Results 1 to 13 of 13

Thread: Why arenít locos rated by pull strength?

  1. #1
    Join Date
    Mar 2016
    Location
    Rockwall, TX, USA
    Posts
    11
    Blog Entries
    1
    Thanks
    0
    Thanked 2 Times in 1 Post
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Why arenít locos rated by pull strength?

    I know that the Right combination of weight and high engine torque means that lots of cars can be pulled. Havenít noticed loco reviews that comment on pulling power. Why not?

  2. #2
    Join Date
    May 2010
    Location
    Cincy-whatzit, Ohio
    Posts
    6,681
    Blog Entries
    3
    Thanks
    8,987
    Thanked 8,087 Times in 3,583 Posts
    Mentioned
    201 Post(s)
    Tagged
    0 Thread(s)

    Default

    I believe the reviews in Model Railroader magazine tend to give drawbar pull in their reviews, and equate that to X number of cars on straight, level track. But there are a lot of factors that can affect actual performance, such as the gradients and curves we have on our layout, how free-rolling the cars are, how we clean or treat our track (for instance, No-Ox-ID might improve electrical pickup but reduce the friction or grip) and so on. So these kinds of ratings are only meaningful in a relative way, to say that one locomotive pulls harder than another in a given set of circumstances.

    I suppose one metric that could be reported would be the average weight per axle. My Life-Like FA-1 is just a little beast of an engine, because it's almost a solid brick of chassis weight that comes down on 4 axles. My Atlas SD-35 on the other hand is a lot lighter, not having as much hood room for a big chassis weight, and it spreads that weight over 6 axles instead. Steam engines might have more axles still, and even less room for chassis weight. I wonder if there is a linear relationship between weight per axle and measured drawbar pull.

    Get yourself a Rail Pass for free travel on the WP&P: wpandp.com
    Could Star Wars: The Last Jedi have been a smarter movie with just one tweak? wpandp.com/how-an-interdictor-could-have-fixed-the-last-jedi/

  3. The Following User Says Thank You to WP&P For This Useful Post:


  4. #3
    Join Date
    Nov 2006
    Location
    Los Angeles, CA, USA
    Posts
    2,954
    Thanks
    1,188
    Thanked 2,742 Times in 969 Posts
    Mentioned
    32 Post(s)
    Tagged
    0 Thread(s)

    Default

    I don't complain because that gives me an excuse to MU.

    Metro Red Ln (Metro Red Line)
    Under the streets of Los Angeles

  5. The Following User Says Thank You to MetroRedLn For This Useful Post:


  6. #4
    Join Date
    Mar 2016
    Location
    Rockwall, TX, USA
    Posts
    11
    Blog Entries
    1
    Thanks
    0
    Thanked 2 Times in 1 Post
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)

    Default Correct.. Model Railroader does give drawbar pull

    Just checked my latest Model Railroader magazine, and there it was, drawbar pull in number of cars. Now if the manufacturers would do the same.
    Quote Originally Posted by WP&P View Post
    I believe the reviews in Model Railroader magazine tend to give drawbar pull in their reviews, and equate that to X number of cars on straight, level track. But there are a lot of factors that can affect actual performance, such as the gradients and curves we have on our layout, how free-rolling the cars are, how we clean or treat our track (for instance, No-Ox-ID might improve electrical pickup but reduce the friction or grip) and so on. So these kinds of ratings are only meaningful in a relative way, to say that one locomotive pulls harder than another in a given set of circumstances.

    I suppose one metric that could be reported would be the average weight per axle. My Life-Like FA-1 is just a little beast of an engine, because it's almost a solid brick of chassis weight that comes down on 4 axles. My Atlas SD-35 on the other hand is a lot lighter, not having as much hood room for a big chassis weight, and it spreads that weight over 6 axles instead. Steam engines might have more axles still, and even less room for chassis weight. I wonder if there is a linear relationship between weight per axle and measured drawbar pull.

  7. #5
    Join Date
    Mar 2011
    Location
    Windsor, Ontario, Canada
    Posts
    1,083
    Thanks
    1,663
    Thanked 2,371 Times in 602 Posts
    Mentioned
    19 Post(s)
    Tagged
    0 Thread(s)

    Default

    I've always thought the drawbar pull metric was a great way to compare the effective pulling power of an engine. Like WP&P said earlier, there's too many factors when trying to convert that to "how many cars" the engine can pull. A straight-up number measuring the pulling force is pretty easy to understand and translates to the bigger the number, the more cars it'll move.


