Yes the ones who knows me will say, he is back again with these motors.
Yes servos a re quiet cheaper than Tortoises.
I have try to use them as a stall motors , following a little article in MR some years ago; even by using different values of resistances, they all burnt after a few weeks, especialy when in a long session.
For some people and again today, stalling motor is the best way to go even if servo's seems cheaper and easier to use.
In the ten last years some empire layout build in this period ( Tony Koester, Georges Sellios, Howard Zane, David Stewart,...) didn't have used servos for turnout but again Tortoises or an another one, which I like.....
Yes I like it and he is an excellent stalling motor, not often proposed, the Hankscraft display motor, the one sold by Builder in Scale as the "switchmaster".
You can order it directly to Hankscraft motor, and in big quantities.
In quantitities you have a price which is largely under the Tortoises and yes, unfortunately over servo's but this is an excellent deal in all case.
I have around 25 of them working since the early 80's and none has ever failed.
I have right now placed a big order of them by Hankscraft for my extension of my layout and the deal is very interesting in any aspect.
Yes I'm a convinced stalling motor man, but not from Tortoises ( I use around 50 Tortoises on my layout) but by Hankscraft display motors, because they are extremly simple to be put in place, need no adjustement, are powerful enough for all the scales ans seems unbreakdable with a good price in mind.
Just my opinion
Not an expert, but I did dabble a bit in RC, more cars than planes, but same deal with the steering servo - you need to pick servos with torque capacities matchign the load so they are not operated in stall. Momentary overtorque doesn;t hurt anything, but constantly in stall - even if the servo doesn;t fry, is a huge battery drain. Last thing you want in your RC plane is for the radio battery to die while you still have half a tank of fuel. Not as big a deal in a car (until it stays stuck on full throttle with no steering and drive off the track and right out on a highway). I would ventur to guess that in my car, the times I smacked a wall, it probably pushed back on the servo enough to take it off position - until the nylon ball joints popped like they were designed to do.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
in Turnout Motors from Servos, the author says the HS-311 has an unusually low stall current.
i wonder what the typical current of an RC servo is in operation with aerodynamic forces pushing against and it essentially not moving. So such servos may be designed to operate essentially stalled.
but if that's really an issue, what about a momentary toggle switch and just holding it down until the motor stops.
greg - Philadelphia & Reading / Reading
I've not played around with those, but I suspect they are not very robust when the load wants to draw more current than it supplies. Some may be overload protected, in case the LED fails short, or the string of LEDs gets shorted, but I'll bet the 'fuse' in most such things is the cheap little CC driver itself.
The DMX LED controller box I got to run LED strips - it's rated for 8 amps I think per output, questionable although they are some pretty beefy MOSFETs and they are secured to the metal case with thermal pads - but thee is NO protection against shorts whatsoever. Connect a piece of wire between the common and one channel, and the full capacity of the power supply will get shunted right through the MOSFET for that channel.
What might work well in this situation is a simple 1-shot circuit - since momentary buttons are being used just to run the servo long enough to flip the Peco spring, there's no reason for the servo to get hot unless someone holds down the button - like a solenoid switch motor. A CD supply won't work for a servo, but some way that only provides power long enough to move the points regardless of how long you hold down the button, or even if you just give it a quick stab (one of the reasons I don't like pushbuttons for a slow motion/stall type system without a driver - a friend had his layout set up like this with pushbuttons for his Tortoises. Granted, when he initially set it up, he had built everythign to N scale fine scale standards with Code 40 handlaid track - so the distance the points actually had to move was pretty miniscule - but you had to hold the button down until the Tortoise moved) so you get a full throw each time but th power is removed so the servo isn't stalled and generating heat.
rrinker Seems like he's not stalling anything - just activating the servo long enough to get the Peco spring to snap over center then turning off the power. I have 20ma regulators? The buck convertors I use CLAIM 3 AMPS, but with no heat sink on the main chip (and the fact that at that price, they aare probably counterfeit chips to boot) I wouldn't load them more than 1 amp - so far the one installed on my servo controller board works fine, no issues with both relays pulled in and both servos moving. --Randy
Seems like he's not stalling anything - just activating the servo long enough to get the Peco spring to snap over center then turning off the power.
I have 20ma regulators? The buck convertors I use CLAIM 3 AMPS, but with no heat sink on the main chip (and the fact that at that price, they aare probably counterfeit chips to boot) I wouldn't load them more than 1 amp - so far the one installed on my servo controller board works fine, no issues with both relays pulled in and both servos moving.
