Coreless Motors - Back EMF - Feedback Controllers - DCC

[RBAGE]

I wonder if anyone might be prepared to share their experience and at the same time help me decide how to proceed with motor selection.

 

I build loco kits only to fill the gaps that RTR has not yet ventured into. My latest is Dave Bradwell J39 chassis that I intend to fit into a detailed Bachmann body. 

My strategy is to build the kit as best as my limitations will allow, run it in DC until I am satisfied it's a goodun, then convert to DCC.

I have almost always used Mashima motors for my kits and they tirelessly run my kit built fleet.

 

Some years ago, I bought a number of Portescap motors but hearing tales of woe with their operation on DCC, I have avoided fitting them and these expensive pieces of engineering have languished in the miscellaneous box.

 

Biting the bullet, I have fitted a Portescap to what I was satisfied is a free running chassis. It sounds like a pig on my DC test track. I cannot take this loco forward for DCC sound operation in this condition.

 

I have a number of problems that I am hoping might be resolved:

 1. Is there a problem with running coreless motors on feedback controllers?

 2. What are these problems and do they extend to the squealing pig that has now gone back in the box?

 3. If I have problems related to feedback controllers, are they likely to carry over to DCC? What I'm saying, can I ignore the squealing pig and go ahead and safely fit a sound decoder?

 4. Has anyone experienced problems with coreless motors on DCC operation?

 5. Should I put my Portescaps back into miscellaneous?

 

Any advice would be most gratefully accepted. However, my Portescaps have spent their lives in their boxes because of unconfirmed suggestions that that DCC would destroy these masterpieces so I would like to avoid thirty year old "he said, she said" type of comment. Direct experience would be most valuable. 

 

Thanks in anticipation. 

 

Bob 

[sparaxis]

Coreless generally don't do well on feedback controllers, but do very well on modern DCC decoders.

The pulse frequency on traditional feedback controllers is low enough for the coreless motors to respond to each pulse due to their very low inertia. Coreless motors are generally more efficient than traditional motors, and the high resulting back EMF also confuses the controller.

 

Modern DCC decoders operate at very high frequencies of 18-50 kHz, high enough to get over the low inertia and inductance of the coreless motors. Also the feedback sensitivity can be adjusted to compensate for the higher efficiency. ESU lok pilots have an automatic calibration function that works very well. For example, the LokPilot 4 in my DJM class 71 works very well.

 

Check your loco on smooth filtered DC first. I have been known to use a 9V battery... <grin>

[RBAGE]

8 minutes ago, sparaxis said:

Coreless generally don't do well on feedback controllers, but do very well on modern DCC decoders.

The pulse frequency on traditional feedback controllers is low enough for the coreless motors to respond to each pulse due to their very low inertia. Coreless motors are generally more efficient than traditional motors, and the high resulting back EMF also confuses the controller.

 

Modern DCC decoders operate at very high frequencies of 18-50 kHz, high enough to get over the low inertia and inductance of the coreless motors. Also the feedback sensitivity can be adjusted to compensate for the higher efficiency. ESU lok pilots have an automatic calibration function that works very well. For example, the LokPilot 4 in my DJM class 71 works very well.

 

Check your loco on smooth filtered DC first. I have been known to use a 9V battery... <grin>

Thanks Sparaxis,

I don't have any alternative to my feedback controller for test and I had considered a 9V battery. I'll give it a go and see if it improves the J39.

Are all feedback controllers similarly affected. Mind you, it is only for test and set up purposes so I'm reluctant to buy anything expensive.

[RBAGE]

I have a couple of Gaugemaster controllers, one is used to power a cheap mini drill, but both seem to have the same effect on the Portescap. 

Is it an issue of the motor or the controller. I have a Faulhauber in an old Millholme kit. Maybe I should give that a go and see what happens. Are all coreless motors likely to be affected?

Are coreless motors becoming more popular/common/recommended?

If I have to buy another make of controller for my set up track, is there something that can be recommended to deal with coreless motors? I friend of mine has had problems and I think he has a KPC controller.

