Reliable frog switching on DC

[spikey]

Points are all Peco Code 75 Electrofrog with the link wire on the underside cut and track feeds attached accordingly, and point motors are all Gaugemaster solenoids unswitched.  How do I best achieve totally reliable frog switching?

[Pete the Elaner]

I mounted microswitches so they are activated by the motor pin. They have proven much more reliable than the SEEP included switch or Peco's PL-13.

[Happy Hippo]

When you say link wire, do you mean the one that is attached to the crossing nose?

 

If it is you will need to re attach it and connect it to either a micro switch or a relay in order to get effective crossing nose (aka frog) switching.

 

See the Gaugemaster GM500 wiring diagram.

 

https://www.gaugemasterretail.com/magento/gaugemaster-gm500.html

 

Pete has already covered the microswitch approach.

 

 

As a general point to all track layers, if you are not sure or do not understand, ask before you start.

 

RMWeb is probably the largest free advice service on the planet.

 

 

[spikey]

Thank you gentlemen.  Points are wired as below, and relays as per those diagrams.

 

The problem I'm getting every now and then is points switching but frogs not, with the result that things get in a muddle.  This is resolved sooner or later by some seemingly-random coincidence of me moving the point rails manually and trying again with point switch in one direction or the other.

 

The problem crops up most often with the pair of facing three-way points which are on a diode matrix (of sorts), but also on one particular pair of ordinary facing points which both change on the one switch.  Funnily enough, another pair of ordinary facing fired the same way hardly ever get themselves in a muddle ...

[Harlequin]

I think @Pete the Elaner's solution has the over-centre spring in the turnout fighting against the spring in the microswitch in one of the two states, if I'm reading the photo correctly.

 

Obviously it works for Pete but that conflict, with the possibility of the microswitch moving the tiebar if it overcomes the force of the over-centre spring, would worry me.

 

I would try to get the microswitch movement roughly at right angles to the tiebar/drive rod/solenoid bar movement with some sort of cam arrangement.

 

 

To switch your frogs, Spikey, it might be simpler to replace the point motors with ones which have switching already built-in rather than trying to retro-fit microswitches.

 

[Free At Last]

10 minutes ago, Harlequin said:

I think @Pete the Elaner's solution has the over-centre spring in the turnout fighting against the spring in the microswitch in one of the two states, if I'm reading the photo correctly.

To combat that, as I am still using the  built in switch of the PM1 I used this method...

[spikey]

1 hour ago, Harlequin said:

 ...To switch your frogs, Spikey, it might be simpler to replace the point motors with ones which have switching already built-in rather than trying to retro-fit microswitches.

 

 

Alas, particularly in the case of those points which are the most troublesome, there's either insufficient space for microswitches or the motors are surface mounted.  

 

As for swapping the solenoids for the switched versions, the reason they aren't already that  type is simply that I'd found those (i.e Gaugemaster PM1) to be unreliable on a previous layout.  I have no idea how reliable any of the alternatives that are now available are reckoned to be.

[WIMorrison]

Swap them out for  MTB MP1 - so much more reliable. You can buy them from several UK shops

[spikey]

Ref the MTB units, my initial impression is that the noise would be an irritation, and that I've been unable to find anything online which leads me to expect that they would last.  It also seems like everybody's waiting for stock!

 

So ... what are my alternatives nowadays then?  In view of my experience of servos in model aircraft and boats and the characteristic noise of them I reckon they're out of the running, and the Peco switched solenoid always struck me as a bit naff, so what am I left with?

[Phil Bullock]

Depends what you want to spend. Stall motors (tortoise, cobalt) will cost you in excess of 20 pounds each and although they are often associated with DCC can also be used with DC. 
 

Tortoises on DCC were excellent on the original Abbotswood, having gone through 3 different motor installations . 

[WIMorrison]

There is a chap in Germany that has tested them for more than 24,000 activations, do you think a Seep or Peco will last that long? 

