Hi all,
I am new here. This is my first post I am a software engineer and I just got into building model trains as my hobby.
I am building my first model rail using a mix of 3D printing, arduino and stepper motors. I am using steel wheels that I ordered off of eBay. My scale is HO (1:87).
I have some concerns regarding my model's "realism". First off, I have ordered numerous branded and unbranded wheels and noticed that the wheel profiles don't match even vaguely to real train wheels. In real trains, the wheels' shape is responsible for steering the trains, the flanges are just a safety device to prevent them from de-railing, but, all of the model wheels I've tried don't have the required conical angel to steer the train even on a super large turning angle. Here's a schematic of a real train wheel.
Here are some sample wheels I ordered and didn't find it to have a satisfactory conical angle.
This means, our wheels will have a lot more wear and tear especially at the flanges unlike a real train wheel.
Here's my first question: Do you have any ideas to solve this problem? How did you achieve conicity in your models? Without decent conicity, if a train relies solely on flanges, it will have a higher probability of derailment at high speeds.
Next, let's talk about the traction motors. In a real train, the traction motors are coupled to each wheel like this:
The advantage of this pinion type attachment is when the train stops accelerating, it will continue to keep moving due to inertia and the traction motors generate electricity that can be sent back to the grid.
However, in almost all model trains I've seen, even in the expensive ones - they always use a single large motor with a worm drive as shown below:
The problem with worm drive is when you stop accelerating, the train simply stops immediately. That is so unrealistic. The other problem, like in the image above, since all the inertial energy is absorbed by the worm gear, it wears out faster like seen above.
My second question: Do you know of any train models that don't use worm drives? What is your opinion on them and I would certainly love some advice on this.
Thank you for reading so far and for your valuable time ^_^. Looking forward to your replies!
To answer your second question, put a flywheel on your motor.
This will give you the realistic " coasting" motion your looking for.
As to the first question, I think your overthinking it. I buy kaydee wheels and they work great.
And at 1/87 scale, no one will really notice the wheel profile.
Rust...... It's a good thing !
"One difference between pessimists and optimists is that while pessimists are more often right, optimists have far more fun."
Hi thaoylee,
First, welcome to the forums and to the hobby!!
Without wishing to discourage you, I have to agree with others that your goals, while admirable, will not accomplish much as far as running models is concerned, especially HO scale models.
If existing model train wheels didn't work properly then the design(s) would have been changed decades ago. The fact is that model train wheels work fine (assuming that they are gauged properly and the track is laid properly). There is no reason to reinvent the wheel, pardon the pun.
Having said that, some wheel sets work better than others in some trucks, but the issues are nothing to do with the wheel profiles. Instead, the differences are in axle lengths, axle materials, and the profile of the tip of the axle. Older so called 'pizza cutter' wheel flanges are an exception. The flanges are too large to run over most Code 83 switches.
As far as how the drive systems are configured, I think the lack of space in smaller scale locomotives has thus far precluded the use of transverse mounted drive motors. A transverse motor would have to be tiny to fit between the wheels. Getting ehough power out of such a tiny motor would be a (likely very expensive) challenge. Small stepper motors might fit, but do they have enough power to pull a long train? At the very least, I suspect that every axle would have to have its own motor. That creates complications with speed matching and makes controlling 4, 6, 8,12....... or however many motors there are in a consist a challenge.
All in all, I think you would get much more enjoyment out of the hobby by using existing wheel and drive technology, and concentrating your skills on building the shells and detail parts instead. Of course, you are free to do whatever you want, so don't let us naysayers put you off.
Dave
I'm just a dude with a bad back having a lot of fun with model trains, and finally building a layout!
thaoyleeThe problem with worm drive is when you stop accelerating, the train simply stops immediately.
That's a bizarre statement. Even with my Tyco F7 60 years ago, when you stop acceletating, the train doesn't go any faster but it doesn't stop.
With DCC, if you cut the trottle to 0, you can adjust the momentum to a quick stop or a gradual stop.