Does anyone have experience of coreless motors on other systems? ECM Compspeed? Modelex feedback controllers.

 

I've tried to keep things simple in the past but this problem is quite taxing but I don't want to throw the towel in. There must be a solution.

Or should I just jump straight into DCC with these motors and hope that technology has resolved the problem.

 

[DLT]

I use Portescap motors on my locos, and can confirm that they run very well on normal dc, but not with feedback, as described by Sparaxis above. 

I ventured into dcc some years ago, and had no problems at all with the running and control of the motors. 

That I didn't stick with dcc and reverted to dc is a separate issue, and not really relevant here.

I use "normal" Gaugemaster controllers, with the cream coloured facing.  The black ones are feedback, and I avoid them.

Not sure if this is much help!

Cheers, Dave.

[Paul Cram]

All my early locos are portescap fitted and 2 have been converted to DCC without issue. The later Portescap have a whine from the gearbox due to wear on the mold of the gears.

[RBAGE]

2 minutes ago, Paul Cram said:

All my early locos are portescap fitted and 2 have been converted to DCC without issue. The later Portescap have a whine from the gearbox due to wear on the mold of the gears.

Thanks. Maybe my squealing pig is actually a whining pig.

[DLT]

You definitely need to source a non-feedback controller to test run Portescaps, and like you, I would get them running well on dc before converting.

Another issue with Portescaps that have been stored for some time is that the lubricant in the gearbox hardens up and glues the gears solid!  I wouldn't attempt to power up a Portescap in  this condition, that is when you are most likely to wreck it.

I removed the motor from the gearbox (two long screws through the gearbox sides and the nylon block) and then washed the gearbox out with white spirit.  Only when it was running freely and re-lubricated did I put it back together and power it up.

Cheers, Dave.

[melmerby]

IMHO bin the Portescaps. Flog em and get a decent 5 pole skew wound motor and suitable quality gearbox with the proceeds.

Portescaps were of their time and not needed in today's world.

 

Coreless motors are however becoming more common in model railway circles because they are cheap, light and compact. Not necessary the ideal specs for a loco.

 

As has been said before coreless motors work fine on modern DCC decoders. No need to fiddle with the BEMF settings.

[DCB]

I would suggest you build a test rig for your DC running if you build locos with coreless motors.
I use a Morley vector which provides variable voltage, No PWM , no feedback. Off load it will show less than a volt and you can run an LED off it towards minimum speed.

These are expensive for just using as a running in rig but there are variable voltage supplies available which will give 3-4.5-6-9-12-15 volts etcs quite cheaply. add a coupe of polyswitches 0,25 amp as overload protection, with a selector switch for one or two in parallel (or more) and a DPDT switch for reversing and you should have a decent power supply for running in.   Don't leave out the overload protection, Coreless motors don't  like being stalled.

I use the uncontrolled DC output from a  very old Hornby Dublo transformer with a very sensitive overload breaker, a selector switch and 11 diodes for my test rig, but its for XO4 motors and ring fields not coreless, I would need to improve the overload protection for coreless, but it works very well, much better than the standard H/D or a Duette.

Edited April 13, 2020 by DavidCBroad

[SR DUNCAN]

Hi have a look at this Link

on High Level  Coreless motors.

 

Duncan

[RBAGE]

You are all coming up with valuable information, so thanks. I just need to pull it all together in a does and don't and what will work with what.

For example, are there feedback controllers that are OK with coreless motor? If coreless motors are becoming more commonly used, we all need to be careful with what we are doing.

If we could come up with a list of controllers that are coreless friendly and those which should be avoided, that would be useful for a great many modellers. Does anyone have experience in both direction so we can pull together such a list, on this forum, so that anyone interested could access the information. Listing them here would be a great start.

I recall that Roy Jackson's Retford used Modelex controllers. Locos on that layout were powered by all manner of motors from Faulhaubers down to ancient open-frame designs such as Pittman's, with no obvious ill effects. On the other hand, KPC controllers don't seem to like coreless motors at all. So, there's obviously some variation in construction, performance or whatever might influence compatibility.