I used the servos with Peco code 83 and momentary contact actuation at 9V. With the centering spring only momentary actuation is required. Stuck the servo to the bottom of my layout with the outdoor quality two sided tape after cleaning the plywood with masking tape. Here is a quick video.
https://www.youtube.com/watch?v=Tyxa3nqnbhY
Thanks, it makes a lot more sense now. It looks like it will be a fantastic layout.
I have a good friend who did something similar modeling Penn Station in Baltimore, getting very close to the actual trackage.
Personally I have never been motivated to try to duplicate prototype trackage that closely.
Seems to me you can seldon do the distances justice even in a big space.
My new layout will have a four track thru passenger station, with a small coach yard and REA/Post Office siding, it will be able to handle passenger trains as long as 15-18 cars, and be able to park two 12 car trains.
That's enough complexity for me.
I like operation, but I am just not into this heavy minutia of duplicating every motion/track of some massive prototype terminal like that.
I also have a strong desire to return to an "old school" ratio of scenery to track, having deeper scenes and more non railroad scenery. The area where my freight yard and passenger terminal are will be five feet deep. A six track freight yard in front, double track freight mainline behind that. At a slightly higher elevation, the four passenger terminal tracks and platforms with an overhead station. That will all take about 30", a depth I can easily reach.
The remaining 30" behind that will be the city scene, which will be built on lift out sections. Some staging tracks will be under the passenger terminal tracks and easily reached from underneath with or without removing the lift out scenery.
Filling a 24" or 30" shelf with track and pasting a backdrop of relief buildings behind them is just no longer acceptable to me.
Even the more "rural" sections of scenery will be 3' to 4' deep, with the trackage comfortably in the front half.
All my operational stuff will be along the front, industries, etc, but again I am interested in developing more "non railroad" scenery for better context to the scenes.
Turnouts, since we have been talking about turnout control, all my trackage is arranged so that turnouts that would be manual on the prototype, will be ground throws on the layout. Only hidden staging and CTC controlled turnouts will be switch machine controlled.
Sheldon
it's the east end of a passenger terminal. He's trying to model the trackage at the New Haven station.
the photo doesn't show most of the turnouts, just the trackage in from of the station
the west side interlock seems even more redundant. I think it's going to take a bit of operation to figure out how to select routes. The software approach will make it easier, even possible to quickly reconfigure the panel during operation.
Greg, it would be interesting to see the actual track plan of the complex interlocking rather than just the "schematic", so as to understand the operation. Is that a passenger terminal throat?
Seems like a lot of redundant trackage?
a very complete basement layout by a club member uses nothing but proto boards for all his electronics, dozens on a panel and several panels. (he's an MIT EE and did switch contact engineering)
I'm not so sure I would use a proto board as a 'permanent' solution for a relay socket. The cheap ones are cheap, for a reason. If the relays are not the soeckets type, consider drawing a small PCB with big solder tabs to use as the relay mount - you can have a place like JLCPCB do a few sheets with them scored for easy snapping apart for not much money, work out to a few cents per board. Have the solder tabs, but spaced and drilled to fit screw terminals in case you might prefer to solder on some screw terminals instead of soldering wires right to the PCB.
THis was nearly unthinkable just a few years ago, but now these places do this for less money than it would cost to buy the chemicals to etch them yourself.
You are quite welcome.
Keep in mind that I am also directing DC track power and providing signaling interlocking info all in the same system.
So being able to use the same relays, with 5 amp contacts, and thereby use any available contact set with the correct logic is a plus.
As I mentioned, the wye is the most complex and actually requires repeaters for functions not covered in this discussion.
But in most applications, one relay per turnout provides route control of DC power, turnout route control and signal interlocking.
The other streamlined feature of my system is the power supply. The 24 volt/12 volt center tap power supply provides Tortoise power, relay control power and signal system power all in one.
And call me neurotic, but I would not really be excited about having multiple different "systems" for each situation.
thanks for the explanation. i'm interested in understanding various approaches for solving problems
there are many approaches to solving problems. being aware of the approaches and using available hardware help determine the most suitable for the situation.
in our immediate case, we're using SPDT toggle switches to control a group of tortoise switch machines for a small yard off the main. The issue was being able to control just the mainline turnout from dispatch. Using a latching relay with momentary switches is a solution.
mounting of the hardware is another concern. we plan on using solderless proto board in the panel to mount the relay an connections to switch and tortoise
for your wye, i believe three latching relays and three momentary switches is another approach. Since each coil of the latching relay is controlled by a single mom-switch, no diodes are required.
| A B C A 1 n r _ 1/ \2 2 r _ r ---B-----C--- 3 _ n n 3
of course other approaches work as well. but depending on the complexity of the track arrangement, can be more difficult to build, debug and maintain.
the interlock below uses an arduino to do route selection and boards controlling latching relays costing ~$400.