    Quote Originally Posted by Eusjim View Post
    I wonder if there is a linear relationship between weight per axle and measured drawbar pull.
    Doesn't measuring drawbar pull negate the need to understand how many axles? The ability to pull is already a measure of grip/power to the rails, regardless of how many axles. Having said that, it would be an interesting relationship to see. Fun topic!
    Peter

    Layout Depot (share your designs with others): www.LayoutDepot.com
    My Build Thread: www.nscale.net/forums/showthread.php?28081-Green-Valley-Railway

  8. The Following User Says Thank You to pbechard For This Useful Post:


  9. #6
    Join Date
    Jun 2013
    Location
    Mill Bay BC
    Posts
    2,858
    Thanks
    4,790
    Thanked 4,024 Times in 1,477 Posts
    Mentioned
    78 Post(s)
    Tagged
    1 Thread(s)

    Default

    Mark @Spookshow usually gives a number of cars a loco can pull in his reviews at
    spookshow.net

    Steve

  10. The Following User Says Thank You to aflica For This Useful Post:


  11. #7
    Join Date
    Apr 2014
    Location
    southeast michigan
    Posts
    1,017
    Thanks
    446
    Thanked 1,158 Times in 462 Posts
    Mentioned
    31 Post(s)
    Tagged
    0 Thread(s)

    Default

    There are so many factors that affect pulling that it might be impossible to control all the variables. Even if you had a dedicated rail section to do it on over the years, and you followed a dedicated wheel and rail cleaning procedure prior to each test, I suspect you still wouldn't get consistent results from the same loco month after month because as you run the loco the wheels get pitted and that changes the tractive effort. Even the speed (pulse width) can affect it.

    Seems to me you can remove lots of noise by not actually pulling cars for the test. Every N scale loco I've ever seen has more than enough power to spin out the wheels under max load plus, so the "horse power" is immaterial. What might be easiest is a setup where you attach a line to the loco's coupler and run it over a coupler-height pulley at the end of a track so it can hang down off the desktop. Then you add weight to the line until the loco starts to skid toward the pulley. Only variables there would be wheel and track condition.

  12. #8
    Join Date
    Mar 2007
    Posts
    514
    Thanks
    6
    Thanked 847 Times in 309 Posts
    Mentioned
    35 Post(s)
    Tagged
    0 Thread(s)

    Default

    I've been doing tractive effort tests for about 15 years or so. I rigged up a spring scale as a dynamometer car (with trucks on it) and do both static pull tests at full slip, and also monitor 'drag' on a train with various wheel conditions. There were two key things I wanted to research that I wanted to collect data on:

    1) The impact of 'carving out' frames for decoders, decreasing weight, also going to lighter-weight cast alloys for frames - how it impacted pulling power
    2) The impact of what appeared to me to be 'slicker wheels' on certain models, that decreased pulling power even more.

    Finally, I wanted to determine just how much drag 'wheel gunk' contributed, and could be corrected with wheel cleaning, hence the dynamic testing ability up grades and through curves.

    It's been pretty successful approach. The biggest issue with me has been the coefficient of traction, and nobody seems to measure that but me. Remember that on a prototype, steel-on-steel, the 'go to' number is about 25% tractive effort to weight; with anti-slip computer controls, etc. you'll see that jacked up to as high as 30%. We don't get anywhere near that kind of performance in N, with plated wheels on nickel-silver. I've seen it go as low as 11% and as high as 22%. "worn" wheels down to brass on a heavy frame perform the absolute best, lightweight frames on 'slick' wheels perform the worst. I've also found a consistent pattern of performance by manufacturer, there really is a difference in wheel metallurgy that can be directly linked to resulting tractive effort. And, surprisingly, traction tires help, but not as much as you might think. And there's no magic solution to hollowing out chassis for decoders and speakers, and the added problem of decreased weight ALWAYS decreases electrical pickup, period. Physics, it's the law.

    There's a lot of BS about gearing, motor poles, torque; axles, etc., it has nothing to do with it. It's basically a factor of two things, weight and adhesion. Fewer points of contact 'may' result in better grip, so the idea that a six axle will outperform a four axle really doesn't translate, and in some cases is just the reverse; more wheels spread the load and don't grip as well, but it never dramatically increases results by adding axles.

    I do everything in grams, both for locomotive weight (using a scientific scale) and the spring scale on trucks, while the spring scale is not of scientific quality it is a rather consistent measurement between locomotives; i.e. you may disagree with the pull in grams but if locomotive X has twice the pull of locomotive Y, that still twice the pull no matter how you measure it. You simply divide tractive effort in grams by weight in grams to calculate the adhesion percentage.

