Storage Sidings Wiring and Control

[mike lee]

Before I finalize my layout, and start track laying and wiring, I thought I would seek out other peoples ideas for 'storage siding' wiring and control? Although I have put this in Electrics Non-DCC section, the design I have worked out 'on paper' is basically a DC system that is switchable so that it can also be used for DCC (Note, it has not been built yet). It is the standard design using Peco solenoids and CDU's via a diode matrix using a single push button to select each route. What I would like is other peoples ideas using Stall Motors (Tortoise) or servo's, I would much prefer the soft action of them rather than the 'smack' of a solenoid? My layout will be a through 'roundy roundy' layout with one up and one down line. The basic storage yard will have six through storage sidings on the up line and six through storage sidings on the down line. I would be pleased to hear of other peoples designs for storage yards, there must be hundreds out there. The main criterion is that it is primary a DC system, but it could also be me made into a DCC system. It should also be fairly simple!! and cheap using standard components if possible, not using Rasberry Pi's, Arduino's, MegaPoints Controllers or the like, although I know servo's etc do need some form of control device?

 

Hoping somebody can oblige, Regards Mike Lee 

[Pete the Elaner]

It sounds like you want to throw a point ladder to select a siding by pressing one button? I have thought about this in the past & because stall motors require constant current, I cannot think of a way to do it without using the complicated systems you want to avoid.

 

They do open up another possibility & the following is an idea I have had:

 

Each motor requires 2 connections instead of the 3 required for a solenoid. The conventional way to do this is to use a DC supply (12v, 5v, whatever the motors actually need) & reverse it with a DPDT switch to throw the point.

Instead of a basic DC supply, how about a something with 0v, 5v & 10v (It will be easier to think of it as -5v, 0v & +5v)? After the supply, things get a lot simpler.

The 0v is used as a return. The -5v & +5v is fed to the switches. This requires SPDT switches instead of DPDT & only 1 wire per point in addition to the common return.

Assuming you have pointwork at each end of your yard, you will need 10 points for 6 sidings. This requires 21 wires for solenoids, 20 for conventional stall motors but only 11 if you can wire them with a common return.

 

I have had a very quick look for a suitable power supply but cannot find one ready made. I did find a video showing someone making one by using 2 separate windings from a transformer (although they didn't explain anything), but that is how I thought of approaching it.

It would make an interesting project.

[Harlequin]

This is just the fuzzy outline of an idea:

 

Could you use the push buttons and the diode matrix to trigger some self-latching relays and then use those relay outputs to feed power to the stall motors?

 

[DCB]

5 hours ago, Harlequin said:

This is just the fuzzy outline of an idea:

 

Could you use the push buttons and the diode matrix to trigger some self-latching relays and then use those relay outputs to feed power to the stall motors?

 

That works but is very expensive as some of the cheapest latching relays are solenoid point motors. 

I have used diode matrix control for around 30 years and as long as you balance the load, If your CDU has power enough to throw 6 points at one time as mine has, then ensure 5 or 6 motors throw together not 1 sometimes or 2/3/4 then it all works very reliably.   The trick is to use spare point motor solenoids as dead loads when you only want to throw 1 or 2 or 3 or 4 points. Failure to do this results in point blades torn from tie bars

[jpendle]

I did exactly what you are thinking of on my old DC layout which used Tortoise motors.

BUT I didn't document what I did.

Anyhow, I used double pole double throw relays to switch the Tortoise motors. These were hooked up through additional relays to control route setting and it was all controlled from a 16VDC laptop power supply and a 16 position rotary switch. The rotary switch outputs were wired to the control relays such that if siding 4 were selected the relays controlling Tortoise motors on sidings 1 - 3 were set to the 'through' route and the Tortoise motor on siding 4 was set for the siding.

I still have the board but probably threw the wiring diagram away.

 

When I went to DCC, I selected a system that allowed graphical route setting (Roco Z21 Black), and that made the relay board completely redundant, as on the Z21 App on my iPAD I could set as many routes as I had sidings in software and when I selected route 4 (for example) the Z21 would know which points to throw.

 

If you do go DCC with Tortoise motors then I have found that the NCE Switch 8 is the easiest thing to use to control Tortoise motors.

 

Regards,

 

John P

[John ks]

I did something like this

 

 

With all switches off the points should be set for track 4

Turn on any switch & the points & power will be switched to that track.

