Flywheels in small 4mm shunting locomotives

[bertiedog]

Flywheels are always beneficial, if properly fitted and as large as practical. They overcome magnetic cogging of the poles of a motor, and with coreless motors add to the momentum considerably.

Theres no need to test or prove it, the use dates back 80 years and encompasses both hobbyist  and professional productions.

The gear ratio does not have a dramatic effect, witness the Hobbytown diesel designs that can coast several inches at slow speed and feet at higher scale speeds.

The momentum overcomes pickups troubles, and generally the loco will run quieter with a flywheel.

The big but is the bearings must be substantial, and nothing must be out of balance.

 

Stephen

[hayfield]

I am interested how you can discern the superiority of the flywheel. I ran one of my own locos (Judith Edge Yorkshire DE2 with Mitsumi motor and HL Loloader 90:1 box) with and without a  (12mm) flywheel and I honestly could not see any difference.

 

Ruston

 

I bought a Bachmann Jinty which had a Comet chassis fitted, using a DS10 motor fitted with a 12 mm diameter flywheel 7 mm thick, the fly wheel does make a difference, I would also expect performance over dirty track is inhanced. There is also a big chunk of metal over the front wheels, I guess for added adheasion

 

The flywheel is not solid as it has a 6 mm hole drilled half way through, is this improving the effect? 

Edited January 26, 2018 by hayfield

[AndyID]

 

The flywheel is not solid as it has a 6 mm hole drilled half way through, is this improving the effect? 

 

Hi John,

 

No, it won't help. It only reduces the total mass of the flywheel. It reduces the amount of energy stored slightly, but not enough to make any significant difference. In a large flywheel it saves a lot of metal and reduces the "dead load" on the bearings, but in in a model it's a waste of potentially useful adhesive weight.

 

Cheers!

Andy

[rope runner]

I have always found them beneficial - even small ones. Although I always use a high level gearbox - nothing else comes close for me. 

 

Paul A, 

[Regularity]

Sorry, a bit late to this one...

 

The flywheel is not solid as it has a 6 mm hole drilled half way through, is this improving the effect?

 

 

No, it won't help. It only reduces the total mass of the flywheel. It reduces the amount of energy stored slightly, but not enough to make any significant difference. In a large flywheel it saves a lot of metal and reduces the "dead load" on the bearings, but in in a model it's a waste of potentially useful adhesive weight.

Andy

Andy’s point should be more widely known.

I did write a brief piece for MRJ some years back (when they still had a permanent editor!) but the then incumbent list the piece.

 

It is true that, for a given diameter, a “hollow” flywheel will outperform a solid one, but only if their mass is identical. So, unless when boring out that 6mm hole the brass was not removed but pushed into the rest of the metal (in which case a Nobel Prize probably awaits the manufacturer!) boring out the flywheel does indeed concentrate the remaining mass towards the rim, but unfortunately at the expense of reducing the available mass. It is possible to work out the effective average radius from the centre of the flywheel to the effective centre of the rotating mass, and to compare these two with the resulting mass. There is no situation where boring out the mass is beneficial to the momentum of the flywheel. But A level physics is nearly 35 years behind me, and having done it once, I have no wish to do it again - besides, if you have the mathematical/physics knowledge to understand the calculations, you know this to be true. And if you don’t know this to be true, the maths might frighten you away!

 

More practically, about 20 years ago I had a small 0-4-0 loco frame under construction, which was driven by a proper (round) 1620 Mashima motor through a DJH 53:1 two stage gearbox. I set up a length of track with an insulating gap partway along, and simply applied 12V and let it go. I tried it without a flywheel, with a “hollowed out” flywheel, and with a solid one. They were the same length (10mm) and the same diameter (20mm - a bit big, maybe) with 10mm bored out of the “hollow” flywheel for about 8mm.

 

The results were interesting!

1, no flywheel. The frames went a lot faster - very little inertia to overcome - but also stopped quickly after the gap, with negligible overrun.

2, hollow flywheel. The motor took longer to get to full speed at full volts, and indeed was a touch slower, but was smoother and had (from distant memory) a couple of inches more overrun.

3, solid flywheel. Overcoming the rotational inertia of a directly mounted flywheel which is of greater diameter of such a small can motor was possibly a bit unfair, and this setup never acquired the same speed as the “naked” version, but it was the smoothest running of the lot. It also overran by 50% more than a hollow flywheel. I remember the 50% more than I remember the actual distances. Having subsequently done the calculations, I should not have been surprised.

