Not wanting to burn out my engines which seem to strain at the 11 car level (2025-Loco. and 8957 Burlington) I was wondering if Lionel has some parameter regarding pulling power (weight?) as a maximum for their engines. I understand there are additional variables like track and wheel condition, inclines etc. but I would assume any such parameter would assume new rolling stock and a level track.
You always have to build your train according to the various weights of the cars making up your consist. Put simply, you can’t put a lightweight high center of gravity car like an empty hopper (MPC 9111 N&W for example) ahead of a long string of heavy diecast postwar freights (especially those with coil couplers) and expect your empty hopper to remain upright on a tight curve. (My 9111 derails with a light caboose behind it)
Motive power pulling capacity is directly related to the weight of the cars produced at the same time as a locomotive. A 2035 with Magne-Traction can out pull a James Gang general because the 2035 was designed to haul heavy cars and the DC #8008 JGG was only assigned 3 very light cars.
Modern locos with flywheel equipped pitman motors can out pull darn near anything from the postwar era on. Especially dual motored diecast behemoths like GG-1’s.
Becky’s given you a very good response that there’s little I can add to. My attitude’s always been if the engine can’t handle the length of the train it’ll let you know PDQ, usually by spinning the wheels.
Just to add a bit about tight curves, they can affect the train by their increased friction on the wheels. Depending on the cars it might not happen all the time but the possibility’s there just the same.
Also take into consideration that a long train with mixed consist also runs the risk of the clothesline effect. the pulling power of the loco, and the weight of cars behind it, can “pull” a car (cars?) clear off the rails just due to it trying to make it a straight line,
Yeah, I learned that the hard way with many derailments till I recognized the pull vector was not a curve! The only answer was to go slow. I wondered how real trains take this into account or are the tracks so slowly curved that it doesn’t happen.
I still wonder, though, if there is a rating of pull weight Lionel recommends for their engines. My refrigerator car alone weighs about a pound.
It’s an interesting consideration “matching the cars to the era” the engine was made. As someone just getting back into my “childhood trains”, I recall as a kid just hooking everything and anything on the engine and opening the throttle as fast as it would go! As a senior senior citizen now my approach is “a little more measured and refined”. I will give more thought to positioning cars based on height and weight. I’m surprised I never discerned such a common sense solution. I’m appreciating more of what it’s like to be an engineer even on a model railroad!
FWIW, if you want a bit more freedom to mix and match locations of cars in a train, there’s nothing wrong too with adding weight to the light ones.
I’d suggest NMRA weight as a starting point. I’ve seen NMRA weight brought up before in postwar/toy train discussions, and a lot of people will turn up their nose because the NMRA is usually associated with scale railroading and the like. With that said, their weight standards weren’t developed blindly, and they are not bad at least as a baseline to keep trains on the track.
The standard for O scale is 5 oz + 1oz for each inch of car body length. That means that say a 10" car should weigh 15 oz. In reality, with as much as we love diecast in O gauge, you’ll often find that cars are either way under this or way over it, but it can be a good target to hit with your lighter stuff.
With that said, too, always factor rolling resistance in. Postwar stuff tends to NOT roll as well as MPC era stuff with fast angle wheels and needlepoint bearings in delrin trucks. A lot of newer pieces with diecast trucks still use needlepoint bearings and delrin inserts.
I was amazed too at some of the first modern(ish) scale Lionel I bought around 2007-2008. Not only was it basically dead on NMRA weight out of the box(give or take an ounce), but the cars rolled so freely that they found every out-of-level spot on my layout. I actually found my track was nowhere close to level-the stuff I was running was just masking it since it would mostly stay put.
While we’re at it too, there are discussions around whether or not modern trucks need to be lubricated. After a few recent incidences, I’ve changed my stance to lubricate everything 
. I had put together a nice set of scale Lionel NYC heavyweights from the late 90s-Lionel had cataloged a set of 4, then two add-on cars. It took me a bit of hunting to find all 6, but I had them, some new in box, some lightly run. I put the set on the track, and thought it seemed really difficult to pull by hand. I hooked a locomotive up to it-the 6-18056 “763E” Hudson that was cataloged along side these cars, and it moved them but was spinning its wheels a lot and couldn’t really get going that well even around O60 curves(the cars spec O48, the loco I think O54). This is a heavy loco with a “high stack” Pullmor motor, a pretty favorable drive ratio at least by typical older Lionel standards(18:1), and has both Magnetraction and traction tires.
After I decided to lubricate the cars, in going through I found one axle out of place in one of the trucks. That made a HUGE difference in how the whole set ran, and once fixed it ran 100% better. That’s still not an easy train to pull, but that locomotive could do it.
I have just bought and am awaiting delivery of a K-line scale Hudson. This is considered by some to be the “best” of the scale NYC Hudsons made in the past ~30 years, but K-line had a nasty habit of putting tiny motors in everything. This loco, like most of their steamers, uses the Mabuchi 385 motor, where most others will use a Mabuchi 500-series motor or a Pittman 9400 series motor(offhand I don’t know if the more recent Vision Line Hudson used the 9431 or the smaller Pittman Lionel sometimes uses). Some will cite 6 scale heavyweight coaches as the practical limit of what the K-line can pull-I’m eying a set of 8 K-line heavyweights, so if I end up getting those I’ll see.
Buy a fishing scale. Hook it to your locomotive and power it up until the wheels start to spin. This will give you an empirical drawbar pull number.
Next attach the scale to a train of cars and pull until the cars move. This gives you a "tractive effort " reading in ounces.
Then you just match numbers and that will give you an idea of how many cars a particular locomotive can pull.
The problem with modern NMRA standards is they don’t consider the toy train curvature and truck mounted couplers as “models” to be standardized. And yet strangely they had NMRA approved truck mounted couplers for HO and N that were ridiculous in most respects compared with Lionel’s slightly oversized 1945 knuckle.
Anyways, most of us would need a large barn to even model a sharp real world curve like Altoona in O Gauge. And string lining of real trains happened there too. 
Becky’s ‘issues’ with model train make up carry through to the 1:1 scale operations - long, light, high CG cars with relatively high amount of trailing tonnage are all but guaranteed to derail to the inside of curves - in the business it is known as a ‘Stringline’ derailment. It happens much too frequently in the real world when terminal supervision gets lax in observiing train make up rules.
Great suggestion. Would never have thought of it. Sounds practical. All I need to BUY now is a fishing scale and learn how to use it. I guess the buying never ends!
Trial and error are generally free at least if done gradually under careful observation. 
An alternative approach is to make a weight setup, with a pulley at the end of the test track and fishing line or string running from the engine around 90 degrees of the pulley and down to a platform holding variable weight. You can adjust weight with sand or other material until the engine is observed to stall or slip, then weigh that on a cooking or postal balance to get the result.
Yup. That’s another creative idea that might be fun to do too.