 

As for noise, it is a lot more pleasant than the thump that you get from Peco and Seep, plus you get a nice slow motion change and a proper microswitch for polarity changing

 

EDIT - This thread has all you want to know on these motors

 

MTB Point Motors - Any Experiences? - Modelling Questions, Help and Tips - RMweb

 

the long term testing is aslo in the thread but here it is again

 

Drehscheibe Online Forums

Edited November 25, 2021 by WIMorrison

[Harlequin]

I have fitted MTB MP1s to my Minories box (to operate the points and switch the frogs) and I'm very happy with them. Yes, they whirr when activated but not really loudly and as Iain says it's no worse than the loud clack of the old solenoid motors - it's just different.

 

They are very simple internally and you can open them without voiding any warranty - in fact, you have to if you want to change the distance of the throw (which is very simple - just moving a peg from one hole to another). It's just a motor, 2 gears, a cam shaft and 4 microswitches (IIRC). No electronics!

 

The control is simpler than servos (no pulse generator needed). Just apply voltage to one pin to go one way and to the other pin to go the other way. When they reach the end of the travel they turn off and don't consume any power until you make them go the other way (no stalling). This means you can hook them up to a standard single pole double throw switch on your control panel, with the direction of the switch giving clear indication of the direction of the points.

 

To operate multiple motors at the same time no CDUs are needed because the power requirements are more evenly spread.

 

Edit: One more thing: they are much smaller than Cobalt and Tortoise motors.

 

Edited November 25, 2021 by Harlequin

[Izzy]

I have to say that these MTB's appear to be a super-dooper commercial version of the hacked servo type of control that I use and not dissimilar to the motor driven types available in the past. I actually think the odd one might still be around today, Conrad? not sure. The lack of current draw except when they actually move is good. The current fad for servo control via electonic boards seems to fall down when more than a few are used due to the very high current draw on start-up with lots of work-arounds needed, separate high amp voltage supply or staggered start-up. What a faff.

[spikey]

 

12 hours ago, Phil Bullock said:

Depends what you want to spend ...

 

As little as possible: my Lady Wife and I are officially rated as living "in relative poverty".

 

Any device that uses power without turning it into movement offends my engineering sensibility, so stall motors are a no-no here.  It does look like MTB might be worth a shot, but alas if any online source actually has any in stock to sell me, I have yet to find one ...

 

Thank you for your input, gentlemen.  

[Stubby47]

You could replace all the motors with a Wire-in-tube system, with a slide switch both moving the tiebar and changing the frog polarity.

 

[cctransuk]

20 minutes ago, Stubby47 said:

You could replace all the motors with a Wire-in-tube system, with a slide switch both moving the tiebar and changing the frog polarity.

 

 

Exactly what I've done with my new-build layout - and it works a treat!

 

CJI.

[Barclay]

I have experimented with a hacked servo as advocated by Izzy and it works very well on the bench. The beauty of it is that you can operate it with a DPDT switch - one set of poles operates the servo and the other gives 100% reliable polarity switching. It's also very cheap - servo, less than £1, plus a DPDT switch and a small SPDT switch. That's it!

[Pete the Elaner]

21 hours ago, Harlequin said:

I think @Pete the Elaner's solution has the over-centre spring in the turnout fighting against the spring in the microswitch in one of the two states, if I'm reading the photo correctly.

 

Obviously it works for Pete but that conflict, with the possibility of the microswitch moving the tiebar if it overcomes the force of the over-centre spring, would worry me.

 

I would try to get the microswitch movement roughly at right angles to the tiebar/drive rod/solenoid bar movement with some sort of cam arrangement.

 

 

To switch your frogs, Spikey, it might be simpler to replace the point motors with ones which have switching already built-in rather than trying to retro-fit microswitches.

 

 

I was concerned about the force used to throw the microswitch but I was surprised how soft it was. I only used them in the first place because the inbuilt switch on the PM1s was unreliable. 2 out of 5 I installed would not switch the frog at all.

That was after suffering failures with Peco's PL-13s.

I have tried Tortoise motors & they worked nicely but stuck out below the baseboards I was using a the time. They are also a continuous current motor so they need wiring differently from a solenoid.