Because this uses a software approach, mistakes in the orignal code were easily corrected and unplanned modifications made.
a second interlock will take a different approach using op-amps to drive the tortoise machines, reducing the total cost to ~$40
of course there's no benefit in redoing the first interlock.
The relays I use are like this, so they can also switch track power. One size fits all for the whole cab control system, cab selection, turnout control, and signaling.
https://www.citrelay.com/view_relay.php?series=J152
Extra contacts on the turnout relays are used for powering frogs.
So in the case of the wye control, one set of form C contacts on each relay is used to switch the polarity to the Tortoise machine, two sets are used for the logic, and the last set powers the frog.
The relays are mounted locally near the turnouts, CAT5 control wires go to the pushbutton locations. 24 volt +, 24 volt -, and one control wire for each button which handles the control input and the LED in the button.
If you look at the diagram, the wire from the button to coil is the same wire that powers the LED. It is always hot when the coil is energized so it lights the LED as well.
When the coil is dropped out, that wire is de-energized, the LED goes out.
Well Greg,
Without the A and C normally open contacts, all the relays will try to energize on power up, so yes they are necessary for B to power up.
All the contacts are necessary at some point in the cycle to get the desired action.
The normally open contact on A is what seals in relay A after you push the button for A.
The normally open contact on C is what seals in relay C after you push the button for C.
When the circuit is powered up, B is energized. When you push button A, B drops out, and A is sealed in.
When you push button C, relay C is energized and A or B drop out and C is sealed in.
And so on........
PS - It should be noted that the wye is actually one of the most complex of these circuits. Even complex multi route interlockings are typically made up of several simpler circuits, sometimes with some electrical interlocks to prevent undesireable redundant routes.
i found a description of relay symbols. If i understand the diagram correctly, the normally open A contact neat the upper left is unnecessary, leaving normally closed C and B contacts in series with relay A. And similarly, the normally open C contact bottom left is unnecessary, leaving normally closed A and B contacts in series with relay C.
Or are they necessary to force the initial state to be relay B energized?
i can see pressing a button energizes a particular relay which opens contacts to deenergize the other two relays.
gregc can you explain the symbols on your schematics these are the relays we plan on using, EA2-5TNG, 10 @ $22
can you explain the symbols on your schematics
these are the relays we plan on using, EA2-5TNG, 10 @ $22
Greg,
A circle with a letter is a relay coil
A break in a wire with two parallel lines is a normally open contact
A break in a wire with two parallel lines with a diagonal slash is a normally closed contact
The letters identify which relay coil the contacts belong to.
The upside down "T" over the break in the line is a normally open push button.
Wires that cross without a "dot" are not connected.
These are all standard symbols for relay logic ladder diagrams that have been used since the invention of the relay.
All contacts are shown in a de-energized condition.
It's not rocket science.
I will happily explain more if needed.
It is relays, everything is discrete, on or off, no "high or low" like solid state.
On the diagram for the wye, powering up the circuit energizes relay B, and aligns the straight thru route at the bottom of the wye.
Pushing the button for a different route, releases that relay and energizes one of the others, which in turn energize the turnouts in the opposite direction as needed.
The lower part of the second drawing shows how the Tortoise machine is controlled by the relay contacts, using the dual voltage power supply.
gregc ATLANTIC CENTRAL So a control panel that controls 5 routes, and may involve 10 to 15 turnouts, only needs 7 wires back to its relay station for both control and indication. where necessary, both for operation from multiple locations (local/dispatcher) and for route selections (+ diodes), latching 2 coil relays and push buttons can be used to control tortoise machines. SPDT/DPDT toggle switches would normally be used members of the PSR typically use a Tortoise as shown below with linkage, no need to precisely position it directly under the tie. i assume a servo could similarly be mounted
ATLANTIC CENTRAL So a control panel that controls 5 routes, and may involve 10 to 15 turnouts, only needs 7 wires back to its relay station for both control and indication.
where necessary, both for operation from multiple locations (local/dispatcher) and for route selections (+ diodes), latching 2 coil relays and push buttons can be used to control tortoise machines. SPDT/DPDT toggle switches would normally be used
members of the PSR typically use a Tortoise as shown below with linkage, no need to precisely position it directly under the tie. i assume a servo could similarly be mounted
Greg, I have explained in the past, and posted schematics showing why and how you do not need latching relays or diode matrix systems for advanced route turnout control.
The two diagrams above show how I do this with simple, inexpensive 24 VDC 4PDT ice cube relays and lighted LED pushbuttons.
In the case of the first diagram, all three turnouts of a wye are controlled with only three push buttons and three relays. You just push one button to select the desired route, all the turnouts move as needed.
