Limiting speed on a curve?

[Realistic_build_Speed]

I tried to search for this but didn’t manage to find anything..

 

I am “responsible for” electrical and safety matters on two layouts, one for my Dad and one for my son. In both cases we’ve got tight curves and fast trains. I’m wondering if there’s an easy way of limiting the speed on the tight curves to prevent the sort of issues which would be front page news on tomorrow’s model newspapers.

 

For example, I wondered about putting insulating fishplates on one rail, then connecting power from the adjacent section via a resistor? In theory the trains see lower voltage and slow for the curve regardless of the controller setting so are safe even when the operator is distracted.
 

I can see a few problems with this approach and would rather not cause fire / damage as I would then be “responsible for” rebuilding the layout / shed / house..

 

Has anyone tried it? Are there any better ideas short of DCC?

 

We’re talking about a lot of locos so I’d rather not go that way!

 

Thanks for reading..

 

Alan

Edited May 13, 2022 by Realistic_build_Speed
Auto “correct”

[Steven B]

A resistor would work but the voltage drop (and hence speed reduction) would depend on the amount of current the motor draws. A pair of locos double heading would reduces speed more than a single loco for example.

 

A pair of diodes connected anode to cathode will give a consistent voltage drop. Adding more diodes in series will further reduce the voltage.

 

 

Steven.

[Edwin_m]

You also need to be aware of the heat dissipation in the diodes or resistor, which is equal to the voltage dropped multiplied by the current passing through.  The power rating of the device needs to be greater than that figure, or it will burn out.  

[Michael Hodgson]

Or you could provide cant, aka superelevation, on your curves so that trains don't derail at speed.

 

If you need to reduce power to limit speed on the curves, the speed reduction needs to apply for the whole of the time the train is on the curve, not just the loco.  This would mean you must also reduce power on the exit straight for the length of your longest train.  If trains run fast in both directions on the same line that also means the entrance straight.  By the time you've done that, unless we are talking about a garden layout with very long straights there probably isn't much track left where you can run fast.  So in practice you might find it simpler just to limit the speed everywhere.

Edited May 13, 2022 by Michael Hodgson

[melmerby]

1 hour ago, Edwin_m said:

You also need to be aware of the heat dissipation in the diodes or resistor, which is equal to the voltage dropped multiplied by the current passing through.  The power rating of the device needs to be greater than that figure, or it will burn out.  

No problem if you use power diodes e.g. 1N54xx, they are rated 3A continuous and are cheap to buy

You will get 0.6v - 0.7v drop per diode. That's about 2W dissipation at 3A per diode.

 

At the current model trains run at, 200-300mA or less, it will be typically a tenth of that or less.

Edited May 13, 2022 by melmerby

[Realistic_build_Speed]

Thanks for the replies. I think I have a stack of diodes somewhere… 

 

Dad’s layout is in a shed, and now that I think about it the only really fast trains are mostly my son’s. Dad has a Hornby coronation with some serious pace, but maybe best to throttle that loco rather than do something to the power. The track took forever to lay so not keen to pull it up again.

 

My son’s is temporary at the moment, we put it out on lino and take down when not in use. His big thing is watching them fly by at speed so the idea was radius 1 set track curves at each end to maximise the straights. We’re debating putting it in the garden for more room, longer straights, higher speeds and frankly to get him out there! :) That way the two running tracks are next to each other along the fence and I don’t have an obstacle to mow under.

 

He likes long trains though so the point about the length of the slow section is well made.

 

Cant / SE may be a good option there to conserve momentum. I’m not sure how it’d work on R1 set track when the aim is to rotate it 180 degrees and back onto a 250mm /9inch baseboard in the shortest space possible.
 

My thought was 90 degree left turn, short straight c 15cm / 6in, another 180 degrees left, then 90 degrees right to face in the original direction of travel. I suspect set-track will be a bit stiff and the transitions to short to cant that.

More research to do!

Edited May 13, 2022 by Realistic_build_Speed
Spelling and detail

[Pete the Elaner]

I say try it. If you get unwanted results, you can always connect the 2 tracks with wires & they should give you a better connection than a rail joiner anyway.

I don't think it will work as you want though. A resistor bank (or diode) will have a more significant impact on the circuit at low voltages when you least want it.

[SR71]

I am not an electrical expert - other than making smoke which I'm highly proficient at.

 

On classic cars that use gauges from different sources some need 12v 9v or 6v to read correctly. I've used voltage regulators to limit the maximum voltage they see.

 

Small black box with three legs and a heat sink. Legs are in, out, and reference. Could this be used to cap the speed on the curves? Regular operation up to limited voltage but can't then be exceeded?

[DCB]

22 hours ago, Realistic_build_Speed said:

I tried to search for this but didn’t manage to find anything..