As far as your concern about wheel configuration and wear. Arrowhead Models claims to have a prototypical wheel set. However for those of us who aren't running a commercial 12 hour a day model railroad, like the Colorado Model Railroad Museum, will probably never had to replace a worn out wheel set.
Henry
COB Potomac & Northern
Shenandoah Valley
If you think the wheel treads are poorly scaled, wait until you look at typical drawn-nickel-silver railheads. I suspect you will be on the track gleaming bandwagon very soon. Order your 3M lapping film accordingly...
The classical reason for worm drives is that they are the easiest way to rotate the torque of a longitudinal motor to turn axles. Other approaches are to use a transverse motor, which has proven difficult over the years, or to use a pinion and crown gear, which has much smaller contact area and tends to be noisy and hard to lubricate to boot. Note that the use of a larger diameter multithread worm (see the Helix Humper as one example) improves the geometry and allows a higher motor speed for a given track speed, and skewing the thread of the worm and the teeth of the driven gear enhances the possibility of backdriving -- as does use of a good coreless motor.
The first great 'enabling technology' for worm drives was the introduction of electronic momentum throttles. The second was the use of backdriving tooth contact... but there are limits to how 'skewed' the drive can be with typical manufacturing tolerances. What was developed fairly early was the 'coasting drive', which put a couple of idlers on a rocking frame so the gear train could mechanically unmesh with the worm if not being actively driven by it. There are threads here that go into considerable detail on the pros and cons of that approach.
In the '50s there was actually a model fluid flywheel sold as a slippable clutch: you could adjust the stiffness of the coupling by changing to a different viscosity of oil. I still can't quite believe someone had the patience to put it together to sell. You could probably use magnetorheological fluid (out of later GM magnetically-adjustable shock absorbers?) for a more controllable version.
My best advise to someone like you is to buy a kit, get it running well and then go back in and make improvements. Just by changing the gears on some models you will get much better running.
First of all, Welcome aboard! Your first post is monitored and it will be a while before you can reply, but please stay with us!
If I may comment here:
Lastspikemike Weight doesn't scale. Neither does volume really. Strength of materials definitely doesn't scale. Prototypical scale model wheels would not work at all. The flanges would be all but invisible and very flimsy. ... The wheels do behave fairly prototypically in part because the weights are really, really low. (What's 400,000 lbs x 1/87?) Model train wheels have massive flanges and very low weight.
Weight doesn't scale. Neither does volume really. Strength of materials definitely doesn't scale. Prototypical scale model wheels would not work at all. The flanges would be all but invisible and very flimsy.
... The wheels do behave fairly prototypically in part because the weights are really, really low. (What's 400,000 lbs x 1/87?)
Model train wheels have massive flanges and very low weight.
I agree with most everything stated so far by our experienced modelers.
To elaborate just a little though, keep in mind mass and enertia don't scale down. The reduction in weight and mass is a cube root equasion. That is, you devide it by all 3 dimensions
Using Mike's example weight, 400,000: (200 tons) a reasonable locomotive weight.
400,000/87 = 4597.7 lbs, your X axis,
4597.7/87 = 52.847 lbs, Your Y axis,
52.847/87 = 0.6074 lbs. Your Z axis. 200 tons comes out to 9.71 ounces.
That, or a 100 ton car at some 4 3/4 oz, won't push a scale train outward in a curve up onto the conical shapes of it's wheels at a given speed like a real train.
That's why, as others have mentioned, the illusion of mass and momentum have to be achieved by other tricks.
Flywheels don't even really give the momentum required to proportionally convey the weight of trains. A real train can take thousands of feet to stop. A generous flywheel model can stop, at least smoothly, in say around 80 to 120 or so. I'm a big believer in how SMOOTHLY flywheels make our engines run though.
That's just the weight and momentum aspect of the discussion. The manufacturers of quality model trains have had the bugs reasonably worked out of the rest of the aspects too, pretty much. Our flanges are massive, wheels wide, (most) couplers too big, and a few other misproportions. But this keeps them on the rails and operating smoothly.