[melmerby]

22 hours ago, spikey said:

Thank you gentlemen.  Points are wired as below, and relays as per those diagrams.

 

The problem I'm getting every now and then is points switching but frogs not, with the result that things get in a muddle.  This is resolved sooner or later by some seemingly-random coincidence of me moving the point rails manually and trying again with point switch in one direction or the other.

 

The problem crops up most often with the pair of facing three-way points which are on a diode matrix (of sorts), but also on one particular pair of ordinary facing points which both change on the one switch.  Funnily enough, another pair of ordinary facing fired the same way hardly ever get themselves in a muddle ...

If you don't mind me commenting, you've made a pig's ear of removing the links in the switched rails.

If you use a small flat bladed screwdriver and exert some sideways force against the spot welds, they will give. It results in a complete removal of the wire and no noticable damage to the plastic.

I have done 60+ electrofrogs and it works every time.

[WIMorrison]

Just now, Pete the Elaner said:

 

I was concerned about the force used to throw the microswitch but I was surprised how soft it was...

 

That is why they were called microswitches - it wasn't because of their size, it was because they took so little effort to operate

[melmerby]

1 hour ago, Stubby47 said:

You could replace all the motors with a Wire-in-tube system, with a slide switch both moving the tiebar and changing the frog polarity.

 

Not practical for many layouts

[melmerby]

These are pretty good:

https://www.dcctrainautomation.co.uk/mp1-point-switch.html

 

Just need a changeover switch to operate and have frog switching.

[spikey]

25 minutes ago, melmerby said:

If you don't mind me commenting, you've made a pig's ear of removing the links in the switched rails.

 

I don't mind you commenting at all. I've probably only done 40 or so of them over the years, and what with deteriorating eyesight and the advancing arthritis in my hands, it's true that the more recent ones are less than perfect.  However, nobody but me will ever see that and it's made no difference whatsoever to the mechanical integrity of the point, so I can live with it. 

 

WIT operating is attractive on the face of it, but if nothing else I really couldn't be doing with all the pushing and pulling required to operate a pair of facing three-way points every time I need to use my storage loops.

[stewartingram]

I'm surprised nobody has commented on an idea published in Railway Modeller - I've no idea when, I have a copy of the article but copyright rules mean I can't put it here. However, I'll try to describe it, so stick with me....

 

I use the traditional 'probe and stud method on the control panel, or rather I did until the major rewire/tidy up that I'm going through at the moment. Just for clarity, as some may not be aware of the stud and probe method:-

 

The solenoids (motors) on the point need a non-permanent supply fed to them, if it was maintained the motor would burn out. So the output of the cdu is fed to a probe, which is touched to a stud/screw on the panel. The screw is connected to the motor coil. Thus there are then two studs per motor.

 

This has no effect on the frog polarity, so I mounted microswitches next to the motor to switch the frogs.Simple wiring, but messy to do to get the actuation of the switches. I disregarded any switches, such as those integral with Seep motors (etc), as unreliable or difficult to set up properly.

 

The new idea from RM has seen me rip out all the microswitches, and the studs. Instead, I've put a double-pole changeover (DPDT) switch - with no centre off position - on the panel instead. The r/h pole is for the motor, with the l/h pole for the frog polarity. Wiring for the polarity is simple; frog to the common centre contact, and the stock rails of the point (+/- if you like) to the other 2 contacts of that pole. So whichever route you select, the switch physically shows the route, and the frog is electrically switched.

The nice bit is that the other pole of the switch , at the same time, selects the motor coil for that direction. But it is not electrically live; only when the probe is touched to the metal part of the switch (ie the lever). This is because the centre contact of the r/h pole is wired to the body of the switch.

 

Long winded to describe, but a very neat solution as I have found.

[spikey]

Thinking further about this, if I were to swap my four solenoids for MTB motors (assuming of course that it's ever possible to buy some), how might I go about switching my pair of facing three-ways?  They form three storage loops, and at present the solenoids are activated by the CDU via a basic diode matrix and three push-buttons, one for each loop.  I can't see an obvious way of keeping things that simple ...