Frog Polarity switching: Relay or Microswitch ?

[DGO]

OK so I've decided to come back to model railways after a short 40 year gap, I've also decided I'm going to switch my turnouts with servos though I've yet to decide on exactly how. But then I come to controlling the polarity of the frog. So as I see it I can either switch the polarity using a relay at the same time as I send the command to the servo (or use sensors to tell the system the servo is moving and change)  or I can stick one or two micro switches on the other end of the servo arm to the turnout and use that to switch the polarity, Are there any clear advantages or disadvantages to either method ?

 

Currently my thinking is that I'd use a pair of microswitches to control the feed, that way whilst the turnout is actually moving the frog will be isolated completely 

[Nigelcliffe]

First question is about your turnout frogs.   Are they fully isolated from the blades, or are they still connected (as per at least some of Peco's products).    If fully isolated, then the chosen method doesn't matter.

 

If "frog live to blade", and you can't easily modify the turnouts, then you need to switch the frog with timing that doesn't conflict with the blades.  Your two microswitch solution could do this if setup correctly. 

 

If "fully isolated, doesn't matter", then a single microswitch is all you need, or a single pole change over relay working off the same switch as the servo controller, or (if the control switch is conveniently local, could use a spare set of poles on the operating switch (eg. a DP switch, one pole for frog, one to operate the servo driver).    Timing of frog change is irrelevant.

 

You could buy/build a servo driver which changes a relay at the appropriate time in servo movement.   DIY servo drivers are pretty simple to build if into basic electronics.    

 

 

( And option if you're running DCC trains, are automatic frog switching electronics ). 

 

 

- Nigel

 

[ikcdab]

On my test track I used microswitches. I had to des8gn and make a more complex servo mount to accommodate the switch and adjusting it was tricky. On some occasions I find the microswitches "bounce back" the servo. I used megapoints drivers which did make the setting of the switches easier.

Learning from this, on my forthcoming opus I am using relays. The servo mount is far simpler and there is no adjustment to worry about and no "bounce back". I have made up switch panels which contain  the relays and all the wiring which I can just plug in when I am ready. I cannot see any disadvantages whatsoever with relays and I would not use microswitches again. 

I attach a photo of one of my switch panels. The yellow leads are the power to/from the crossings and the red leads are to trigger the servos.

[nick_bastable]

Prefer relays given you can get built boards of that auction site 8 channel 5v  or  12v   negative or positive switching  for about £6 less hassle easy to hookup and robust however I'm working in 2mm with fully isolated frogs  I would suggest investigate MERG 

 

Nick B

Edited July 20, 2020 by nick_bastable

[DCB]

I use micro switches on DC but only outside, point blade contact has been fine inside for the last 35 years or so.   DCC point blade contact doesn't last long so you probably need relays to cope with the power. My micro switches are only rated at 1 amp the same as my controller.

[Andy Reichert]

Stall motors usually include frog switching contacts.  I have a high tech engineering background, but (or so) I don't see the point of using (wasting) a wide angle analog positioning device to move point blades between two fixed points for pretty much the same money and more complexity.

 

Andy

[Chris M]

I use Unifrog points and DC. I went for solenoid motors and latching relays and have found this to be cost effective and reliable (so far).  If you must use servos then you will need something to control the servo. There are a number of electronic switching devices from companies such as Megapoint that will provide electronic switching. If you are running DC the use of servos to change points seems to be a good way of increasing both cost and complexity. 

 

This is how I wire mine, its probably the cheapest solution to powered points:

 

 

[Chris M]

On one project I am involved in we installed servos with the MERG c-bus to control the points. Its a great idea in theory. After a couple of years of trying to make this set up work reliably we are about to strip it all out and fit solenoid point motors with traditional control. Sorry to say this but I really can't recommend using the plastic MERG servo mounts with micro switches for the frog power. The plastic mounts were not robust and we couldn't get the the servo to move both the point and the microswitch consistently. Also older Farish locos would somehow interfere with the system and create "phantom" point changes. I don't know whether the same problem occurs in 00. I know the MERG system has been successfully used by some but it has been a disaster for us.

[Godders]

3 hours ago, DavidCBroad said:

I use micro switches on DC but only outside, point blade contact has been fine inside for the last 35 years or so.   DCC point blade contact doesn't last long so you probably need relays to cope with the power. My micro switches are only rated at 1 amp the same as my controller.

Hi David, 

your statement is puzzling. If I read you correctly, you are saying a dcc loco uses more power than a dc loco and therefore needs micro switches. However, you only use 1A ratings. Please explain.

 

My friend, who has a large US layout, uses the switch blades only, to my disdain, and has done for 20 years of dcc operation, without problems. The layout is 30ft x 10ft and only has about 6 feeds. It was converted from dc to dcc by simply changing the controller.

 

Thank you

 

Godders

[kevinlms]

12 hours ago, DavidCBroad said:

DCC point blade contact doesn't last long so you probably need relays to cope with the power.

 

Why would anyone wire a DCC layout in such as way that the entire current, for the whole layout relies on the blade contacts?

 

That's the idea of a bus, so that current is only locally from the nearest dropper.

