&RL #6 hauled two coaches and a combine up a 2.04% grade.

SR&RL #18 hauled two coaches and a combine up a 2.9% grade.

SR&RL #6 hauled six freight cars up a 3.6% grade.

SR&RL #18 hauled seven freight cars up a 3.6% grade.

Amount of slip/wheel-spin allowed on test: I allow a minor amount of locomotive driver wheel slip when starting the train on the inclined test board, but do not allow wheel-spin once the train is moving at a steady pace. This is judged by eye, so it may be somewhat variable in terms of accuracy and reproducibility.

I recorded the weight of the train of freight cars that the loco could pull up the grade slope, which was 692 grams (6 cars) and 798 grams (7 cars). The passenger coaches weighed 352 grams and the combine weighed 364 grams.

The modified SR&RL #6 weighs 474 grams, so in total it was lifting 1166 grams up a 3.6% slope (freight cars) and 1190 grams of passenger cars and loco.

The modified SR&RL #18 engine only weighs 500 grams and its tender weighs 184 grams, so in total it was lifting 1482 grams up a 3.6% slope (freight cars) and 1388 grams of passenger cars and loco.

The Custom Brass SR&RL #6 tested above has been re-motored with a Faulhaber 1624 coreless motor driving through the original gearbox, and has the OEM boiler weight modified to add more weight over the drivers, as well as extra lead in the boiler. Additional power pickups have also been added.

The Custom Brass SR&RL #18 also has the same 1624 Faulhaber coreless motor, but in this case it is married to an RG4 gearbox fitted to the rear driver axle. The PFM Sound System fitted has an on/off switch allowing the loco to run as a non-sound loco on conventional controllers, and it was slope tested in this configuration.


The foam board set up is easily converted into a grade test rig by adding packing on top of one of the trestles, as shown here.

Rather than relying on measuring the height of the packing, Terry prefers to use a spirit level with a shim piece (electrical connector block used here) to determine the true grade. The level is levelled by moving the shim piece, and the height of the end of the level above the board is measured by a ruler. Dividing this height by the length of the level gives the grade.

In this picture Terry’s modified Custom Brass SR&RL #18 hauls a three car passenger train uphill.


The topics of locomotive drawbar pull and truck friction are related to the maximum grade capability, and have now been published on these FAQ’s is a separate topic. Click here to view on another page.

Text corrections: 03- July-2018. Text additions and photo’s added: 09- July-2018. Further additions and presentation changes 11, 12, 13, 16 & 24 – July – 2018.



4 thoughts on “Model trackwork, grades…..!

  1. Terry, thanks for sharing your tests. I’ve been executed similar tests for Sn2 and may do a some more before building my layout. One of my multi-deck concepts requires large amounts of 2% grade and tops out around 3% are the furthest reaches to/from Bigelow and Rangeley.

    You are right on to consider truck resistance. For Sn2, the HOn3 Blackstone trucks are a blessing, I’ve used them to convert my T&T brass Rangeley from a sled to free rolling car. The minor imperfection in look out weigh the lack of roll in the brass trucks.

    Similarly Sn2 has Brass trucks that do not roll well and some HOn3 trucks that do not roll as well as I would have hopped. The HOn3 Blackstone trucks roll tremendously but do not look right under the cars. Most recently, I’ve been testing Shapeways printed trucks, converted HO tichy trucks (with shapeways printed bolsters) to determine if they are acceptable.

    I wrote about my earlier testing here:
    My Sn2 WW&F #7 can pull about 17 cars on the level with a mix of trucks but with pull trucks it can only pull 3 cars up a 2% grade. With lighter and better trucks it can pull a more acceptable 8 cars and my Sn2 SR&RL #7 (equivalent to your #6) pulled about 6 cars.

    Your #18 seems disappointing in my eyes. I would hope that a 2-6-2 could pull 1/2 again more cars than a 2-4-4. So I’d be looking for 10-12 cars on a 2% grade.

    In my dreams, I want my model railroad to haul trains that seem worthy of operating. As on operator, it seems goofy to be pull trains of 2 freight cars and a caboose.