There are "Proto-87" (HO) or other scale groups that do concentrate on getting these visual misproportions much closer to scale. But they're otherwise still dealing with the same "mass" disproportions.
If you are bringing other technologies or innovation into the hobby, that's great. There's always room for improvement and we'll love to see whatever you bring, new or traditional.
In all practicallity, rrebel nailed it! Dan
Didn't Tangent or exactrail try that a few years ago? I remember somebody trying to do proper tread contours. Haven't seen them in a few years though.
shane
A pessimist sees a dark tunnel
An optimist sees the light at the end of the tunnel
A realist sees a frieght train
An engineer sees three idiots standing on the tracks stairing blankly in space
Code 110 NMRA RP-25 wheelsets operate flawlessly on well laid trackage.
Wheels and rails are the meat and potatoes of model railroading. Functionality is far more important to me than scale fidelity.
-Kevin
Living the dream.
SeeYou190 Code 110 NMRA RP-25 wheelsets operate flawlessly on well laid trackage. Wheels and rails are the meat and potatoes of model railroading. Functionality is far more important to me than scale fidelity. -Kevin
Hi everyone, and thank you all for your wonderful inputs, that was very insightful. I didn't know of this NMRA standard for wheels, thank you so much. I am going to order a bunch of these and try them out.
As for the traction motor, I am going to take all your advice and proceed with a worm drive version for my second engine's bogie. My first engine's bogie is now built (kind of already before I could get your replies) with this tiny little stepper motor:
I chose these motors because they resemble real traction motors and perfectly sit inside my HO scale wheels with a rack and pinion type gear. But, will they be able to pull all that load is a big question mark I am yet to test. I will test them and post results here.
As for the braking, I thought about this - I know worm drives offer better instant braking support, but I have managed to recreate real train brakes using linear drives using electromagnets. So, the wheels will actually have something scraping them off just like with real trains. I will post more pictures once this is fully functional.
I tried some micro DC motors (brushed), but the problem with brushed DC motors in general (in a rack and pinion setup) is you can't get realistic ultra slow speeds and exponential acceleration curves like you can simulate with a worm drive. Brushed DC motors behave differently under different loads Here's the motor model I tried:
So, here I am, using the stepper motors above and I am using 6 of those for my engine. Just like in a real rail engine. I will update here of my progress once I have something to demo
You might want to look into RailFlyer Model Prototypes. They were a modeling outfit years ago that attempted to deliver a product similar to what you are doing, with traction motors individually powering each axle. The company went under before they could bring the models to market, but the development forum posts remain on various sites.
The original wheels you posted don't really conform to North American modeling standards besides the one on the right. Most wheels that I've seen in the North American market do have a slight taper to the wheel treads, while the wheels you posted are from Chinese bulk manufacturing. Those parts can be useful, I have used some myself, but are not standard fare for models here.
I think it would be helpful for you to buy yourself a ready-to-run North American locomotive model just so you have an idea of what a standard locomotive model looks like. An Athearn or Kato diesel would be good for that, they have kind of set the standard for many mechanical design choices.
-Peter. Mantua collector, 3D printing enthusiast, Korail modeler.
Looks like what he is trying is similar to a Stanton drive.. it works in G scale, if he gets it to be reliable In Ho. That would make it interesting.
Little Timmy To answer your second question, put a flywheel on your motor. This will give you the realistic " coasting" motion your looking for.
Not in my experience. I have a wide range of engine types, steam and diesel, across at least six different importers, and none of them coasts worth a darn when power gets cut to them. They stop dead short. I am almost of the impression that this behaviour is worse FOR HAVING the flywheels because when power cuts they torque up the drive very quickly...probably not doing a lot of good. If I'm wrong, I'd happily hear how, but doesn't that also suggest that they don't do much good the other way either?