If anyone can name names, based on experience, maybe we can summarise the whole lot and help modellers, present and future. Such information would be very useful. Well, to me anyway.

Thanks again.

[melmerby]

1 hour ago, RBAGE said:

You are all coming up with valuable information, so thanks. I just need to pull it all together in a does and don't and what will work with what.

 

Well if you do come up with the definitive list it will probably be enough for a book.

 

Coreless motors were originally developed for electronic kit that required fast response, which coreless motors with their lack or armature mass were ideal for.

I used to service X-Y plotters and these usually had Canon or Maxxon motors. Expensive but extremely high quality. Built like a Swiss watch!

I wish I had kept the surplus ones I used to have!

 

The current crop of coreless motors are as developed in China for cheap mass production, the lack of core makes them cheaper than a similar rated iron cored motor and more compact, ideal for cheap mass produced lightweight items, the side effect they are now starting to be used in model railways where the lack of rotary mass and light weight is not an advantage.

 

Oxford replaced the original 5 pole skew wound iron core motor in the Dean Goods with a 3 pole coreless design which IMHO is definitely inferior.

 

 

Edited April 13, 2020 by melmerby

[mdvle]

2 hours ago, RBAGE said:

For example, are there feedback controllers that are OK with coreless motor? If coreless motors are becoming more commonly used, we all need to be careful with what we are doing.

If we could come up with a list of controllers that are coreless friendly and those which should be avoided, that would be useful for a great many modellers.

 

I suspect the general advice of avoiding feedback controllers when using coreless motors is going to be the best option going forward.

 

Even if someone was able to find a feedback controller that currently worked with coreless motors, there is no guarantee that a production change wouldn't mean an identical controller made the following year would still be coreless safe (this frequently happens in the electronics world, where a change in a part can lead to subtle differences).

 

 

[Chuffer Davies]

Hi,

I think Sparaxis has explained the reason for the apparent problems well.  Melmerby's suggestion is perhaps slightly drastic and not one I would immediately support.   

 

With regards which controllers can be used successfully our experience at the Shipley MRS is as follows:

 

- Non feedback controllers of most types work well.  These days we typically use the Gaugemaster (cream faced) controllers and their GMC-W black hand held unit but not their GMC-HH black hand held controller because this has feedback and Gaugemaster themselves warn against using this with coreless motors. The old H&M controllers work fine such as the Duette but don't use the half wave option.

 

- The two feedback controllers we have used successfully are the Modelex and the Pentroller range.  We would definitely not recommend ECM controllers.

 

- As to DCC it is all down to the chip you select to put in the locomotive.  Some chips are more suitable than others (or at least that has certainly been true in the past) so it is worth doing research before purchase.  Personally I use the small Zimo chips. It is also worth checking the chip's user manual for guidance on adjustments to the CV settings specifically for coreless motors.

 

With regards the whine experienced with Escap units it is my understanding that the noisy units resulted from moving production from the UK to overseas.  As I understand it the material used for the plastic gears was changed and it is this change of material that has caused the increase in noise rather than warn moulds.  I don't know if anyone else has more information about this.

 

I hope this helps.

 

Frank

[melmerby]

2 hours ago, Chuffer Davies said:

Hi,

 

 

With regards the whine experienced with Escap units it is my understanding that the noisy units resulted from moving production from the UK to overseas.  As I understand it the material used for the plastic gears was changed and it is this change of material that has caused the increase in noise rather than warn moulds.  I don't know if anyone else has more information about this.

 

I hope this helps.

 

Frank

Were they ever made in the UK? I thought they were Swiss.

[Chuffer Davies]

30 minutes ago, melmerby said:

Were they ever made in the UK? I thought they were Swiss.