The switch for track 4 only has to power the track as the points are already set to track 4

On small disadvantage is you can only have one switch on at any time

If you turn 2 switches on at the same time then power will be on to both sidings but the points will be aligned to the lower numbered track 

 

Hope this helps

John

[DCB]

If you go the rotary switch route with Tortoise motors then you only need the Rotary switch and the diode matrix, just leave the power on, no need for latching relays.   I have some rotaries controlling Solenoid motors through a diode matrix with a separate energise switch which works but isn't very satisfying to use, I much prefer the old electric pencil.    With Tortoise and matrix you can either ensure one end or the other of the tortoise is powered all the time the layout is powered up or let any points not used for your route float unpowered and unlatched.  I can see problems with tortoise motors, some of my solenoids have feeds from more than one matrix, which you can't do with Tortoises, 

[Ron Ron Ron]

The OP, Mike, states..........

 

"It should also be fairly simple!! and cheap using standard components if possible, not using Rasberry Pi's, Arduino's, MegaPoints Controllers or the like, although I know servo's etc do need some form of control device?"

 

I'm not sure how much "cheap" means, unless DIY electrics and/or electronics are involved; but there is a very simple solution that may be worth looking at, which does not involve the technical complexity or costs associated with the methods cited above as not being desirable.

 

Plug and play.

Usable with both DC and DCC layouts, with no conversion work needed to swap between the two methods of loco control.

i.e. a stand-alone system.

 

 

.

 

 

 

 

 

.

[Pete the Elaner]

12 hours ago, Ron Ron Ron said:

The OP, Mike, states..........

 

"It should also be fairly simple!! and cheap using standard components if possible, not using Rasberry Pi's, Arduino's, MegaPoints Controllers or the like, although I know servo's etc do need some form of control device?"

 

I'm not sure how much "cheap" means, unless DIY electrics and/or electronics are involved; but there is a very simple solution that may be worth looking at, which does not involve the technical complexity or costs associated with the methods cited above as not being desirable.

 

Plug and play.

Usable with both DC and DCC layouts, with no conversion work needed to swap between the two methods of loco control.

i.e. a stand-alone system.

.

A slight twist. The complexity is there, but the manufacturer has taken care if all that so the user does not have to concern themselves with it.

I think that is as close as you can get to the ideal answer for this.

Edited August 29, 2019 by Pete the Elaner
spelling

[Pete the Elaner]

12 hours ago, DavidCBroad said:

If you go the rotary switch route with Tortoise motors then you only need the Rotary switch and the diode matrix, just leave the power on, no need for latching relays.

Stall motors require a constant current which has to be reversed (or changed form +ve to -ve). A diode matrix is incapable of doing this because the same voltage will also be fed by the other 'slave' motor.

[Sol]

I do have circuit for diode matrix & stall motors - found on the internet in 2007 & the actual website is no longer accessible.

DIODE MATRIX MADE EASY.docx

 

when you open the attachment, you will need to cursor down part the blank area to get to more info

Edited August 29, 2019 by Sol
added an extra line of info

[Pete the Elaner]

28 minutes ago, Sol said:

I do have circuit for diode matrix & stall motors - found on the internet in 2007 & the actual website is no longer accessible.

DIODE MATRIX MADE EASY.docx 107.61 kB · 3 downloads

 

when you open the attachment, you will need to cursor down part the blank area to get to more info

Good find.

You will need a +/0/- supply as I described earlier & it will require a bit of trial & error. The easiest way would probably be to Just select the route you want & play about with the diodes until it works.

The maximum sidings is 6 (which is fine for the OP) because you will need 2 pole switches & the biggest I have seen support 12 connections (or 6 DPs).

[mike lee]

Thanks for the replies so far, keep them coming, the more variations the better?

 

There must be dozens of large club layouts with storage sidings, having lots of different variations on how they operate?

 

I am away from home at the moment so will have a look in detail when I get back.