 

I have always found them beneficial - even small ones. Although I always use a high level gearbox - nothing else comes close for me. 

 

Quite the best solution!

 

Fitting a flywheel should make for smoother running, removing some jerkiness resulting from less than absolutely perfect track and such things as wiper pick-ups. Only at high speed will you gain any noticeable momentum, and the aim should not be to glide over dead frogs or similar, just to smooth things out when the voltage supply gets low. Even with feedback controllers, if the current us interrupted then the motor stops turning.

Conversely, it is possible for a flywheel to unbalance the motor and make things very much worse - for example, you can mount a flywheel which is perfectly balanced onto a shaft which has slight bend in it. In such cases, bin the motor.

 

However, since I am in the process of moving away from DC, then some of the reasons for this will disappear.

My intention was to go towards DCC, which has 3 distinct advantages:

1) with its constant supply voltage, issues relating to low voltage over the rails go away;

2) feedback control is from the decoder directly to the motor, and not determined via a long route of copper wire, nickel silver rails, wheel to rail contact and wheel to pickup contact;

3) the adding of energy storage capacitors, even small ones, means that if the supply is interrupted by a speck of dirt, there is no problem.

Given a good motor and a decent gearbox, I am not sure if a flywheel is warranted - but I have them already fitted so they shall remain!

 

That said, I am moving towards on-board battery power, with direct (wireless) control. I am working through lots of permutations at the moment re charging the batteries (via rail, via sockets, even via induction loops) but that is by the by: anyone who has seen a small, light, 4 wheeled loco moving under battery power (whatever the control system) is immediately impressed by the uncanny smoothness that an uninterruptible power supply creates. It is almost eerie.

 

So, to answer your original question, David, if working in DC I would not consider going without a flywheel, and would fit a solid one, preferably 10mm long and to about the same diameter as the “round” dimension of the motor, so for a 1420 flat can that would be 16mm. If that were not possible, I would fit the largest I could get in, sacrificing length before I sacrificed diameter, and going down a step in motor size rather than not having a flywheel at all. If working in DCC or on-board battery power, I would still want a flywheel, but if push came to shove, would be prepared to go without.

 

As an aside...

The equal-smoothest running engine I have ever had, the one which set the standard for me, was a Triang-Hornby Ivatt class 2, with an XO4 motor and 20:1 gears with a double-start brass worm. Work that one out! Always liked that motor, but it was a bit bulky and there wasn’t much room for a flywheel. They stopped using them because one of the two armature winding machines they had broke down, and in the cash-strapped 70s they couldn’t - or wouldn’t - buy a new one, so started to outsource the supply of motors. They couldn’t make enough motors in-house, but also weren’t able to spend the money on a new machine. First the dreadfully implemented “ring-field” came along (it could be a good motor, if properly engineered), then those tiny slotcar motors, etc. Triang, when it was the world’s largest toy maker, was like the LNWR: it outsourced as little as possible, although it didn’t make its own bricks or steel. When you outsource, you lose the inherent “ownership” of the production - it’s just stuff for someone else - and no amount of quality control checking will make up for that. Plus, you have to take into account the supplier’s need to make a profit.

 

Anyway, rambling aside, I hope that helps.

 

PS If anyone wants an original, tested-only Dynadrive clutch to fit onto a 2mm shaft, send me a PM and I will see how much you are prepared to be extorted.

[bertiedog]

The point about a solid flywheel is it has greater mass and the out of balance forces plus gravity, tend to hammer the bearings.

Hollowing out the flywheel reduces  the overall mass  and leaves the effective flywheel effect, but with less wear on the motor bearings.

With ball raced motors use solid flywheels....

Stephen

[AndyID]

Hollowing out the flywheel reduces  the overall mass  and leaves the effective flywheel effect, but with less wear on the motor bearings.

 

Hi Stephen,

 

That sounds like a rather "wavy arm" generalization. How much will hollowing out the flywheel reduce the wear on a model locomotive's motor bearings? Do you have some test data you can share with us?

 

Andy

[bertiedog]

Straight forward engineering science, I have no particular references to quote, but the effects taught in college and university studies. Above all is practical experience  on working will full size flywheels on both Steam engines and  Electric motors.