Its a Swiss company but they have manufacturing plants in other countries.  Originally the gear boxes were built in the UK and paired with Swiss made motors.  Or at least the 1616 and 1624  motors were their own but I believe the 1219 motors were sourced from Faulhaber

Frank  

[Michael Edge]

The Portescap gearbox was designed and made in the UK, I say "designed" but it is actually scaled down from a 7mm gearbox made by Leeds Model Co. back in the 1940s or early 1950s - the only difference was the substitution of plastic bevel gears for the original crown wheel and pinion. The coreless motors were Escap from Switzerland or in the case of the 1219 Faulhaber, there was an assumption (backed by an advertising campaign) the the coreless nature of the motors was the big advantage but the big change was the efficiency of the gearbox without any worm drive. The same can be seen today with the colossal power available from the tiny n20 motors on spur drive boxes - out of all proportion to the size of the motor. At the time they appeared there was nothing available with anything like to power and smoothness of a Portescap.  I've used hundreds of them in scales from TT to O, I liked them because they were so easy to fit and the customers paid for them anyway. There must be over 150 of them running on the vast Carlisle layout and while some of them are a bit noisy they all perform well, the only motor/gearbox which can match them is the Comet GB1 with its Maxon motor.

They all run on DCC with no problems, I have a few on my own layout which run with all sorts of DC control systems, including feedback types. None of them have ever suffered any damage, the only slight problem is that the best running ones don't quite stop when the controller is turned right back - you have to switch it off as well. I have seen a very small (single figures) number of faulty Escap motors and no faulty Faulhabers, the fault has always been worn out brush gear.

Efficient gearbox design is the key to performance of models, not the motors, that's why the High Level ones are so good. The best 7mm ones all have a skew gear for the first stage gearing and are similarly efficient regardless of what motor is attached to them.

[WIMorrison]

I wonder if people confuse Coreless with Brushless - which are most definitely not the same thing?

[Izzy]

I think there are two separate aspects here. One, what type of electrical control is best suited to coreless motors. And Two, the noise/efficiency of gearboxes. In regard to the Portescaps these are combined issues.

 

Coreless motors are meant to work on very low current draw, one of their main attributes, and respond/react to very small current changes. Since their arrival in model railway use, and they were used by some well before the Portescaps arrived on the scene, I have never found them to be happy with any type of DC feedback controller. Because of their nature they over-react to the low frequency pulsed power, which is most commonly seen in the way they can ‘jitter’ along track. They also don’t have the iron core mass to absorb the higher levels of current these supply, and so can tend to become too hot, not being able to dissipate it. It is therefore best to use them with simple DC current as supplied by non-feedback types to avoid these issues. 
 

The current DCC decoders use feedback, generally known as BEMF (PWM- pulsed width modulation), but this uses much higher frequency, often with a choice of rates with better decoders as well as cv adjustments specifically for coreless, and they will run well

without issue when properly set up.

 

Gearbox noise, and gear efficiency, is another can of worms entirely. In respect to spur gears they do tend to become noisy when run at high speed because of the transfer of loading from one tooth to the next, which results in harmonics, eg, whining. When the gears are either worn, or not of optimum meshing distance or quality, then this just gets worse. Helically cut gears, where the teeth are at an angle - some of the better gearboxes use this type - will be quieter because of a smoother load transfer. Although they absorb more power, worm gears are quieter because they maintain a constant loading.

 

At present I feel that a High Level gearbox coupled with any type of motor, cored or coreless, is a good combination for either DC or DCC for 4mm. Perhaps the most flexible given the range of combinations possible.

 

Izzy

[melmerby]

30 minutes ago, WIMorrison said:

I wonder if people confuse Coreless with Brushless - which are most definitely not the same thing?

In practice I doubt it, as Brushless motors can't work on DC without a dedicated control circuit.

Terminology wise possibly.

 

Motors that work on DC must have some form of commutation, either by using brushes or by fancy electronics.

The latter has been adopted for many devices in recent years such as battery operated drills or motors in computer equipment but not much for such things as model railways in any mainstream form.

 

Theoretically you could have one in a model train running on DCC as you could have some circuitry to convert the output of the DCC decoder to that required by the motor.

The Märklin Sinus motor is an AC motor with control gear.

Straightforward DC control via the voltage on the track would be very difficult, if not impossible.