 

For example, I wondered about putting insulating fishplates on one rail, then connecting power from the adjacent section via a resistor? In theory the trains see lower voltage and slow for the curve regardless of the controller setting so are safe even when the operator is distracted.

Thanks for reading..

 

Alan

The resistor on the curve  concept is flawed, the last thing you want is to suddenly brake as the train enters the curve. in 00  the tension lock couplings will ride over each other as the train bunches up/  Also slow trains will get extra drag and less power and probably stop dead.   Cant or banked curves are a good idea.  (My teacher told me almost 60 years ago there was no such word as Cant ! ) ,   I really fancy 45 degrees on a 00 12" curve, my outdoor layout has cant on a 5 or 6ft radius good for the best part of 5MPH, 350 scale mph when I get runaways,

However a Diode based voltage limiter for the whole layout would work.  A pair of diodes in inverse parallel would drop voltage a bit, some people say with authority it's 0.7 volts, It's somewhere round there, and its pretty linear so 1 = 0.7  2X = 1.4, 3X = 2.1 etc     You have to have inverse parallel if you are putting it between controller and track or it would only conduct one way .
Diodes reduce voltage hence speed, Resistors only reduce the voltage when a load is applied    I have a home made diode controller with 11 diodes and a 12 way switch and it works very well, a lot better than a Hornby wall wart or H&M Duette, More steps would be useful as some steps need 2 diodes in series to get a good low speed crawl, but the concept works and it lets lets trains draw as many amps as they like which gives plenty of power, but limits the voltage which limits the speed.  Diodes are about 10p each, I haven't blown even a 1 amp one up yet and 4 amp are readily available so spend a few pence and give it a try.  Otherwise build a nice curved barrier along the trackside to catch the derailed trains

[melmerby]

5 hours ago, SR71 said:

 

Small black box with three legs and a heat sink. Legs are in, out, and reference. Could this be used to cap the speed on the curves? Regular operation up to limited voltage but can't then be exceeded?

Problem is they only work in one direction

Trains might be going in both directions and will have the opposite polarity one way

Pairs of diodes is the most practical. Cheap, simple and bulletproof.

[melmerby]

If you had enough track you could have graduated speed with one pair of diodes, then two, then three etc. as you approach the bend then the reverse as you come out of it.

[WIMorrison]

There is always the option of simply turning the controller dial to a realistic speed (rather the default Hornby 250mph) and then the trains will go around the bends OK 😆

[Michael Hodgson]

6 hours ago, WIMorrison said:

There is always the option of simply turning the controller dial to a realistic speed (rather the default Hornby 250mph) and then the trains will go around the bends OK 😆

I think the problem here is the age of the operators, and they see it as being a kind of Scalextric racing set.

 

If they won't turn the speed knob down, perhaps the answer is to put the diodes directly on the controller output, so they effectively can't get full speed anywhere.

[Realistic_build_Speed]

I like the diodes direct to the controller idea for the shed. For the four year old full whack needs to be retained on the straights.

 

The controller is a Model D so if there’s only one loop it would be possible to feed the curves and braking zones from one source at say 9V and have the others fed from the other side at “up to 12v” but then you get the bunching / decoupling effect as per DCB’s post.

 

My preferred concept is for a diode set between consecutive short sections so that the speed comes off slowly, as per  melmerby’s suggestion. May be quite jerky for the locos but they’re mainly very old ebay stocks - albeit still rather spry in their old age.

 

If he’s still interested in railways when older we could remove it and let him drive. Alternatively if it goes go in the garden then a larger radius + can’t solves everything..

[WIMorrison]

You could also reduce the voltage on the actual locos to reduce max speed rather than doing it on the track. 

[melmerby]

4 hours ago, Realistic_build_Speed said:

 

My preferred concept is for a diode set between consecutive short sections so that the speed comes off slowly, as per  melmerby’s suggestion. May be quite jerky for the locos but they’re mainly very old ebay stocks - albeit still rather spry in their old age.

 

 

Inertia will mean that you probably won't notice any jerks as you are only reducing the voltage in perhaps 0.7v increments, same when accelerating out of the curves as it increases again..

[Realistic_build_Speed]

Sold - I’ll get insulating fishplates on order and find the diodes.

 

Thanks all for the valuable insights. 

[SamThomas]

23 hours ago, Realistic_build_Speed said:

Sold - I’ll get insulating fishplates on order and find the diodes.

 

Thanks all for the valuable insights. 

Be interesting to see how this works out in practice.

[Realistic_build_Speed]

On 16/05/2022 at 22:08, SamThomas said:

Be interesting to see how this works out in practice.

Before getting too excited reflect for a moment on my username… 🙃