Cheap servos to control turnouts... Am I missing any unknown unknowns?

[Huskisson]

Hi I was wondering if any of you have used servos to control turnouts?

The official LGB motors for turnouts are extraordinarily expensive IMO plus you need to make or buy another bit of kit to drive them anyhow so I thought I'd make my first proper electronics project something that would do the job, hopefully at a fraction of the cost.

I already had an Arduino and a servo from my Arduino starter kit. So that is where I started. After several weeks of Googling and learning how to actually use an Arduino I've somehow muddled together this to automatically operate the turnout on my reverse loop (Its analogue so the loco always goes around in the same direction.)

When the train passes over the light sensors, the turnout switches to the correct orientation negating the need for me to worry about remembering to switching it myself. A task I had previously, manually, consistently forgotten, leading to several de-railings.

It seems to work well. The turnout switches at a satisfyingly leisurely pace and although the box I made to contain the servo is possibly a bit huge. I'm hoping that I can slim it down a little and maybe disguise it a something more tracksidey. Obviously I need to replace the breadboard with a proper soldered circuit board and make a box for it the Arduino and a power supply of some sort. I'm thinking that i'm going to make both the servo housings and the Arduino easily removable in case of really bad weather/long periods of track inactivity and just leave the wiring in place. Is this what people normally do?

Total cost under £21 compared to $40 ish for one LGB motor. Subsequent motors would only be the cost of the servo (about £3 each on EBAY) as the as the Arduino can control up to 12 of the little fellahs. I have another 4 manual turnouts (my "layout" is very small) at present so I'm wondering whether to take the plunge and splash out on an equal number of servos and save myself a couple of hundred quid.

However, after more Googleness, I'm now wondering whether servos are really the best motor for this job.

My question, should you choose to answer it, is this.

Does using servos for this task have any major disadvantages/pitfalls that I may not have considered. (Unknown unknowns.) Reading around I see that solenoids seem to be used in the majority (if not all) of the commercially available turnout motor "solutions"

I'm wondering if that's because...
(a) They are just as cheap as servos...
(b) Or maybe cheaper to run...
(c) Perhaps more durable/longer lasting...
(d) Some other reason I haven't thought of...

If anyone has any experience or advice in this area I'd love to hear it. I've learned a lot from my servo project so even if it turns out to be a dead end I won't too bothered.

Also I anyone has any ideas what I can disguise my (rather large) servo housings as that would be a help too. I'm thinking the Arduino and power could live in a signal box.

Finally, if anyone thinks this is actually a workable idea and would like more details about anything let me know. (With the caveat that I'm a total novice at both garden railways and electronics so Google is probably more reliable than me)

Ta

Mark

 

[PaulRhB]

The motor solution and gears are more prone to dirt and moisture in the garden as the tolerances are tighter. If it works go for it especially inside. Servos are now quite common in the smaller scales for indoor layouts and there are various dcc drivers for them too.

 

[GAP]

Looks good to me.

For the light sensors I suggest you look at "Sandstone & Termite" under electronics there has been some work using 2 LDRs to provide an input to a microcontroller that might help if you intend to use this in the outdoors.

Most servos are sealed quite well so should be OK outdoors.
If you are worried about moisture and dirt ingress to the servo you could always look at the ones used on model boats they should be well sealed but may cost a bit more.

I am open to controlling model railways with arduino.
I think arduino is going to be a handy system into the future with a heap of experimentation going on.

Once you have perfected this might I suggest putting it on the arduino forum to share with others in keeping with arduino's philosophy.