 

Thanks once again,

 

Mike

[AndyID]

On 27/08/2019 at 06:09, mike lee said:

Before I finalize my layout, and start track laying and wiring, I thought I would seek out other peoples ideas for 'storage siding' wiring and control? Although I have put this in Electrics Non-DCC section, the design I have worked out 'on paper' is basically a DC system that is switchable so that it can also be used for DCC (Note, it has not been built yet). It is the standard design using Peco solenoids and CDU's via a diode matrix using a single push button to select each route. What I would like is other peoples ideas using Stall Motors (Tortoise) or servo's, I would much prefer the soft action of them rather than the 'smack' of a solenoid? My layout will be a through 'roundy roundy' layout with one up and one down line. The basic storage yard will have six through storage sidings on the up line and six through storage sidings on the down line. I would be pleased to hear of other peoples designs for storage yards, there must be hundreds out there. The main criterion is that it is primary a DC system, but it could also be me made into a DCC system. It should also be fairly simple!! and cheap using standard components if possible, not using Rasberry Pi's, Arduino's, MegaPoints Controllers or the like, although I know servo's etc do need some form of control device?

 

Hoping somebody can oblige, Regards Mike Lee 

 

I agree with what you're saying about solenoid point motors. Personally I think they should be called "seismic simulators". The old H&M motors could knock stuff off the track

 

I'm not sure if you can make this work with Peco points but one thing I have done in the past is take advantage of the fact that the exit points are trailing and just let the trains "self set" the blades. That immediately halves the number of point motors. One snag with it is that unless you use frog juicers you do have to use dead frogs.

 

If you don't mind doing a little work with Veroboard, using 555 timers you can control as many servos as you care to for very little cost.

 

You'll need three 555's. One of them is configured to run continuously as a clock. The clock timer kicks-off two other 555's. They are configured to produce a single pulse (aka monostable or one-shot) every time they get "clocked". The servo inputs are connected to the outputs of either of the two monostable 555's via a single pole change-over switch. The widths of the monostables' pulses determines the extents of the servo travel.

 

So what you are doing is connecting the servo input either to a long pulse or a short pulse through a simple switch and you can use the same two pulse generators to feed as many servos as you like. The "clock" generator can be adjusted to determine the servo speed. Normally for model aircraft you want the servos to move as quickly as possible and they are clocked at 50 times per second but you can make the clock as slow as you like. If you make it really slow you will see that the servo travel is actually a series of steps

 

What you don't get with this method is a means to adjust the throw of each servo individually. They will all throw through approximately the same angle so you have to use some sort of mechanical arrangement that takes that into account.

 

You can also do this by programming a micro-controller as a pulse generator. That's the way i do it but I can't blame you if you don't want to get into that.

 

EDIT: (More info)

 

If you rig the servos so that the short pulse drives each point straight (not to turnout) and supply short pulses to all the servos the route will be to the last loop in the ladder (you might call it "Loop #6"). To select any other loop you only need to supply its servo with a long pulse to make it "turnout". You can "logically OR" the existing short pulse with a long pulse supplied from a single pole six-position rotary switch. The logical OR can be pair of diodes at each servo's input although it would be better to use some OR gates to maintain the right input voltages at the servos inputs.

 

Note: Depending on how you construct the loops the first point might be "short pulse to turnout" rather than "short pulse to straight".

 

 

 

 

Edited August 31, 2019 by AndyID
Gramer and spelin

[RobjUK]

13 hours ago, AndyID said:

You'll need three 555's.

 

You can drive a servo perfectly with a single 555.

 

See my old article in Practical Electronics, it's on pages 54 - 56 (as shown in Acrobat reader) or 1260 printed number, of the magazine archived here:

https://www.americanradiohistory.com/Archive-Practical-Electronics/70s/Practical-Electronics-1978-12.pdf

(Tony - Anthony - is middle name, which family always use but confuses people with my initials...)

[And Just realised that's over 40 years old now! Time flies.]

 

To just generate a fixed servo pulse, all you need is the 555 section on the left.

 

You can use two of those and switch between them, or use the full circuit and feed the "input" from another 555-only section.

Turn the input one on and it will take over, turn it off and it switches back. A 74HC00 is probably easier to get now and will work the same.

 

Or just use a single 555 section and switch another higher value preset in parallel with the 5K one, possibly 47K ?

Start with that at maximum. It will allow a second position to be set relative to the first one, when connected.

 

Edited August 31, 2019 by RobjUK
typo

[AndyID]

8 hours ago, RobjUK said:

 

You can drive a servo perfectly with a single 555.

 

 

Yes you can. The advantage of using three is that you can adjust the end positions and the speed all independently of each other.