Jerky running on down slopes

[HOTTODDY]

Forgive me if this has been discussed before, but are there any component gismos that can be added to a track circuit or controller to make downhill running smoother? I have mostly H & M or Gaugemaster controllers, but I also have a feedback type of unknown origin. GM advise against using feedback control  with some types of motor. My locos have a variety of motor types, old and new, so I am reluctant  using anything but straight control. Its just annoying with locos stop/starting causing wagons (usually) derailing on down slopes. So any ideas would be welcome please, but as I'm not an electronic savvy person, please keep it simplish as you can! 

[Crosland]

What sort of gradient?

 

Stop/start sounds like poor pickup to me. Does the same happen if you turn the loco round and run it backwards downhill?

 

What about when running light engine?

[locoholic]

The only time I've observed this was with an old Farish 00 gauge 94xx pannier. The weight of the metal body was such that when going downhill the armature of the can motor was pulled forwards by the worm gear, so that the brushes lost contact and the motor stopped, and the loco kangarooed down the gradient. Never did manage to sort it. The body got mazak rot, so said loco was disposed of. Not sure if there would have been any electronic way to counteract it?

[brianusa]

Another similar reason is that if the gradient is steep enough, the train speed will exceed the engine speed causing a momentary stall.  This usually shows up in the same position on the grade.

 

Brian.

[Dublodad]

Do you mean stop/starting, I find that some locos surge on 1;50 down when feed with a Gaugemaster feedback controller. I believe this is caused by the feedback "over controlling£ the speed. I had this effect with my previous petrol Focus when in cruise control descending hills. Basically the cruise control was over compensating when slowing the car. This effect does not occur on my present diesel Focus.

 

Terry

[Nearholmer]

Dublodad - try a Volkswagen. Ours will maintain speed within a gnats, up hill and down dale on cruise control. It's as if a ghost is driving!

 

Hottoddy - hard to be sure without a stack more detail, but if the locos are geared such that it is possible to turn the motor by turning the wheels (spur gears, bevels, and some double-start worms are efficient enough to permit this), the load of the train could push the loco into "over speed", thereby causing the back-EMF (voltage generated by the motor in opposition to the supply voltage as it turns) to rise. It is the back-EMF that a "feedback" controller senses, and if it senses a rise, it will compensate, maybe setting-up a cycle of trouble. But, as a poster above suggested, it could equally be a mechanical effect, caused by excess end-float in the motor armature (common in old open-frame motors), or slop in the drive somewhere.

 

K

[DCB]

I have this problem with Hornby and Triang locos with the X04 motor down a 1 in 36. Locos like Hornby Halls puling 7 or 8 coaches or 20 heavy Hornby Dublo wagons.

 

The cause is the non reversible worm drive, the moment the loco begins to run faster than the motor is driving it the worm gears resistance increases slowing the motor further at some stage the loco slows so much that the motor speed catches up with the loco speed and the resistance drops away and it accelerates only for the whole process to repeat.    

 

Locos with a lot of side play on the driven axle and lots of  armature end float  are particularly bad, those with the worm wheel shimmed to reduce side play and the armature shaft end float shimmed suffer less.

 

The problem is worse with smooth DC than with pulsed power, my OnTrack and H and M Safety Minor power units are particularly bad.

 

I suspect DCC equipped locos with worm drive must be particularly bad.

 

The cure is surprisingly simple, on the H and M  I just use half wave and a high setting on the controller, the jerking is so rapid it is virtually imperceptible.   With the OnTrack I simply don't use it on the downhill stretch any more.

[bertiedog]

The actual gradient is not mentioned and might be the basic cause, as many layouts have far too steep a gradient, or are simply built without testing or measuring the slope.

 

Anything above about 1 in 50 should be used with caution, and 1 in 20 is about  the most any loco can take with a load. This assumes plain wheels with no traction tyres. Running badly down the slope is almost always mechanical play in the mechanism.

 

The other problem not addressed very often, is the transition between the flat and the slope, too sharp, and wheels leave contact in the dip at the bottom or the hump at the top. The slope transition should be a couple of feet at least, with the track shimmed to give a smooth transition from level to the slope angle. Flexible track is better at transitions than set track where joints should be kept from the transition area.

 

Most diesel models are better at slopes than steam, as the drive is simpler. Modern free running coaches can add problems to down slope running as they can push the loco to move faster, the older coaches always dragged, acting as a brake.

 

So check the track and the locos, and service the mechanisms to cure it.

 

Stephen.

[Gwiwer]

I have always believed this issue - of jerky running down steep hills - to be caused by back electro-motive force (BEMF).  

 

One of my branch lines is very steep as its construction was something of an afterthought.  Not everything goes up it nicely (or at all in the case of lightweight steam locos) and some trains exhibit jerking on the downhill run when operated at a suitably slow speed.  These include the Bachmann class 150, Hornby BoB / WC, Hornby class 153, Dapol O2 but notably not the Hornby bubblecars nor the Lima generic suburban DMUs.

 

If I apply more power the trains run smoothly but too fast for the situation.  The explanation I was given is that there is low current potential across the motor which is being affected by gravity on the slope with the train trying to run faster than the current permits.  This sets up a BEMF across the motor briefly causing a jerk until things are back in harmony but then as soon as that occurs gravity takes over one again and tries to accelerate the train thus perpetuating the cycle.

