Magnet Drive instead of spur gears

[DCB]

I have been dreaming of Dyna Drive like propulsion. Basically a huge flywheel on the motor shaft driving the wheels through spur gears.  The problem with the early Dyna Drive was the spur gears stripped if the loco stopped abruptly as in hitting buffer stops, and the later ones with worm drive was not as free running which sort of spoiled the effect.   My devious plan is to use a couple of super neo ring magnets  as de facto spur gears between the motor/ flywheel shaft and the output shaft to the spur gear final drive.  I know the super neo ring magnets can take a plastic  bush to fit them to the shafts, that they turn each other just like spur gears or belt drive and deliver quite a bit of torque before slipping.  What I don't know is how long they will survive before losing their magnetism, especially if they slip.   

Bit of a long shot but has anyone experience  of using magnets for drives like this?   

If its feasible I also have plans for using a magnet train to drive the rear axle of a 14XX class loco and  possibly turn a turntable deck!

Edited November 20, 2019 by DavidCBroad

[Gordon A]

David,

That sounds like an interesting idea.

I know someone who uses a magnetic clutch on his crossing gate mechanism.

 

Gordon A

[Miss Prism]

The principle of dynadrive was having the clutch between the motor and the flywheel. If the clutch is placed between the flywheel and the final gear train, the motor will probably blow up when trying to accelerate the flywheel.

 

[Stanley Melrose]

You might find this website interesting <http://www.magnomatics.com/>.

 

Stan

[rab]

16 hours ago, DavidCBroad said:

I have been dreaming of Dyna Drive like propulsion. Basically a huge flywheel on the motor shaft driving the wheels through spur gears.  The problem with the early Dyna Drive was the spur gears stripped if the loco stopped abruptly as in hitting buffer stops, and the later ones with worm drive was not as free running which sort of spoiled the effect.   My devious plan is to use a couple of super neo ring magnets  as de facto spur gears between the motor/ flywheel shaft and the output shaft to the spur gear final drive.  I know the super neo ring magnets can take a plastic  bush to fit them to the shafts, that they turn each other just like spur gears or belt drive and deliver quite a bit of torque before slipping.  What I don't know is how long they will survive before losing their magnetism, especially if they slip.   

Bit of a long shot but has anyone experience  of using magnets for drives like this?   

If its feasible I also have plans for using a magnet train to drive the rear axle of a 14XX class loco and  possibly turn a turntable deck!

i suspect we'll get some polarised opinions on this topic

[RAF96]

12 minutes ago, rab said:

i suspect we'll get some polarised opinions on this topic

 

The topic has gone south already.

[Stanley Melrose]

As I was involved in the incorporation of  <http://www.magnomatics.com/> while I worked at Sheffield University, I had suggested/hoped that they might develop a version of their motor/gearbox for model railway use.  Unsurprisingly, the founders, with my support, were seeking large amounts of investment from major industries and model railways were a long way off their radar.

 

Over 40 years ago, I recall seeing the first model railway application of coreless electric motors - one by Bernard Laycock in a GWR 0-4-2T and another by Guy Williams in a King at Pendon.  These motors were developed by Faulhaber and Escap and came with prices consistent with their German and Swiss origins.  Now that the patents have presumably expired, they can be bought very cheaply from China and are commonplace in drones, etc.  I can't help wondering if the Magnomatics situation will be similar and that when their patents expire we will see wide-spread applications of the technology, including model railways.  EVs are obvious, as are wind turbines, microwave ovens, washing machines and tumble dryers but I fully expect model railway applications in due course.

 

Remember you heard it on RMWeb first!

 

Stan

[kevinlms]

12 hours ago, RAF96 said:

 

The topic has gone south already.

Or north!

[34theletterbetweenB&D]

On ‎20‎/‎11‎/‎2019 at 02:40, DavidCBroad said:

... What I don't know is how long they will survive before losing their magnetism, especially if they slip...

It's typically heating that causes loss of field. I would suggest the simple pragmatic test of putting one magnet on the motor shaft, aligning a fixed magnet, then run the motor at whatever will be full output for as long a period as you feel is a realistic risk, and see what happens.

On ‎20‎/‎11‎/‎2019 at 09:44, Miss Prism said:

The principle of dynadrive was having the clutch between the motor and the flywheel. If the clutch is placed between the flywheel and the final gear train, the motor will probably blow up when trying to accelerate the flywheel.

Magnetic couplings both ends of the flywheel then?

 

On ‎20‎/‎11‎/‎2019 at 02:40, DavidCBroad said:

...   If its feasible I also have plans for using a magnet train to  ... possibly turn a turntable deck!

Has been tried, and found problematic.

[Miss Prism]

2 hours ago, 34theletterbetweenB&D said:

Magnetic couplings both ends of the flywheel then?

 

No. Putting a clutch between the kinetic energy of the flywheel and the final transmission is what bedevilled the ModelTorque system. Because model trains require little force to accelerate them (after overcoming initial stiction), the torque transfer characteristic of any clutch on the output side is such that the train will accelerate far too rapidly after the flywheel exceeds a threshold rotational speed. Similarly, when the motor power is backed off, the flywheel rotational speed will diminish, and the train will come to a rapid halt because there is no longer sufficient torque transfer in the clutch on the output side. In other words, two clutches make the flywheel pointless. The advantages of a flywheel come in only if its rotation is linked directly to the final drive.

 

Dynadrive owners will know the best way to slow a train down is to put the motor in reverse. It's the only effective brake. Backing a big Dynadrive class 37 onto a train, just nudging the coach buffers to compress them a little bit, is great fun, but I'm not sure I would call it relaxing.
 

[polybear]

On 20/11/2019 at 20:46, Stanley Melrose said:

I can't help wondering if the Magnomatics situation will be similar and that when their patents expire we will see wide-spread applications of the technology, including model railways. 

 

Little details like Patents never seem to be high on the worry list of the chinese.....  

 

[billbedford]

On 20/11/2019 at 02:40, DavidCBroad said:

My devious plan is to use a couple of super neo ring magnets  as de facto spur gears between the motor/ flywheel shaft and the output shaft to the spur gear final drive.

 

I don' think this would work very well. With a single pole on each of the rings, you would get something that acted like a dog clutch, ie the magnets would either lock in sync or not, leading quite noticeable cogging. The cogging could be mitigated by using multi-poled rings, maybe with unequal numbers of poles.

 

However, an even better solution would be to use a single multi-pole ring and a plain aluminium disc. This will act as an induction motor and since eddy currents in the disc are induced by the magnetic fields of the ring the rotational speed transitions are going to be smooth. The downside of this arrangement is that the air gap between the ring and the disc has to be kelt to a minimum so some degree of fine engineering is required.

Edited November 22, 2019 by billbedford

[DCB]

Many thanks for all the replies.  Wow, simple bodgery to the transmission of the future in one page.   Actually my plan used the sides of the super neo magnets not the poles. I can't see why it wont work but I will build a test rig later and find out why!  If I'm right it could take all the precision out of spur gear trains as the magnets hold themselves together instead of having to mesh the spur gears accurately.   Plan is to araldite the axles to the magnets and let the idlers float with a lot of clearance to their guide pin.

[Stanley Melrose]

The first version of the Magnomatics gearbox that I saw can be described as follows:

 

Imagine a pair of spur gears in mesh.  As one gear turns the teeth progressively engage with the teeth on the second gear and transmit the torque.  Now, instead of teeth on the gears, imagine pieces of rare earth magnets but rather than having the "teeth" in mesh, they are separated by a minute air gap.  As each magnetic pole moves under power, it attracts the adjacent pole on the second gear which follows it until the air gap allows separation by which time the next pole is engaging with the next pole on the second gear and transmitting the torque.  This sequence continues just as in the toothed gears with which we are familiar.

 

In an application in a wind turbine, one of the major problems with toothed gear drives is that a sudden gust of wind can spin the blades so quickly that before they can be feathered, the gears break under stress putting the turbine out of action until a repair can be effected.  With a magnetic gearbox, as described above, a sudden gust would see the gears slip without damage.  Once the blades were feathered and rotation speed reduced the gears would function as desired once more.

 

HTH,

 

Stan

[billbedford]

Yes, the trick with Magnomatic gears is that each 'magnetic' tooth has the opposite pole to the ones on either side of it. The intention is that the areas of each wheel where the magnetic lines of force are tangential to the circumference of the wheel are kept to a minimum. It is possible to make such wheels by glueing thin magnets around a plastic wheel, and easier now we have 3D printing. But it's not easy. While there is no problem fixing all the magnets of one pole orientation to the wheel, fixing the others is always fraught since they will do their damnest not to be put in the place reserved for them.

[Stanley Melrose]

And Magnomatics found that keeping rare earth magnets of a size relevant to wind turbines tested the adhesive powers of available sticky stuff!

 

Stan

[RAF96]

I have lots of approx 5mm diam neos that my daughter uses as jewellery clasps. They have a hole in the middle hence the N  &S poles must be on the flat faces. 

Initially I was going use them as coach couplings, but now I wonder if they can be used in locos to replace cardan shaft Universal Joints.

Edited November 23, 2019 by RAF96

[34theletterbetweenB&D]

On ‎23‎/‎11‎/‎2019 at 14:52, RAF96 said:

I have lots of approx 5mm diam neos that my daughter uses as jewellery clasps. They have a hole in the middle hence the N  &S poles must be on the flat faces. 

Initially I was going use them as coach couplings, but now I wonder if they can be used in locos to replace cardan shaft Universal Joints.

Worth a trial. There are bound to be some wrinkles to sort out. Could be rather neat where the mechanism design leads to something of a fishing expedition to relocate the shafts in the couplings when bogies are being reassembled to the mechanism. Self aligning magnetic couplings would make life easier.

[AndyID]

On 23/11/2019 at 06:52, RAF96 said:

I have lots of approx 5mm diam neos that my daughter uses as jewellery clasps. They have a hole in the middle hence the N  &S poles must be on the flat faces. 

Initially I was going use them as coach couplings, but now I wonder if they can be used in locos to replace cardan shaft Universal Joints.

 

You can make a pretty good CV joint with thin-wall silicone tubing (the stuff they use for model aircraft fuel lines). Just force it over the ends of the two shafts you want to connect. You can use concentric square brass tube if you need the length to vary. Hexagonal tube is even better if you can find any.

[AndyID]

On 22/11/2019 at 03:11, billbedford said:

However, an even better solution would be to use a single multi-pole ring and a plain aluminium disc. This will act as an induction motor and since eddy currents in the disc are induced by the magnetic fields of the ring the rotational speed transitions are going to be smooth. The downside of this arrangement is that the air gap between the ring and the disc has to be kelt to a minimum so some degree of fine engineering is required.

 

The snag is that the torque transmitted is proportional to the speed difference between the input and output so unless there's a lot of slip you don't get much torque.

[billbedford]

6 hours ago, AndyID said:

 

The snag is that the torque transmitted is proportional to the speed difference between the input and output so unless there's a lot of slip you don't get much torque.

 

But maximum torque would be needed starting the loco, i.e. the opposite of what electric motor give, and similar to the way a steam loco behaves. Thinking about it an induction disc drive may not need the usual gearing, just a bevel or contrate gear to the dive axle. Though learning to drive such a loco may be a challenge to those who are used to 'turn-the-knob-and-go' type controllers.

[AndyID]

18 hours ago, billbedford said:

 

But maximum torque would be needed starting the loco, i.e. the opposite of what electric motor give, and similar to the way a steam loco behaves. Thinking about it an induction disc drive may not need the usual gearing, just a bevel or contrate gear to the dive axle. Though learning to drive such a loco may be a challenge to those who are used to 'turn-the-knob-and-go' type controllers.

 

Unfortunately some amount to gearing is unavoidable. Small electric motors can deliver quite a lot of power but they don't produce much in the way of torque. It's always best to use the largest diameter motor that will fit.

[RobjUK]

On 26/11/2019 at 06:35, AndyID said:

The snag is that the torque transmitted is proportional to the speed difference between the input and output so unless there's a lot of slip you don't get much torque.

 

That reminds me of a gadget I read about some decades ago - a kind of drive unit or "torque converter" using two closely spaced parallel conducting discs connected to the input and output shafts, in a casing filled with mercury and a strong end-to end magnetic field.

(I can't find the name for the overall system so far, google results are swamped with free energy junk...)

 

The whole setup forms a homopolar generator directly connected (via the mercury) to a homopolar motor.

 

If I remember the details in the book correctly, it gives both the direct magnetic drive and a torque increase as "slip" increases, in a similar way to a hydraulic torque converter - a pretty good system for a locomotive drive?

 

Obviously mercury is not a good idea but galinstan alloy could serve the same purpose, as long as an appropriate metal is available for the discs - aluminium is definitely not suitable for contact with gallium, for one example. Possibly a non-magnetic grade stainless steel would work?

 

With far stronger neodymium magnets now available, it could be an interesting experiment; it may be possible to scale it to 0 gauge, i don't know about smaller scales though.

 

[APOLLO]

What did the "Western" say to the "Warship" ?

 

"I can't Torque, I've lost my Voith" !!!!!

 

Brit15

[roythebus]

Mention of magnets for coach couplings, it's nothing new. In the late 1950s I got a "train set" from Santa a somewhere like Selfridges.

 

It was 3 coaches and an A4 style loco made from wood, painted and with wooden wheels, maybe TT size. the whole lot coupled together with round magnets in the corridor connections, not very strong ones, but they done the job.