I agree with your other comments. Machst nicht for all intents and purposes because the only problems I get with speed still depends on gaps at my swing-up bridge if they're not aligned properly, or if I fail with points lining. My trains track well, every one of them.
selectorI have a wide range of engine types, steam and diesel, across at least six different importers, and none of them coasts worth a darn when power gets cut to them.
All my diesels have flywheels, and none of them coast very well when power is cut off. The flywheels will get them over tiny dirty spots without stopping, but they still hesitate.
I do have this beast, but I have not installed a body on it yet.
-Photograph by Kevin Parson
The chassis on the left is a Proto Power West chassis with a coreless (whatever that means) "Micro-Motor" and two massive flywheels. This bad boy will coast about 12 inches when power is cut off.
thaoylee SeeYou190 Code 110 NMRA RP-25 wheelsets operate flawlessly on well laid trackage. Wheels and rails are the meat and potatoes of model railroading. Functionality is far more important to me than scale fidelity. -Kevin Hi everyone, and thank you all for your wonderful inputs, that was very insightful. I didn't know of this NMRA standard for wheels, thank you so much. I am going to order a bunch of these and try them out. As for the traction motor, I am going to take all your advice and proceed with a worm drive version for my second engine's bogie. My first engine's bogie is now built (kind of already before I could get your replies) with this tiny little stepper motor: I chose these motors because they resemble real traction motors and perfectly sit inside my HO scale wheels with a rack and pinion type gear. But, will they be able to pull all that load is a big question mark I am yet to test. I will test them and post results here. As for the braking, I thought about this - I know worm drives offer better instant braking support, but I have managed to recreate real train brakes using linear drives using electromagnets. So, the wheels will actually have something scraping them off just like with real trains. I will post more pictures once this is fully functional. I tried some micro DC motors (brushed), but the problem with brushed DC motors in general (in a rack and pinion setup) is you can't get realistic ultra slow speeds and exponential acceleration curves like you can simulate with a worm drive. Brushed DC motors behave differently under different loads Here's the motor model I tried: So, here I am, using the stepper motors above and I am using 6 of those for my engine. Just like in a real rail engine. I will update here of my progress once I have something to demo
With everything being miniturized today it will be interesting to see if you are successful with a true traction motor. Would be neat to see a six axle diesel with six motors. Good luck, and keep us posted with your tests.
Ray
This is not true of most electric motors nor is it possible in most circuits containing electric motors.
Keviin, from you photos, I believe you.
SeeYou190 This is not true of most electric motors nor is it possible in most circuits containing electric motors. -Kevin
selector Keviin, from you photos, I believe you.
Rich
Alton Junction
SeeYou190 This is not true of most electric motors nor is it possible in most circuits containing electric motors.
any armature turning within a magnetic field produces a BEMF, even when driven by an exernal voltage.
in a DC motor, the BEMF reduces the effective voltage across the armature so that the current is limited to the difference in voltage/winding resistance instead of the applied-voltage/winding resistance.
under a heavier load the motor slows, the BEMF decreases allowing the current to increase along with torque to compensate for the increase in load. similarly, the current, torque decrease when the load decreases (e.g. going downhill) as the motor speed and BEMF increase.
generator shafts are turned externally, resulting in BEMF and power output. the torque required to turn the shaft of the generator is ~proportional to the power (V*I) produced, hence electrical power output is ~proportional to mechanical power input
unpowering a motor and shorting its output terminals results in a braking action that will cause the motor to reduce its speed. of course the braking force is proportional to the speed of the motor. shorting the output terminals won't prevent a motor from turning
of course instead of wasting the output power as heat (e.g. dynamic braking) by braking, that power can be used to charge batteries -- regenerative braking.
1088
greg - Philadelphia & Reading / Reading
I guess the rules of physics have changed since you wrote this:
Anyway, another new participant has had his thread SPIKED!, way to go. Par for par.
At 12 volts then immediate cut-off an Athearn TrainMaster will go 3.7 inches. A GP-35 will go 4.5 inches.