 

The same trains do not exhibit this problem on level or gently graded track.  It might or might not be relevant that the ones which perform well have cheap pancake motors which are also light in weight while most of the others use heavyweight units of one sort or another.  I cannot find any undue slack in the linkages anywhere.

[BR60103]

On downhills, I feel it's usually due to play in the mechanism, and the loco running away faster than the motor.

Check is the armature moves back and forth inside the motor. I'm not qualified to tell you what to do to correct this, but it often means adjusting bearings, or adding thrust washers.

[Nearholmer]

Gwiwer

 

See my post above wherein I mention back-EMF.

 

I hope you won't be offended if I say that the explanation of the effect that you give is very close, but not quite correct.

 

Back-EMF is produced all the time that the motor is turning, and the faster in turns, the greater the back-EMF. The motor is simultaneously acting as a generator, all the time it is turning, and it is the difference between the supply voltage and the back-EMF that the motor generates which sets the current flow around the circuit.

 

If gravity drives the loco downhill, turning the motor faster than the speed set by the controller output, the back-EMF from the motor rises above the supply voltage. A feedback controller detects this and thinks "ah, train going faster than required, I'll turn myself down", and if things go well speed is neatly regulated, but if there could be circumstances where there is over-compensation, and surging can result. In the same circumstance, a non-feedback controller will do nothing, simply carry on feeding-out set voltage, but the current in the circuit will become zero, as the back-EMF exceeds the supply voltage, the motor will then cease to be driven electrically, and the loco may well stall due to friction in the gears etc, then will re-start as back-EMF drops below supply voltage and current flows again (and that cycle could repeat).

 

A lot will depend upon how efficient the gearing is, and I suspect that many locos simply won't be affected, even on a very steep grade, with a heavy load pushing them, because their gearing is so inefficient that the motor always has to work to overcome friction in the gearing, and can't be driven into "over speed". My gut feel us that "stop-start" will be most likely with some old Hornby Dublo locos (efficient gearing) and a non-feedback controller with a nice smooth output. Putting "bigger lumps" in the supply, by switching to half-wave rectification, or switching any smoothing circuit out, should, I think reduce the problem.

 

But, after that long ramble about electrical things, I still think it is always worth looking carefully for mechanical affects too.

 

Kevin

 

(Who really wishes he hadn't woken up in the middle of the night as a result of a draft from a window left wide open, and then found himself unable to go back to sleep!)

[Gwiwer]

 

 

I hope you won't be offended

 

Offended?  Not at all.  I come here to learn and would be the first to say I no next to nothing about electrickery.  

[34theletterbetweenB&D]

...Its just annoying with locos stop/starting causing wagons (usually) derailing on down slopes...

 Is it annoying enough that you are prepared to do some experiments to determine root cause(s)?

 

In addition to what Crosland very sensibly asked in the second post to begin characterising the problem- all of which need to be tried and answered:

 

What do the trains that derail do if released onto the down gradient without loco, at the normal speed they start the descent?

Is there any difference in the stop/start performance on a train between feedback control and no feedback control by each loco type you operate?

 

 

 

Or the short cut. In the search for a 'gismo' I suspect the answer will not be anything electrical/electronic. Either ease the gradient, 1 in 80 or shallower is usually trouble free with RTR OO product (there's a good reason for this); or if that cannot be done add drags on the axles of all the trains that have this trouble, sufficient that the locos actually have to pull the train downhill.

[cliff park]

Just a thought that no one seems to have mentioned:- quatering. Is it only steam outline?

[DCB]

On the full size railways they used to pin down brakes on the wagons and pull trains down hill against the brakes.

 

I wonder if a brake van with working brakes would be a partial solution.

 

It is something I have been wondering about as rakes of Bachmann coaches have a habit of rolling out of carriage sidings...   It is not too easy to make happen in DC but should be a cinch with DCC or radio control...

[brianusa]

Happens with all scales and most of these replies come up with similar results; that the train pushes the loco.  Comments about the steepness are also relevant - mine are very steep and it should only affect worm drives.  Side winders will just go along with the extra push  and as mentioned, speeding up helps the problem but by then you're going too fast.  One of those things to live with!

 

Brian.

[kevinlms]

I've certainly experienced surging before on the downhill running, but certainly never to the extent of causing wagons to derail. As many have suggested the causes to be varied.

 

As you stated that its mainly goods trains, I'd try to modify a brake van, so that the wheels drag on it. A bit of brass strip mounted & bent so it drags on the axle, may be enough. This will cause the train to stretch as the train is being dragged downhill. A bit of experimentation is required.

[34theletterbetweenB&D]

On the full size railways they used to pin down brakes on the wagons and pull trains down hill against the brakes.

 

I wonder if a brake van with working brakes would be a partial solution.

 It should work. Powerful permanent magnets are now cheap. Secure an iron strip on the sleepers from before the downgrade, and have a permanent magnet on a beam pull on the brakes of the van perhaps?

[sp1]

It should work. Powerful permanent magnets are now cheap. Secure an iron strip on the sleepers from before the downgrade, and have a permanent magnet on a beam pull on the brakes of the van perhaps?

Completely off topic, but I wonder if this idea would work for a slip coach?