The heavier the flywheel the better, but with a high mass then gravity forces the weight down on the shaft bearings more as the mass increases.. The lightening by hollowing reduces the wear, it also lowers the flywheel effect, but the max diameter mass is doing most of the work, the loss is proportionally small when hollowed out.

Stephen.

[Regularity]

I think that given the small masses involved here and hardened steel shafts, and the infrequent low usage on most models, this may be a case of theoretical concerns and practical experience of full size machines not being particularly applicable to our situation.

[bertiedog]

It's not the shafts that wear, it is the sintered bronze motor bearings that wear oval that is the cause for concern. They are initially soft, but work harden in use and surprisingly large wear can happen  in running in, aggravated by any strain from the flywheel forces.

Although small, the forces are proportional, and still can cause wear that is preventable.

[Regularity]

I meant the shafts are quite rigid.

To quote yourself:

Straight forward engineering science, I have no particular references to quote, but the effects taught in college and university studies. Above all is practical experience  on working will full size flywheels on both Steam engines and  Electric motors.

 

No reference to, or practical experience of, this actually happening.

The heavier the flywheel the better, but with a high mass then gravity forces the weight down on the shaft bearings more as the mass increases.. The lightening by hollowing reduces the wear, it also lowers the flywheel effect, but the max diameter mass is doing most of the work, the loss is proportionally small when hollowed out.

 

With our small flywheels, the mass is not that great, and the loss is not proportionally small when hollowed out.

With our small flywheels, we need all the mass we can get as the space is limited.

If the flywheel is unbalanced, then yes, it will create problems but that will be immediately obvious due to noise and jerky running.

 

The only time I have had this was with a (commercially supplied) hollowed out flywheel, where insufficient metal had been left at the end to be able safely mount it squarely on the shaft.

[bertiedog]

It  all depends on the design,, the hollowing should be equidistant from each end  and  leaves a sleeve over the shaft as big as the thickness of the  flywheel. Some commercial designs are just shoddy.

This is all about getting the best smooth performance  out of a small shunting  loco. not an engineering thesis on Flywheels..

[hayfield]

I think that this is s case perhaps personal preference. Those who prefer the system being happy with minor improvements. Where as those who are sceptical looking for better gains

[AndyID]

I think that this is s case perhaps personal preference. Those who prefer the system being happy with minor improvements. Where as those who are sceptical looking for better gains

I think it is a good example of how effects don't scale. There are good reasons for reducing the interior mass of large flywheels but they don't apply at small scales. In other words, hollowing out flywheels at 00 scale is a waste of time.

 

I can run the numbers but I doubt if anyone will bother to look at them

[Regularity]

This is all about getting the best smooth performance  out of a small shunting  loco. not an engineering thesis on Flywheels..

An odd remark, coming from the only person who has made it so...

I think it is a good example of how effects don't scale. There are good reasons for reducing the interior mass of large flywheels but they don't apply at small scales. In other words, hollowing out flywheels at 00 scale is a waste of time.

I can run the numbers but I doubt if anyone will bother to look at them

I already did for my own benefit, so I’ll take a rain check on that...

[DCB]

Wrote almost seventy years ago, but still revellant. Draw your own conclusions.

COAL TANK 2.jpg

Now that is inspirational!   It seems to answer all the questions, separate bearings for the flywheel to stop the armature bearings being over loaded etc.

 

Maybe someone could send the Chinese a copy of the Coal Tank article for inspiration.

[TheGunslinger]

I will say that flywheels can help smooth motor running, but if you want solid performance over points and dirty track, nothing comes close to keep-alive capacitors. A far smaller package (the TCS KA-3 is 25 x 12 x 8mm) that can run a locomotive for a good 5-10 seconds completely unpowered. Keep in mind there are two kinds; very, very small keep-alives that essentially maintain momentary power to the decoder to prevent the infamous "stammer" of a sound equipped locomotive on bad track, and the larger ones which are capable of fully powering the entire locomotive for a short period, motor and all. I've only used them in DCC and swear by them, but by their very nature they should be adaptable to DC applications as well. In fact for a DC application combining them with a flywheel would be very beneficial as then the motor only requires very minimal power input to maintain motor speed.

https://www.youtube.com/watch?v=zxEQDVI6c3E
 

Edited February 26, 2018 by TheGunslinger

[Ian Kirk]

Not quite on topic but the manufacturers of O gauge RTR shunters seem to  have opted to include a flywheel.  I was at the Glasgow Show at the weekend with the Gauge O Guild Stand where we had "Have a go in O" a 6ft long shunting plank where we were letting some "juniors" do a bit of shunting with a Dapol 08.  Flywheel plus low gearing meant that acceleration was limited yet did not introduce too much of a run on and even the youngest of our trainee drivers seemed to manage not to ram the buffers. Some of the parents panicked a bit though. The loco seems to have survived unscathed which after three days of amateur operation is a tribute to its design/manufacture.