[sparaxis]

Interestingly enough after three different versions of the "sinus" concept and a bankruptcy, Marklin dropped the whole idea of brushless motors and went back to 3 and 5 pole iron cored motors, (With the odd coreless here and there). 

 

They worked "well enough" when supplied with the constant 22 V or so of the Marklin digital system , but when run on "analogue"  they were miserable as they needed at least 8-9 V on the track to power the electronics enough to move.

 

The first version was a home made pancake that would run away downhill.

The second version was  "can" with worm drive, that had noticeable jitter at slow speeds. (Off-the-shelf stepper motor I believe)

The third version was their so called "soft drive" sinus with skew magnetisation.

 

A major problem with these is if the drive circuitry dies, your loco is dead as there are no spares available from Marklin.

 

(Quite unlike the "iron age" Marklin locos from 60 years ago that I have been buying on eBay for light entertainment. Generally they run just fine. Maybe need new tyres or pickup shoes...but that's it)

 

Lionel offered/proposed a new generation of locos with brushless motors probably 20 years ago, but I am not sure if they ever reached the market or died a natural death.

[DCB]

Has any one tried several small or one large ceramic capacitor across the coreless motor to smooth out the PWM pulses?

I always suspected the benefit of the Portescap was actually the spur and bevel gear drive.  I know the bevels are a pain to keep in adjustment but I reckon an XO4 would run nearly as nicely with that sort of gearbox.

 

[RBAGE]

On 13/04/2020 at 20:38, Chuffer Davies said:

Hi,

I think Sparaxis has explained the reason for the apparent problems well.  Melmerby's suggestion is perhaps slightly drastic and not one I would immediately support.   

 

With regards which controllers can be used successfully our experience at the Shipley MRS is as follows:

 

- Non feedback controllers of most types work well.  These days we typically use the Gaugemaster (cream faced) controllers and their GMC-W black hand held unit but not their GMC-HH black hand held controller because this has feedback and Gaugemaster themselves warn against using this with coreless motors. The old H&M controllers work fine such as the Duette but don't use the half wave option.

 

- The two feedback controllers we have used successfully are the Modelex and the Pentroller range.  We would definitely not recommend ECM controllers.

 

- As to DCC it is all down to the chip you select to put in the locomotive.  Some chips are more suitable than others (or at least that has certainly been true in the past) so it is worth doing research before purchase.  Personally I use the small Zimo chips. It is also worth checking the chip's user manual for guidance on adjustments to the CV settings specifically for coreless motors.

 

With regards the whine experienced with Escap units it is my understanding that the noisy units resulted from moving production from the UK to overseas.  As I understand it the material used for the plastic gears was changed and it is this change of material that has caused the increase in noise rather than warn moulds.  I don't know if anyone else has more information about this.

 

I hope this helps.

 

Frank

Hello Frank,

What was your experience or knowledge of ECM controllers that would definitely not recommend them for coreless motors?

Thanks for you feedback.

Bob

[Chuffer Davies]

42 minutes ago, RBAGE said:

Hello Frank,

What was your experience or knowledge of ECM controllers that would definitely not recommend them for coreless motors?

Thanks for you feedback.

Bob

Hi Bob,

My comments are based upon what I have read in the model press particularly when the Portescap units were first launched in the 1980's.  There was quite a lot of discussion about the compatibility of these motors with various controllers.  There was talk about a modification for ECM controllers to make them compatible with Portescaps.  I can't remember whether it was a specific ECM unit or a factory modification that you needed to request.

 

My personal experience early on with standard ECM controllers and Portescap units was that not only did the motor sound unhappy, slow running was significantly worse than with smooth DC control.  I believe this is because the commutator has negligible effective inertia and is constantly adjusting to the pulse from the ECM feedback controller resulting in the proverbial Kangaroo effect.  

 

I know that others have suggested that there are no issues but I have to believe that the people who wrote the articles in the 80's knew what they were talking about and I have always avoided running my Portescap powered locomotives (which most of mine are) on layouts with either ECM or Gaugemaster GMC-HH controllers.

 

Regards,

Frank