 

[gregh]

I've been using servos for turnouts, outside for years. I use Picaxe to control them. Here's my 3 year old article...
http://www.gscalecentral.net/tm?m=42887&high=servo < Link To http://www.forum.gscalece...m=42887&high=servo

The biggest problem I had originally was that I 'direct coupled' the servo to the turnout. This meant I couldn't move the turnout by hand (hence the little 'handle' on the servo in the above article). I've since developed a 'sloted link' method of connecting the servo to the turnout that does allow me to just push the turnout blades across manually. The servo pushes a pin in the slot to move the blades, then returns to a central position so the blades can still be moved manually.
The only problem I have had is some interference causing servo jittering. You must have the servo controller mounted right adjacent to the servo to avoid these problems.


I've also used servos to control a level crossing boom gate. All my servos are just in styrene 'boxes'.

 

[tramcar trev]

To throw in another method that I will eventually use would be to use electromagnetic actuators. They can be got inexpensively but would need cabling back to a central 'panel'. Power on to move the points one way, power off and the return spring snaps them back.

 

[Mice 1]

I too use Picaxe to control servo actuated turnouts. I agree with Gregh you can get servo jitter. Allowing hand operation and Picaxe control requires some sort of special linkage.
However if the control signal to the servo is set high a short time after the operation of the turnout the servo will still hold the turnout blades in the new position without any jitter but allow hand operation.

 

[gregh]

I agree with Gregh you can get servo jitter.
However if the control signal to the servo is set high a short time after the operation of the turnout the servo will still hold the turnout blades in the new position without any jitter but allow hand operation.

Thanks for that tip. I'll try it out.

 

[Huskisson]

Thanks for all the advice. I'll investigate PICAXE. I like the way you've disguised your motors and board Greg, they look great!

I might see if I can't do something similar if you don't mind me nicking your ideas... Being able to switch manually as well seems like a good one too.

Also the info about proximity to the controller causing jitter is something I hadn't considered. Is this caused by drop off over longer cables?
I'll do some tests with my set up to see if I have the same problem.

I have been disconnecting the servo control from the output pin on the Arduino through software when it isn't actually moving to avoid any stress on anything if a train should somehow go through the turnout in the wrong direction. I wonder if this would help alleviate any jitter also (or perhaps make it worse) ? Perhaps this is something similar to Mice's technique?

I agree Gap that the Arduino seems to be a really useful and easy to learn piece of kit. Especially for someone like me with no science or engineering background. There is such a large online community of users that it easy to find layman friendly walkthroughs for almost any application you can think of.

Tinkering with one has given me a great introduction to electronics. 6 weeks ago I didn't even know what a diode was.

That said, I'm now backtracking to learn some fundamentals as, like all things, once you learn a little you realise how much you don't know.

Anyway you guys have given me a lot to think about for version 2.0

Thanks again.

Mark

 

[PhilP]

I think the biggest problem for reliability will be our wet climate..
Yes, I know the rainfall levels are not to bad, but we get dew / condensation on many days of the year.
Ours would not be such a 'green and pleasant land' if we didn't!

Even minor surface corrosion on parts, or algal growth, will have a marked affect on the energy needed to start things moving over time.

It will be interesting to hear reports back over time.

 

[Gizzy]

The official LGB motors for turnouts are extraordinarily expensive IMO plus you need to make or buy another bit of kit to drive them anyhow....

Whilst I agree the LGB motors are expensive, they are well engineered for outdoors use and very reliable. If they do fail, they can often be repaired. I had problems with the screw connectors rusting on one or two, so I took the connector out, soldered a long flying lead to the coil inside and sealed up the holes with some mastic.

It's often possible to pick up cheaply 2nd hand on the forum or EvilBuy.
Also, they can be controlled cheaply by using the GAUGEMASTER GM 511 switch which can be easily mounted in a track diagram panel;

http://www.gaugemaster.com/item_details.asp?code=GM511&style=&strType=&Mcode=Gaugemaster GM511. Cost are £3.60 each.

I also have a Train Line point fitted with their motor, a servo type. This is DCC ready and cheaper than the comparable LGB items. I would certainly consider these for my future purchases.

However Mark, your article is very interesting and I shall be following to see how you get on with motorising your points.

Good luck with your project....