    One concern I have with setting grades too high is the wear on the locomotives. Driver slip will wear off the nickel plating. Stiff down grades might put wear on gear boxes and mechanisms. I’d have some 200 feet of mainline at 2% grade.

    Sorry for the long comment, but I thought your readers might like to understand one modelers thought process on grades. At this time, I’m trying to decide if I should build a full SR&RL with significant grades or limit my railroad and only have a single deck.

    Dave K.

    PS. John Roger’s On2 SR&RL layout was multi-deck and had grades up to 4% (I think). I operated it once, it seemed OK. I’d think your readers might find his efforts of interest. I think he posted about that on the On2 list (or one of the maine 2-foot modeling lists). His layout is no more.

    • Dave,

      First off thanks for replying.

      Regarding On2 trucks used on the grade tests, then I would advise that wherever possible I will use freight cars equipped with Grandt Line trucks and fitted with NWSL wheels. This represents, as far as I know, the de-facto standard for On2 and the use of any other trucks in the mix would add an extra uncontrolled variable to the tests.

      On2 Brass trucks typically have higher friction than the Grandt Line freight trucks by a factor of around +50%, but the Custom Brass tender trucks fitted to the #18 tender are particularly poor at around three times the Grandt Line levels. This has a very severe impact on the grade test performance.

      The updated post/topic now includes “Amount of slip/wheel-spin allowed on test: I allow a minor amount of locomotive driver wheel slip when starting the train on the inclined test board, but do not allow wheel-spin once the train is moving at a steady pace. This is judged by eye, so it may be somewhat variable in terms of accuracy and reproducibility.”

      I believe that you are right to be concerned with setting grades too high on your proposed dream layout for a number of reasons. As a former professional engineer involved with the field of power transmission (gearboxes), I would be more concerned with the amount of wear that occurs when the transmission is heavily loaded pulling uphill, rather than the much lighter loads that will occur going downhill. However, I do recognise that some model railroaders may see the possible bucking of a lightly loaded loco going downhill as a problem.

      Two of the FAQ editors remember a number of threads and conversations with John Rogers on the Maine On2 Yahoo group board about the supposedly poor quality and limited lifetimes of gearboxes in On2 loco’s. John was sufficiently mechanically skilled with a good enough workshop to cope with the maintenance demands of running heavy trains up and down 4% grades in On2. I suspect that most Maine On2 enthusiasts would happily settle for lower grades with lower maintenance demands.

      Details of John’s layout and his various conversations can be found on the Maine On2 Yahoo group message board with further entries in the both the photo and file sections.

      Note that John’s layout was one of the last topics/albums posted to the original Photo FAQ’s on Fotopic before that site went belly up without notice. It was unfortunate that the backup copy that was eventually made available, was from an earlier date and did not contain John’s layout, so all the original work was lost.

      Regarding your view of the performance of my SR&RL #18, I was rather pleased with its performance, as it out pulled my SR&RL #6, which has added lead in the boiler directly over the driving wheels. My recent focus in this field has been to try to improve the performance and pulling power of the smaller 0-4-4 Forney’s in my collection, such as the Putnam & Stowe SR #1, the Portland Products F&M #1 and the Custom Brass SR&RL #6, so that they can reliably pull realistic trains.

      I think that the recent tests demonstrate that my modified SR&RL #6 now has pretty good performance and pulling power, but that is just my personal opinion.

      I will acknowledge that the previously published pulling power of my Car Works B&SR #7, now looks weak by comparison, but that is of little concern to me as the techniques that I have used to improve my SR&RL #6 can be used on any engine.

      Do let us know what you eventually decide for your own dream layout and why,


      • Thanks for the lengthy reply. I agree that your #6 improvements are very positive and if you can apply them across your fleet of locos, including #18 you’ll be happy and achieved your On2 Master Mechanic certificate.

        A DCC decoder with back emf turned on, might not slip at all.

        Any future progress toward my Sn2 dreams will be posted at

  2. Pingback: Locomotive drawbar pulls and truck friction measurements | The Maine On2 FAQ

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.