    My tractive effort pull champs remain an odd lot:

    1982 Kato GP38, odd-on best puller for a diesel
    First-run Kato SD40-2 chassis in an FP45 shell, loaded up with more weight

    All original Kato six-axles are excellent pullers

    Kato consistently outperforms Atlas for adhesion, there really is a difference in wheel metallurgy. Best demonstrated by the Kato NW2 in testing, as it's not particularly a heavy locomotive yet grips better than heavier Atlas ones.

    Recent Atlas (most recent run) actually outpull some of the earlier runs, Classics generally are heavier than the DCC compliant ones and perform as such.

    For steam, the original Life-Like 2-8-4 was an absolute wimp. The champ remains a Kato 2-8-2 with traction tire and the GHQ cast pewter PRR L1 boiler, that thing is a complete beast.

    I've got an Excel spreadsheet I've assembled on everything I own or have tested, PM me if you want it, but my professional background is to take the data and measurements and let it speak rather than just throwing out theories and seeing if they stick. And because N scale is about long trains, I'm on a campaign to make sure that all the efforts to add everything but weight to a locomotive don't result in virtually everything available won't pull itself out of a wet paper bag. My continuous poking of Atlas on wheel adhesion may have hit a nerve, because I can tell the difference on current productions vs. about 3-4 years ago, the wheel metallurgy appears to have changed slightly for the better.

    Translating pulling power in grams to 'cars' is a different matter entirely, and I don't even get into that. I know what my layout will do for given grams of TE. What I can tell you is that locomotive X has three times the tractive effort of locomotive Y, even though Y is a six axle and X is a four, and X and Y are virtually the same weight. The worst 'makes no sense at all' is that the Kato 1982 GP38 outpulls the current Atlas GP38 by a factor of about 3:1. That's what mattered to me, and why my new Atlas GP38 is in the display case and the old Kato is still out there dragging trains around.
    Randgust N scale kits web page at www.randgust.com

  13. The Following 6 Users Say Thank You to randgust For This Useful Post:


  14. #9
    Join Date
    Apr 2014
    Location
    southeast michigan
    Posts
    1,017
    Thanks
    446
    Thanked 1,158 Times in 462 Posts
    Mentioned
    31 Post(s)
    Tagged
    0 Thread(s)

    Default

    Sounds like a good systematic approach. One thing though, but for comparison purposes it probably doesn't matter: pulling force goes down when the wheels slip, similar to spinning your car wheels on snow and ice.

  15. #10
    Join Date
    Nov 2017
    Location
    Ariege, France
    Posts
    387
    Thanks
    1,328
    Thanked 774 Times in 237 Posts
    Mentioned
    38 Post(s)
    Tagged
    0 Thread(s)

    Default

    @randgustWow, that is an excellent study and highly praiseworthy. Thank you for taking the time to share that with us.

    PM sent
    Thanks, Tom

  16. #11
    Join Date
    Dec 2014
    Location
    FL
    Posts
    206
    Thanks
    37
    Thanked 256 Times in 84 Posts
    Mentioned
    17 Post(s)
    Tagged
    0 Thread(s)

    Default

    [QUOTE
    ]Haven't noticed loco reviews that comment on pulling power. Why not?
    [/QUOTE]

    Any professional machinist will tell you that no two identical parts are exactly alike.
    Same with automobiles one right behind the next off the assembly line.
    The same goes for toy train engines.
    Model Railroader tests, or @randgust tests, or @Spookshow tests, or anybody else’s tests, only supply meaningful data to the ones that performed them.
    No valid conclusions can be drawn from a test sample of one or two from lot size of a few thousand, at least nothing scientific or meaningful.
    Without a controlled environment the tests are inconclusive at best.
    Take them all with a grain of salt, way too many variables to consider them an accurate
    representation of the entire production lot.
    Your own tests, in your environment, are just as accurate (or inaccurate) as everybody else’s.
    Last edited by donzi; 12th Feb 2019 at 05:23 PM.

  17. #12
    Join Date
    Mar 2007
    Posts
    514
    Thanks
    6
    Thanked 847 Times in 309 Posts
    Mentioned
    35 Post(s)
    Tagged
    0 Thread(s)

    Default

    There's probably more consistency between manufactured locomotives at least 'out of the box' than you'd think, because the inconsistencies are relatively miniscule. What changes, and changes a lot, are the factors of wear into performance. What's odd is that worn locomotives (at least in my testing) almost always outperform new.

