Another dead motor, and fixed

[ntpntpntp]

@Porfuera   There are many PWM based controllers out there, it is a common method for driving a DC motor and helps with low speed control especially.  Also every DCC decoder uses PWM to drive the motor. The difference is the PWM frequency used in modern decoders is usually much higher than that of PWM controllers - especially the older designs. Decoders often have other "tuning" settings to get the best out of whatever motor type they are driving.

On top of PWM control you can also have "feedback" where the controller senses the motor speed by the back-EMF generated in the gaps between the pulses  (look up "back-EMF" 🙂 )  Modern coreless motors don't generate so much back-EMF and this can cause the controller to overcompensate.

I have PWM controllers with switchable feedback on my N gauge exhibition layout. Back in the 90s with mostly older motors I'd use the feeback quite a lot, but nowadays it's mostly switched off other than for locos made in the 1970s/80s.  Some recent locos absolutely hate it and become very jerky!

Even worse than PWM for coreless motors is the "Half-Wave" setting on ancient controllers such as the H&M Duette and Clipper still which are still loved by older modellers. It results in very low frequency pulse power of a variable voltage.  I always recommend if you MUST use one of these then keep it on Full-Wave and High Resistance settings, but you're better off changing it out for a pure DC controller.     

Edited July 16 by ntpntpntp

[Fishmanoz]

Matt, I agree with your suggestion for a direct replacement.  It should include the gear wheel as that will significantly increase the number of people who can do the swap.  Offering it free to A1/3/4 and 08 owners would also reduce warranty replacement costs.

[ateshci]

PWM is not detrimental to coreless motrs provided the switching frequency is high enough to compensate for their low inductivity. Maxon's have an in-depth treaty of that on their website. Sometimes two SMD 100µH chokes will straighten out any jerkiness. It is not true, however, that modern coreless motors generate lower BEMF compared to iron-core ones. Due to their high efficiency, they need less current to run, which is only possible when the BEMF is higher. Another advantage of them is the absence of commutation spikes and a generally lower value of energy stored in the magnetic field.

[Simon M]

This is interesting and well worth getting further examples of failures. 
for me (touch wood) I have had no issues yet on my A1 and A3, both over 1 year now, or my A4 and class 08 running for 10 months ish. 
when you consider 3x 20 mins per week would give you ~50 hours per year, and Matt stating failures were ~30 ish hours. 

It would be useful I think to understand conditions, I.e DC/DCC/HMDCC, load such as number of coaches? , all pulling or pushing(direction), speed setting, inclines involved? 
 

all of mine are HMDCC including 08, speed very rarely over 50% on HM App. A1 pulling 4 pullmans, A3 and A4 each normally 3 x Mk1’s, class 08 shunting coaches and pulling 10x 7 plank wagons.

hopefully my 50’s and 43’s won’t be affected. 

[Purple]

My 08 shunter is with Hornby for a motor replacement but hasn't really done many hours at all. I'm hoping that the next one lasts a bit longer 

[Rallymatt]

Good points Simon; 

no inclines and locos only used pulling not pushing loads. Maximum train loads , 6x Mk1 coaches although fitted with illumination so slightly heavier, or 12 BR vent vans. V High CV is set to 195 rather than 255 and in operation these locos see speed step 80/128 maximum on DCC. Track voltage is 14.8v, surge protector on PSU and snubbers on a truncated power bus. Duty cycle is measured as actual running time per loco. There are many examples of these motors failing and it’s guaranteed they will. The early signs are that slow speed running will get lumpy and possibly ‘stick’ on reverse, that’s because the finger contacts burn out one by one generally and they are set up as trailing contacts in normal forward rotation. The positive terminal fingers will always burn out first. The running characteristics deteriorate gradually. My repaired motor runs so much better it’s obvious the old one was worn for quite a while. Trigo will be next to turn up its toes, it’s on 32 hours and slow speed running is not as smooth as it once was. I’ll let it get to stage it stops running. 

 

[Simon M]

6 minutes ago, Rallymatt said:

The early signs are that slow speed running will get lumpy and possibly ‘stick’ on reverse,

Thanks Matt - will keep a close eye on this now and thanks for the work around repair if (when) I need it!. Bit concerned now as was gonna move my Flying Scotsman and WW on soon as I focus on my diesel collection, but wouldn’t want to sell on knowing a failure could be imminent. Hmmmm. 

[Rallymatt]

I think it’s just a fact of reality that sealed can motors are a wearing part. At least the fix is under £2 😁👍

[ateshci]

If you want to read about the pro and cons of brush types, take a look here.
There are N20 carbon-brushed motors available, alas because they see use in robotics, as geared motors only. Pololu sells them as HPCB motors.
The open frame gearbox is incredibly noisy and from personal experience I would not recommend to use it as is. What I have done in that case is to separate the motor from the gear assembly and use it as a stand-alone, but that requires some tinkering. These carbon-brush motors are really heavy-duty and last for years.

[dBerriff]

An interesting link. The precious metal brushes seem to suit being run at a steady speed in the forward direction (which I presume means commutator rotation). That should be fine for the mainline expresses but not so good for shunting. 

[Rallymatt]

Thanks for that link @ateshci 👍 The stand out disadvantage for EB motors stop/start/reverse! All the things that people do with model railways. 
I have seen the motors with gearboxes  attached, although the specifications didn’t identify the brush type on ones I have seen. Do you have a links to the carbon brush variety and if the gearbox is removed, it’s only screws and the worm gear, which have the similar running characteristics as the OE motor. I’m happy to try some other options. 