And you can add as many push buttons to the "stack" for as many multiple locations as you desire. The diagram only shows local and dispatcher, but you could add more.
Similar logic is easily designed for any situation, usually requiring only one relay per turnout, sometimes two, and sometimes less than one per turnout.
Latching relays actually make this more complex than it needs to be.
I was not really looking to debate this with you again, I was simply curious as to the motive.
I'm not much for re-inventing the wheel just for the fun of it. The circuits I use for my controls are 100 year old motor control circuits and relay logic that is basically the same as what has controlled machines in industry and operated the orginal signal systems on the prototype - a proven system that ran the machine age in this counrty for 80 - 100 years - before the age of solid state and computer logic.
I'm not against solid state, but I'm not into change just for the sake of change either. Show me why it is better or less expensive and just as good, that is a reason for change.
Also, Circuitron makes a remote link for the Tortoise for difficult mounting situations and to allow one machine to operate a crossover.
http://www.circuitron.com/index_files/Tortoise.htm
rrinker Do you have a photo of your mounting Mel? Are you just gluing it together? --Randy
Do you have a photo of your mounting Mel? Are you just gluing it together?
ATLANTIC CENTRALSo a control panel that controls 5 routes, and may involve 10 to 15 turnouts, only needs 7 wires back to its relay station for both control and indication.
ATLANTIC CENTRAL Ok, one more question. With servos you have to rig up a linkage? I know that's likely pretty easy, Sheldon
Ok, one more question. With servos you have to rig up a linkage? I know that's likely pretty easy,
Expensive - and big. If room under the layout is important, anyway.
Yes, a servo needs a linkage - most common is to do the exact same thing as the way a Tortoise works. As the servo arm swings side to side, it moves a piece of piano wire side to side, which pushes the points. You can glue the servo to the bottom of a turnout, Peco style, an SG90 is about th esame size as Peco's solenoid anyway. You can hde the servo in a trackside building and run a pushrod under the roadbed to reach a turnout several tracks over. Theoretically, if you had the skill, you could put the servos inside a tower and have the rodding actually work to move the points. Bonus points for the levers in the tower also moving - if someone hasn't already done that, I would love to see that happen. HO and larger shouldn't be insanely ridiculous.
Related - the control over a servo is quick enough that you can use them to drive semaphores and actually have them 'bounce' when they drop. That HAS been done, many times.
Price is just one factor, but it's what got me to try servos in the first place. At the time, you could get the controller from Tam Valley, with buttons and LED indicators, plus it could also be used as a DCC decoder, plus the servo and a mounting bracket for the servo, all for less than the price of just a Tortoise, no switch or relay, no LEDs for the panel. Prices have gone up, but if you factor in buying a toggle switch and LEDs to control the Tortoise, it's still a wash, and that's buying the Tortoises always in bulk packs. DIY on the servos and the controller, and it still is significantly less expensive than a Tortoise.
But that's just one part - the servo is also much smaller, and can fit in smaller spaces. And off fthe shelf custom linkage parts - well, while many hobby shops no longer have trains, there are still plenty with RC cars and planes.
rrinker Here's the thing, if you DON'T already have the Tortoises - servos are MUCH cheaper. To use them the proper way, a single $3 Arduino will easily drive 4 of them plus use your same LED pushbuttons to provide 3 control locations for each of the 4 turnouts. ANd still not use all the pins. Without resorting to cleaver multiplexing tricks - if you really know what you're doing along those lines, you can expand the number of control locations, or total number of turnouts controlled by the one Arduino, or some combination thereof. If you already have boxes of Tortoises and relays, no, it does not make sense to junk it all and go buy servos and Arduinos. --Randy
Here's the thing, if you DON'T already have the Tortoises - servos are MUCH cheaper.
To use them the proper way, a single $3 Arduino will easily drive 4 of them plus use your same LED pushbuttons to provide 3 control locations for each of the 4 turnouts. ANd still not use all the pins. Without resorting to cleaver multiplexing tricks - if you really know what you're doing along those lines, you can expand the number of control locations, or total number of turnouts controlled by the one Arduino, or some combination thereof.
If you already have boxes of Tortoises and relays, no, it does not make sense to junk it all and go buy servos and Arduinos.
With my relay circuit, I can in theory stack as many additional location push button sets as I want with no additional inputs. All the push buttons are in parallel, the relay only knows that one of them closed. There could be 10, or 100, the circuit would not care.
Once the hot and neutral of the control circuit is available at the panel location, each pushbutton and it's LED only need one wire.
So a control panel that controls 5 routes, and may involve 10 to 15 turnouts, only needs 7 wires back to its relay station for both control and indication.
But I get it, Tortoise machines have become expensive.