 

best wishes,

 

Ian

[t-b-g]

I have built locos with and without flywheels and it is quite possible to get either to run well. There is just an indefinable something that the flywheel fitted ones have. They just respond to the controller slightly differently and are more "touchy feely" in their operation.

 

I can't measure it or say why. They just do! There is an unexplainable pleasure in switching the power off and seeing the loco run on, even for a couple of inches.

[cctransuk]

I have built locos with and without flywheels and it is quite possible to get either to run well. There is just an indefinable something that the flywheel fitted ones have. They just respond to the controller slightly differently and are more "touchy feely" in their operation.

 

I can't measure it or say why. They just do! There is an unexplainable pleasure in switching the power off and seeing the loco run on, even for a couple of inches.

 

It's because that's what the prototype does.

 

Regards,

John Isherwood.

[micknich2003]

And we moan about not being able to buy a motor, Mr Jackson (that Mr Jackson?) had to make his own!

 

Andy G

The same Mr Jackson who invented the coupling, he was by proffesion, an Electrical Engineer.

[Allegheny1600]

I will say that flywheels can help smooth motor running, but if you want solid performance over points and dirty track, nothing comes close to keep-alive capacitors. A far smaller package (the TCS KA-3 is 25 x 12 x 8mm) that can run a locomotive for a good 5-10 seconds completely unpowered. Keep in mind there are two kinds; very, very small keep-alives that essentially maintain momentary power to the decoder to prevent the infamous "stammer" of a sound equipped locomotive on bad track, and the larger ones which are capable of fully powering the entire locomotive for a short period, motor and all. I've only used them in DCC and swear by them, but by their very nature they should be adaptable to DC applications as well. In fact for a DC application combining them with a flywheel would be very beneficial as then the motor only requires very minimal power input to maintain motor speed.

 

https://www.youtube.com/watch?v=zxEQDVI6c3E

 

Hear, hear!

I would go so far as to say that a keep-alive capacitor of the same physical size as the flywheel will have a greater effect on smooth running than the flywheel will.

So, if you're building a small 0-4-0 shunter, leave out the flywheel and fit as large a capacitor as possible in its stead.

This is because a fully charged capacitor can store much more energy than the flywheel can.

Cheers,

John.

[AndyID]

Hear, hear!

I would go so far as to say that a keep-alive capacitor of the same physical size as the flywheel will have a greater effect on smooth running than the flywheel will.

So, if you're building a small 0-4-0 shunter, leave out the flywheel and fit as large a capacitor as possible in its stead.

This is because a fully charged capacitor can store much more energy than the flywheel can.

Cheers,

John.

 

Keep-alive caps work well with DCC, but it takes quite a lot of on-board electronics to use them with DC.

[TheGunslinger]

That's interesting. You may well be right in what you say. Another question springs to mind though. Will a worm driven locomotive that weighs more take longer to stop when the power source is removed when travelling on the flat at a given speed?

No, but it will put more strain on the gearbox when it comes to an abrupt halt!

 

One of the reasons why the High Level and Comet gearboxes command a pretty penny is because they use helical cut teeth on the worm wheel, which have a different moment of interaction and can apply some movement back through the gearbox.

 

I think I've actually inadvertently discovered why modern manufacturers use flywheels! The vast majority now run simple worm drives for their high gear ratio (the worm is effectively a 1-tooth gear, so combine with a single 40-tooth wheel for an impressive 40:1 reduction). When power is lost the loco will abruptly stop as momentum cannot transmit back through the gearbox and force the motor to continue turning. This in itself is already bad for prototypical running, but that jarring halt also puts a lot of force on the drivetrain. Remember how Bachmann models with nylon axles love to split in half and tear drive gears to pieces? It might be a whole lot worse without a flywheel on the motor.

[Regularity]

Hear that noise?

 

The sound of a penny dropping...