[ateshci]

@Rallymatt
Here's Pololu's 5:1 variety

 

[ntpntpntp]

13 hours ago, ateshci said:

It is not true, however, that modern coreless motors generate lower BEMF compared to iron-core ones.

Interesting, I'd always believed what I'd read elsewhere about it being less.  Either way, the negative feedback loop of older controllers over-compensates and causes poor / jerky running.  Seen it happen with my feedback controllers which is why the feedback circuit is turned off most of the time. 

[ateshci]

Which is understandable because the controller frequency is much too low for coreless motors. An iron-core motor can be driven best at somewhere between 50 and 500 Hz, for smooth operation over the whole rpm range the frequency should vary with duty cycle from low to high, too.

Addendum:
Should you really want to dive deeply into that topic, read this

 

Edited July 17 by ateshci

[Rallymatt]

An update on Motors; 

Silver King was returned from Hornby with a new motor fitted… few faults, lubricator rod not refitted, grease all over the wheels and the insulation not re instated on the motor connections. 
I appreciate a motor was fitted out of warranty FOC but the bad news is that the motor is defective. It grinds and stutters in reverse and is hardly capable of running an A4 to a scale speed of 70mph and is very noisy at that. 
I have spent 2 days testing and altering settings on the decoder to see if that could coax some decent running and it’s definitely a dud. It runs but not as it should. Rather than return again I am going to fit something different as fitting the same thing to fail is a pointless exercise really. I have a couple of options here and have been speaking with another source too. 
Hornby really need to get a grip of this motor issue and quickly because on the failure rates that are coming through, this is what will damage the future of the scale. I am disappointed for Hornby as it really is ‘spoiling the boat for a ha’Peth of tar’ 

[Fishmanoz]

This needs high priority reporting back to Hornby Towers that the current motor has the potential to sink the scale if an alternative is not found very soon - motor and gear direct replacement needed.

[Rallymatt]

Fishy, I’m pretty sure they know but so far it’s either a huge issue finding a suitable alternative or admitting there is an issue is too painful. 
For modellers like you, how would this be dealt with halfway round the world? 

[Silver Fox 17]

It certainly is a worry about how people view and respond to the poor motor design.

A few people have given up saying they had had too many faults although not all specifically stating it was motor issues.

I have bought a couple of these cheapish and not had a problem (so far) and wonder if many are linked to how they are handled and run.

We know the valve gear is fairly delicate and some seem to think speed is stop or maximum which will not help.

I run dc analogue only but the majority, note I said majority, of issues are related to dcc locos.

I have been to a couple of events where the Hornby layout was on display and, with spending most of the day at the event, I have to say each time the layout by about mid afternoon has nothing running yet the same locos on just staionary. I have often wondered if they had issues but staff did not want to show anything was amiss while the public were around. I am not the only one who has noted this too but it makes you wonder why a demo model railway with the staff there was not always running????

It would be nice for Hornby to get a better more reliable motor and offer it to those with these older ones but with over 12000 sold it is unlikely.

Edited July 24 by Silver Fox 17

[RB51]

3 hours ago, Rallymatt said:

I’m pretty sure they know

They do now, it's been passed to Hornby Towers. R-

[Fishmanoz]

The problem here is there is no in-country Hornby support for TT:120.  And I can’t even go via a supplier here as my 3 locos with these motors were all purchased online from Hornby UK, one delivered here from a purchase delivered to me here prior to the SMS ban kicking in, and the other 2 via a UK address then on-forwarded.

So warranty support is on the other side of the world. Consequently, my clear preference is for Hornby to own up to the problem and have a drop-in replacement delivered here for me to install.  That would save expensive return and re-delivery.

And as for 12,000 sold, much cheaper to supply 12,000 new motor/gear assemblies than to refund the lot and start again. 

Edited July 24 by Fishmanoz

[Rallymatt]

Well as was out on bike riding along I have decided that I am pulling the ‘new motor’ out and first going to fit the £1.50 EBay Special and see how it fares, TBH it can’t be much worse the brushes I swapped over run really well so far. I’m going to open up the ‘new motor and take a look, I have a feeling the ‘precious metal’ fingers are going to look a bit not precious

Thanks @RB51, I really think this needs to be a top priority for Hornby because it won’t be long before 1000’s of locos will be hitting the mark at which the motors seem to go bad. Its not a blame game for me, it’s about finding solutions and improvements 😁

[ntpntpntp]

Being a bit of a traditionalist used to motors you can service and replace brushes etc, this is more the type of motor I wish were in the TT:120 locos.  5 pole skew wound.  I have several N gauge locos with this general style of motor (Arnold, Brawa, and Kato models in particular) though with both shafts the same length.  Yes it's a bit larger than an N20 (shown for comparison) but if it can fit in N gauge surely it can fit in TT:120?   These ones I bought on ebay a few years ago for probably £2 each ish. 

[Purple]

What's the average time for Hornby to return a failure with a new motor? When I rang to ask about mine I was told within 28 days 

[Skelton Junction]

I’ve recently experienced between 10 days and two weeks for the two 50s (The second one is due back tomorrow.)

[Silver Fox 17]

6 hours ago, ntpntpntp said:

Yes it's a bit larger than an N20 (shown for comparison) but if it can fit in N gauge surely it can fit in TT:120?   These ones I bought on ebay a few years ago for probably £2 each ish. 

Size wise, the motor alone will probably fit in a 120 body but the issue would be due to the height of the gears, and, needing the motor centre line in the same place as the original then it will most likely be too high up for the body to sit correctly.

If you are looking at altering the gear train then it is a good chance it will fit.