I have a Scale Trains SD45 with duel fly wheels & dule capacitors, although it is programed for momentum yard switching and braking operatons i too have not found the engine to roll more than a few inches regardlees by decresing the speed manually. My ATH Genesis 2 Engines do the same. Steam and diesel powered turbines power electric generators (rotors) will contine to spin down (inertia) when the turbine power is cut off, any subsequent electric power generated decreases accordingly out to a main transformer. An electric motor works in a similar manner, except there is no main output transformer. Instead installed electric capicitors store a minute amount of motor generated power for intermittent use until the main power suppy is re-activated, such as an engine running over a dirty or problem section of track. In my opinion the concept of converting real life railroading into model railroading is a subjective "state of mind" no matter how much we try to acheive reality, trains, planes, ships, trucks, automobiles, etc.
I have a Scale Trains SD45 with duel fly wheels & dule capacitors, although it is programed for momentum yard switching and braking operatons i too have not found the engine to roll more than a few inches regardlees by decresing the speed manually. My ATH Genesis 2 Engines do the same. Steam and diesel powered turbines power electric generators (rotors) will contine to spin down (inertia) when the turbine power is cut off, any subsequent electric power generated decreases accordingly out to a main transformer. An electric motor works in a similar manner, except there is no main output transformer. Instead installed electric capicitors store a minute amount of motor generated power for intermittent use until the main power suppy is re-activated, such as an engine running over a dirty or problem section of track. In my opinion the concept of converting real life railroading into model railroading is a subjective "state of mind" no matter how much we try to acheive reality, trains, planes, ships, trucks, automobiles, etc. Bayway Terminal NJ
My first engine's bogie is now built (kind of already before I could get your replies) with this tiny little stepper motor:
"I chose these motors because they resemble real traction motors and perfectly sit inside my HO scale wheels with a rack and pinion type gear. But, will they be able to pull all that load is a big question mark I am yet to test. I will test them and post results here."
Hi thayolee,
I want to apologize for my somewhat negative response to what you are trying to do. I hope I didn't discourage you. I am really curious to see how much power the stepper motors can put out, as well as how smoothly they will operate.
Do you know what the draw and max. rpms are on the stepper motors?
Cheers!!
hon30critter My first engine's bogie is now built (kind of already before I could get your replies) with this tiny little stepper motor: "I chose these motors because they resemble real traction motors and perfectly sit inside my HO scale wheels with a rack and pinion type gear. But, will they be able to pull all that load is a big question mark I am yet to test. I will test them and post results here." Hi thayolee, I want to apologize for my somewhat negative response to what you are trying to do. I hope I didn't discourage you. I am really curious to see how much power the stepper motors can put out, as well as how smoothly they will operate. Do you know what the draw and max. rpms are on the stepper motors? Cheers!! Dave
Hi everyone,
Thank you again for your replies. I can confirm that DC motors DO product voltage when the train/bogie is coasting. Enough to light up some LEDs!
As for the motor specifications that you had asked for, here's the info I was able to find online:
Holding torque: 5gf.cm (3.3V to coil at 0 pps)
Voltage: 3.3V
Hope this answers your question.
I chose stepper motors because they have many useful properties that resemble real world traction motors - A real world traction motor is usually DC (depends on the country, make and model). But, even the AC ones behave very similar to a stepper motor. This makes stepper motor as close to spec as possible in this given size.
I am driving these with a gear ratio of approximately 1:3 with pinion and gear setup at 5V. So, I'm making roughly about 15gf-cm per motor at 3.3V and probably close to 25gf-cm per motor at 5V. For 6 motors that's about 150gf-cm of force. Again, these are theoretical values, I'm still waiting for my stepper motor drivers to be delivered so I can run proper tests to see how much actual weight they can pull.
I'm using CNC milled Aluminium chassis for 70% of my train's construction, this is by design because again, I want it to add heft to my model, so it is as realistic as possible. I have also ordered NMRA spec wheels. I feel the flange's friction will greatly affect my stepper motors' ability to pull the load.
Until next time,
Thank you