    This entire TE study evolved from a weird set of circumstances when I got my first Atlas SD24. I normally rate a single B-B unit for about 8 cars on my layout eastbound up the 2.5% ruling grade. So I got it, brand new, out of the box, ran great, and began the run-in process. Put 8 cars on it, a few test laps, then set it to just 'run' and left the room for about half and hour. When I left it was happily running its train. When I got back it was stalled on the hill, sitting there and spinning. Backed it off, tried again. Couldn't make the hill. Had to knock it back to 5 cars. What changed?

    Out of the box, it had blackened wheels. Examination showed that they were now polished smooth. Could simply wearing off the blackening really result in cutting TE by about 30%? Did wheel adhesion make that much difference by itself?

    Then I tested 'new' vs. 'old' Trix U28's. The oldest ones with the plating worn off (down to brass) consistently outpulled a brand new one with plating by about 40%. Wow. OK, I may be on to something here. I started testing everything, measuring with the spring scale. More to this than I thought. I knew it was related to weight but wheel material and condition was something I'd never really heard anybody comment about.

    Back to the SD24, removed the rear headlight board and put in lead, and then deliberately took to the wheels with a needle file and took the layer of 'plating' off down to raw brass. Bang, now I'm back up to 9 cars again. And so that's how all this started. Then the observations of types of wheels started to kick in.

    My old Life-Like GP20 has excellent adhesion, but it also oxidizes up way faster and requires a good wheel cleaning about every 3 months. Different material, that's the tradeoff.

    There's no deep science here to simply weighing a model and checking the pull in the same units and dividing it out. And it's a really easy way to compare.

    My opinion that I can't necessarily back up is that the necessity for perfect pickup for DCC results in the need for a harder, slicker, wheel alloy that won't be as suspect to oxidation and stalling, but the result is not only a lighter model, but one that is driving on slicks. Add it all up and your tractive effort is dropping like a stone if you compare older production to current. Wheel metallurgy is a lot more significant than I ever thought.
    Randgust N scale kits web page at www.randgust.com

  18. The Following 3 Users Say Thank You to randgust For This Useful Post:


  19. #13
    Join Date
    Feb 2018
    Location
    SW USA
    Posts
    92
    Thanks
    6
    Thanked 182 Times in 55 Posts
    Mentioned
    10 Post(s)
    Tagged
    0 Thread(s)

    Default

    Supporting Donzi's post, and to give an idea HOW different two "identical" locos can be...

    My 2017 release Athearn Challenger would pull ~20 MTL freight cars max up the 1.8" grades (on 17" curves) on my small layout. Unfortunately (making a very long story short), it had an drive mechanism issue that required it to be sent off to the factory in China for refurbishment. After a long delay, Athearn sent me a replacement Challenger under warranty, but the new loco would only pull 8 cars up my grades... I THINK due the traction tires were too thin, and not making contact with the rails...

    As I was communicating with Athearn about how I might improve the mediocre tractive effort (the new challenger was "out pulled" by my Bachmann 2-8-0!), the repair tech told me that my original loco was due back from China any day... Not knowing how refurbishment might have affected performance, I asked if he could ship me the original loco for a head to head comparison, offering to pay a deposit from my CC).

    Athearn sent me the refurbished original challenger, with no deposit other than my word... it ran as smooth as it did when new, and still pulls ~20 cars around my layout, so I sent the wimpy replacement back. The point is... the tractive effort of two supposedly "identical" locos differed by more than a factor of 2.

  20. The Following 2 Users Say Thank You to NDave For This Useful Post:


Similar Threads

  1. Getting the most pull out of your locos
    By P-LineSoo in forum Motive Power
    Replies: 16
    Last Post: 19th Jul 2016, 10:46 AM
  2. T-Trak Structural Strength
    By KCS1894 in forum Modular Layouts
    Replies: 17
    Last Post: 8th Jun 2015, 10:59 PM
  3. Awwwwwwwwww, aren't they cute?
    By bc6 in forum Motive Power
    Replies: 10
    Last Post: 18th Dec 2012, 12:32 AM
  4. Bridge Strength
    By MikeS in forum Scratchbuilding & Kit Bashing
    Replies: 7
    Last Post: 17th Jun 2007, 12:10 PM
  5. Plastercloth strength?
    By jake in forum Structures & Scenery
    Replies: 3
    Last Post: 12th Mar 2007, 03:58 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •