the MRH Forum

Coreless Motors in Model Trains Locos and rolling stock
Reply
Prolific poster 8,843 posts railandsail Prolific poster 8,843 posts
railandsail

This was a recent posting over on the Repower/Regear groups.io   
I thought some folks here find it interesting.

Quote:
Link to a brief article on the Faulhaber website. 
 
 
In my model railroad club back in Belgium we have been remotoring engines -both DCC-equipped and not- with Faulhaber motors since the early '80s, to the utmost satisfaction: some are still going strong after nearly 40 years. It helps of course that the club's president was a Faulhaber importer/dealer...
 
Wouter  J.K. De Weerdt - SP in Mali

Brian

1) First Ideas: Help Designing Dbl-Deck Plan in Dedicated Shed
2) Next Idea: Another Interesting Trackplan to Consider
3) Final Plan: Trans-Continental Connector
Reply 0
herronp Prolific poster 1,847 posts
herronp

I have had mixed results with coreless........

......motors in O scale.  They are fabulous with straight DC but so so with DCC.  I think the reason is the Back EMF circuitry in the decoders gets "confused" as coreless motors don’t seem to generate much if any. You really have to play around with the BEMF settings a lot to get smoothness out of them.  I’m no electronic wizard but have shied away from them for this reason.  Some work well, others I can never get right.

Any electronic wizards out there that could shed some light on this issue?

Peter
Reply 1
forsyth.sandy New member 16 posts
forsyth.sandy
I'm looking at replacing brushed dc motors with coreless motors. The DCC Wiki tells me that coreless motors can only be used with Multifunction decoders. Is there a difference between loco decoders and multifunction decoders?

South African Railways fan
Reply 0
Ken Rice Super User 5,939 posts
Ken Rice
It’s not multifunction that’s important with coreless motors, it’s high frequency pwm drive.  Which almost all decoders are these days.  Back in the early days of DCC most decoders used a lower drive frequency which created an audible buzz or whine, and which could also cause problems for coreless motors.

My blogspot blog: http://rices-rails.blogspot.com/
My MRH blog index
Reply 1
Larry of Z'ville Frequent poster 984 posts
Larry of Z'ville
In last years motor study, I found a couple of excellent coreless motor choices.  These are used for tattoo guns normally.  Their torque & RPM functions are really good for HO engines.  The ones I tested were single shaft, but there are versions that are dual shaft.  

So many trains, so little time,

Larry

check out my MRH blog: https://model-railroad-hobbyist.com/node/42408

 or my web site at http://www.llxlocomotives.com
Reply 0
Marc Prolific poster 1,369 posts
Marc

I have Used Faulhaber motor in several N scale model

In any case excellent results with powerful and smoot running engines especially Old Atlas Rivarossi models

They comes from the 1016 and 1018 series 

Some  have gear head to have more slower  capabilities 

Decoders troubles had existed with old ones which worked with low frequency but for sure this didn't affect the quality of these motors and there's no more low frequency decoders on the market since a while 

Without hesitating I invite you to use them they are much over quality of many others careless motors

On the run whith my Maclau River RR in Nscale
Reply 0
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
don't coreless motors require 3-phase drive circuits?   
or are there different types of coreless motors?

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
Prof_Klyzlr Moderator 7,960 posts
Prof_Klyzlr
Dear Greg,
greg ciurpita  gregc wrote:
don't coreless motors require 3-phase drive circuits?   
or are there different types of coreless motors?
A: There are different types of "coreless" motors

https://www.faulhaber.com/en/products/dc-motors/

Also, for reference:

https://www.faulhaber.com/en/products/brushless-dc-motors/

https://www.promodeler.com/askJohn/talkRADIO/about-RC-servo-motors
(Permanent Magnet  VS Coreless VS Brushless)

Happy Modelling,
Aim to Improve,
Prof Klyzlr
Reply 0
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
do you know what makes the DC faulhaber motor different from more conventional coreless motors that require 3 phase drive circuits?

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
Prof_Klyzlr Moderator 7,960 posts
Prof_Klyzlr

greg ciurpita  gregc wrote:
do you know what makes the DC faulhaber motor different from more conventional coreless motors that require 3 phase drive circuits?
Sounds like you may be confusing "Coreless" with "Brushless". Both regular "permanent magnet" ("permag") and Coreless motors still have a hardware segmented commutator, which contacts a pair of brushes. As the motor rotates, the make/break of the brushes to the relative commutator contacts dis/connects the respective coils, thus sequencing the required magnet fields to invoke motion. Simply apply DC, and let the mechanical-switching commutator + amplitude of the incoming DC-->magnetic field strength and interaction work it out.
(The physical relationship between the 2x permanent magnets, the electromagnetic core/armature, and the commutator "switchgear" is fixed).

NB The reason Coreless motors are generally considered more fragile than regular permag motors is that the commutator coils and brushes are constructed with much more precision, using much smaller wire. Smaller wire = more susceptible to heat damage, either due to magnetic-field create/collapse (Vicious square-wave PWM abuse), or over-current conditions (stalling) .

"Brushless" in contrast does not have a hardware commutator and brushes, but uses a series of electromagnets around the outside of the permag core to invoke the motion, and a series of hall-effect sensors to work out how hard/fast/which-order to switch power into the electromagnets to spin the magnetic core. The means the motor itself carries circuitry to:
- receive the multiple Hall-Effect Sensor Inputs
- receive the "source power" input, and interpret speed+direction commands from that Input
- and discretely-drive each of the multiple electromagnets as required

https://en.wikipedia.org/wiki/Brushless_DC_electric_motor

For the astute, a brushless-motor without the onboard electronics would be very-similar, if not the same to what most would recognize as a stepper motor. Steppers have the individual electromagnet terminals exposed, and it's up to the User to connect them to a H-bridge or similar "stepper motor driver" circuit...
...Brushless motors just have the stepper motor driver circuitry built-in to the motor itself... ;-)

Happy Modelling,
Aim to Improve,
Prof Klyzlr
Reply 2
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
Prof_Klyzlr wrote:
Sounds like you may be confusing "Coreless" with "Brushless". 
probably
does coreless mean no iron armature?
if so, wouldn't an iron armature provide some inertia that would be a good thing on a model locomotive?

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
Prof_Klyzlr Moderator 7,960 posts
Prof_Klyzlr
greg ciurpita  gregc wrote:
probably
does coreless mean no iron armature?
if so, wouldn't an iron armature provide some inertia that would be a good thing on a model locomotive?
Elimination of the relatively bulky and heavy iron "core" lowers the mass and inertia, avoids "cogging" where the cores pole-pieces are attracted to the surrounding permanent magnets, and makes for a more electrically efficient armature.

Elimination of cogging, and greater Power In --> Rotation Out efficiency, are both generally considered positives for small/micro electromechanical systems (IE model train mechanisms), so we hold this as self-evident.

Reducing mechanical inertia or inherrent "flywheel effect" is a topic which has had much discussion since decoder-control of motors (IE DCC control and similar) has become common. There is no commonly accepted "winning ethos", but the arguments for both sides, In short:

- back in Analog, there was little/no active tracking or feedback-control of the motor's behaviour. It was a straight "power In --> motion out" situation, no finesse or sophistication. Ergo using a mechanical flywheel to smooth-out the end-result at the end of the electro-mechanical signal path made sense, and still does for less-sophisticated or non-BEMF decoders.

- However, in the DCC decoder world, where BEMF and similar "feedback control" systems actively change the Input to the motor, in response to the measured motor behaviour Output, anything which (artifically or externally) affects the motor's behavior and messes-with the measurement the feedback system takes and the Control algorithm relies on, is arguably considered a Bad Thing.

Electronic control systems do not take well to mechanical output "edge devices" which run away from them, show excessive or unexpected hysteresis, or otherwise ignore/usurp their authority. Therefore, a motor with minimal uncontrolled inertia, which Starts when the Control System tells it to Start, and Stops when it is told to Stop, is considered "the goal". (Yes, the Digital Control system wants the End Output Analog device to "act more digitally", and in extreme cases can actually get grumpy and snarky when it doesn't!).

In terms of creating the Scale Impression of Inertia or Momentum, such systems prefer to do the "inertia calcs" at the Control "stage" (commonly in the decoder, see CV3 and 4), and deterministically tell the motor how-to-act to emulate the effect of inertia. They do not want the mechanical End-device to "have inertia of it's own, which we do not explicitly and deterministically control"...

SOOO, is mechanical inertia (per a flywheel or heavy motor armature) a good and desirable thing in a model locomotive?
- From an mechanical and old-skool "coarse, one-way, zero-feedback control" perspective, yes
- From a tightly-controlled, feedback-loop-locked digital-control and scale-movement-emulation perspective, no

Happy Modelling,
Aim to Improve,
Prof Klyzlr
Reply 1
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
Prof_Klyzlr wrote:

- From a tightly-controlled, feedback-loop-locked digital-control and scale-movement-emulation perspective, no
but that's not the goal in model railroading -- especially if you want to more realistically simulate the inertial of a loco/train

plus, it helps ride past dirty track

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
Ken Rice Super User 5,939 posts
Ken Rice
greg ciurpita  gregc wrote:
probably
does coreless mean no iron armature?
if so, wouldn't an iron armature provide some inertia that would be a good thing on a model locomotive?

Yup, no armature.  More inertia is a good thing.  But what we’re really after is smoothness, and the inertia is the easiest means to that end.  Coreless motors, since they have no core, also have no constraints on how shaded the poles can be, which means a good one can run more smoothly than a good cored motor.

My blogspot blog: http://rices-rails.blogspot.com/
My MRH blog index
Reply 1
Ken Rice Super User 5,939 posts
Ken Rice

Prof_Klyzlr wrote:
"Brushless" in contrast does not have a hardware commutator and brushes, but uses a series of electromagnets around the outside of the permag core to invoke the motion, and a series of hall-effect sensors to work out how hard/fast/which-order to switch power into the electromagnets to spin the magnetic core. The means the motor itself carries circuitry to:
- receive the multiple Hall-Effect Sensor Inputs
- receive the "source power" input, and interpret speed+direction commands from that Input
- and discretely-drive each of the multiple electromagnets as required

There is a lot of variety to brushless motors.  All of them have the coils stationary and the magnets spinning the shaft.  The traditional form has the magnets on the shaft with the coils around the outside - those are called “inrunners” at least in the RC field.  The other type has the coils stationary in the middle with a bell like shell on the shaft with magnets around the rim, some the magnets spin outside the coils.  Those are called “outrunners” in the RC field.
It’s also possible to leave out the Hall effect sensors.  The motor controller can figure out what’s going on from back-emf.  The common powerful outrunners in the RC field do without Hall effect sensors.  There is sometimes a little hiccup at startup as the controller figures out how to get itself in sync with the motor but they work quite well and put out oodles of power.

Brushless doesn’t mean coreless - the brushless outrunners I have for my planes all have iron cores the windings are on.

My blogspot blog: http://rices-rails.blogspot.com/
My MRH blog index
Reply 1
Prof_Klyzlr Moderator 7,960 posts
Prof_Klyzlr
Quote:
Prof_Klyzlr wrote: - From a tightly-controlled, feedback-loop-locked digital-control and scale-movement-emulation perspective, no
greg ciurpita  gregc wrote:
but that's not the goal in model railroading -- especially if you want to more realistically simulate the inertial of a loco/train

Greg, pls re-read the whole post in context.

If the Control System (Decoder CV3 and 4, for example) is creating the model-railroading-desirable "scale simulation of inertia and momentum", then uncontrolled mechanical hysteresis at the motor-end is arguably counter-productive. A fantastically  "extreme edge case" example of this is Bruce Kingsley's "Ultimate Throttle", which uses a Win10 PC to do a monster "loco + train inertia simulation", and then sends very tight and explicit commands to a Loksound-equipped loco which has had all of it's "autonomous features and functions" inc in-decoder CV3/4 momentum disabled, and literally only does Exactly what it is told to do, when and where the Control System simulation tells it to.

Note the comparison at around 4:00, Bruce has configured the loco's Loksound decoder to run in a manner most modellers would consider unrealistically "slot-car" while under "direct DCC throttle control", but that's because, as he states, he wants to eliminate hysteresis out of the Command Station --> Decoder --> Motor path, and let the upstream Control Simulation dictate what "inertia and momentum" looks like. The result, once the loco/decoder/motor plays it's part exactly as it is told to, no embellishments, run-aways, or ad-lib solos, is actually pretty amazing, inc "loco shutdown, brakes-bleed-off runaway" simulation capability!



Quote:
plus, it helps ride past dirty track

Agreed, but this 12"/1'-scale physics concern is separate and different from the desire for a 1/87th Scale-model-inertia-simulation... ;-)

Please NB, you asked "why Coreless?" and "why reduce motor mass/inertia?". These are just some of the example reasons commonly stated and demonstrated. For my own models, I'm more than happy with an NCE Cab04p or ISE PT "Tactile User Interface" Input device, an NCE or JMRI Command Station "Brain" handling the Assign/Consist/Function maths, with a properly-tuned TSU or Lok decoder "edge device" doing the "scale simulation of inertia and momentum" maths and subsequent motor-output drive to a stock Ath Mashima can motor... ;-)


Ken Rice wrote:
Coreless motors, since they have no core, also have no constraints on how shaded the poles can be, which means a good one can run more smoothly than a good cored motor.
Sounds like another description of the phenomena of "cogging" ;-)

Happy Modelling,
Aim to Improve,
Prof Klyzlr
Reply 1
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
Prof_Klyzlr wrote:
Greg, pls re-read the whole post in context.
not sure what context you mean

i understand that if you are using a motor that requires very precise control of position, with accurate speed changes, you would want a low inertia motor.   you would also like very precise feedback, not speed but shaft angle to do so.

having spent some time studying engine simulation, of course you don't want any momentum in a decoder.  i believe DCC decoder momentum is far more significant than the inertia of a cored motor.    of course  you  want the the controller to dictate the speed of the locomotive.     ideally in such a simulation, the controller would have sufficiently accurate of motor speed.  BEMF would probably be good enough.   having a decoder that uses BEMF to regulate the speed of the motor it is given may adequate.

i don't believe such a simulation would require a low inertial motor.   i don't believe the inertia of the motor would be significant considering the inertia of a train, much less the inertia of the  locomotive by

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
Ken Rice Super User 5,939 posts
Ken Rice
Prof_Klyzlr wrote:

Please NB, you asked "why Coreless?" and "why reduce motor mass/inertia?". These are just some of the example reasons commonly stated and demonstrated.


He doesn’t mention flywheels or motor mass (same idea) in that video at all as far as I can tell, and he say it’s a coreless motor either.  In fact when he runs it with a straight throttle it reacts much like you’d expect the average diesel with a regular cored motor, flywheels, and no momentum in the decoder to do.  I think reducing motor mass is a red herring here.  The coreless benefit for model railroaders is smoothness due to better shading which, as you note, is the cure for cogging.  Eliminating motor mass is a detriment, not a benefit, it’s just a detriment which is more than outweighed by the better shaded poles.  Flywheels (within reason, and pretty much any stock loco doesn’t have excessive flywheels) also help with smoothness, and I can’t imagine the  adding enough inertia to get in the way of any realistic loco simulator.

My blogspot blog: http://rices-rails.blogspot.com/
My MRH blog index
Reply 0
Greg Amer gregamer Regular poster 496 posts
Greg Amer gregamer
Prof_Klyzlr wrote:

….Bruce Kingsley's "Ultimate Throttle" ….





WOW! I really hope that is the future of model railroading. 
Greg Amer The Industrial Lead

MRH Blog Index
Reply 0
greg ciurpita gregc Super User 1,153 posts
greg ciurpita gregc
Ken Rice wrote:
 The coreless benefit for model railroaders is smoothness due to better shading which, as you note, is the cure for cogging. 
that make sense

greg - LaVale, MD     --   MRH Blogs --  Rocky Hill Website  -- Google Site
Reply 0
MannsCreekRR Super User 158 posts
MannsCreekRR
I have been using Faulhaber motors in my On3 shays.  The only decoder that seems to work with them are ESU decoders.  I tried putting in a TCSwow decoder in one and it was almost impossible to operate it.  I contacted TCS and they said at the time they did not know anyone else having an issue, but then again their decoder was never tested with a coreless motor and they wanted me to send them my locomotive for testing, which I was not going to do.  I have converted all my coreless motors to use ESU decoders.  My 3D printed shay has a 5 pole can motor so I could use a TCS decoder.

Jeff Kraker

Read My Blog
Reply 0
Reply