[DCB]

23 hours ago, Godders said:

Hi David, 

your statement is puzzling. If I read you correctly, you are saying a dcc loco uses more power than a dc loco and therefore needs micro switches. However, you only use 1A ratings. Please explain.

 

My friend, who has a large US layout, uses the switch blades only, to my disdain, and has done for 20 years of dcc operation, without problems. The layout is 30ft x 10ft and only has about 6 feeds. It was converted from dc to dcc by simply changing the controller.

 

Thank you

 

Godders

Yes, and no, A DCC lash up can draw a lot more power than a standard 12 va DC set up can provide.  My layouts are powered by H&M and Morley 12 volt nominal 1 amp power units and I have no trouble worth mentioning with blade contact. some points on the lower levels are buried under other parts of the layout pretty much inaccessible and haven't seen the light of day for well over 10 years.  All the points are peco and have the little tags on the point blades which make contact under the running rails . I had a few without the tags and they were useless.  A lash up, 4 locos can easily take 2 amps and the standard fish plates and point blade contacts can't take that sort of power.  I know that from using two 12 VA controllers in parallel  to run 4 loco lash ups on a DC layout. I had fish plates glowing red hot.  Several DCC power units are rated at 4 amps or more.   If you want to run 4/5/6 loco lash ups a la USA then you need droppers, point polarity switches etc.  If you are happy with the 1/2/3 locos per train a standard DC set up can provide then you don't need droppers polarity switches etc with Code 100 etc track. If you use DCC with 1/2/3 locos per power district Code 100 etc  and limit the amps by using polywswitches or other 1 amp overload cut outs on the power district feeds you don't need droppers polarity switches etc  But if you use scale track with non conductive fish plates you need droppers, potentially hundreds  of them and they are very very tedious to wire up.   Even without 4/5/6 locos at a time sooner or later an engine gets across a rail break and causes a partial short, usually by running into the frog end of a wrongly set point. On 12va  DC the trains run slowly on DCC it may or may not trip, if it does not trip that's when the damage is done to the point blade contact.   Outside we had little trouble with point blade contact using a 12 volt 1/2 amp controller for power but when we used a 12 volt 1 amp problems quickly appeared, again partial shorts due to bridging insulators caused the contacts to arc and oxidise, 1 amp microswitches in parallel with the point blade switching fixed it but with 12 month life per microswitch.  Current fix on a totally exposed point outside  two sealed reed relays operated by a magnet on a wire in tube operated by pushing the blade. Only did it Sunday so watch this space and I report on its success in 20 years time,

[WIMorrison]

I think that the term you meant to use is 'consist' and not 'lash up' which has a rather different connotation.

[Izzy]

On 23/07/2020 at 06:39, Chris M said:

On one project I am involved in we installed servos with the MERG c-bus to control the points. Its a great idea in theory. After a couple of years of trying to make this set up work reliably we are about to strip it all out and fit solenoid point motors with traditional control. Sorry to say this but I really can't recommend using the plastic MERG servo mounts with micro switches for the frog power. The plastic mounts were not robust and we couldn't get the the servo to move both the point and the microswitch consistently. Also older Farish locos would somehow interfere with the system and create "phantom" point changes. I don't know whether the same problem occurs in 00. I know the MERG system has been successfully used by some but it has been a disaster for us.

I have come to understand that there is no real problem with radio controlled servos per se, but that trying to adapt them for model railway use can cause issues that can arise with them becoming an obstacle to usage, depending on the particular layout setup and where the layout is situated, where it is erected for use at any particular time. 
 

The problem is RFI/EMI (radio frequency/electro magnetic interference). It is this that is the issue of the older Farish locos disrupting the servo signals. This disruption can arise from many sources, mine went haywire whenever my DCC system was switched on, it didn’t even have to be plugged into the layout! Needless to say I dumped servo control as such, managed to sell off the MERG servo 4 boards - as you say a lot of people have no problems at all - and now use ‘ hacked’ servos. They just run off 1.2 -1.5v DC via DPDT centre off switches. 
 

I did use an alternative design which I posted in the 2mm section which might suit your club. https://www.rmweb.co.uk/community/index.php?/topic/112916-hacked-servo-point-control/

 

Anything is better than sledgehammer solenoid point motor control.

 

Izzy

Edited July 24, 2020 by Izzy

[Stubby47]

I use merg  powered servos, but with a DPDT switch - one half is the on/off for the servo, the other half is the frog feed.

 

No micro switches or relays required.

[WIMorrison]

How do you bc avoid the shirt that appears when the switch pole has changed but the servo delay hasn’t yet moved the turnout blades?

[Stubby47]

Using DC , so there is no power to the track.

[ikcdab]

On 23/07/2020 at 06:28, Chris M said:

DC the use of servos to change points seems to be a good way of increasing both cost and complexity. 

Not sure why you say this.  A peco point motor retails around £7. A 9g servo (hobbyking hk15168) is £2.29. If you use MERG drivers, these are £5 to power four servos (servo4 kit). So servo drive is around £4 plus a few pence for aluminium channel mount. Switches are the same in both cases, though with the servo just needing a simple on/off switch rather than a momentary contact it is simpler and easier. If you need to add a relays or mixroswitch than that added another £1 to the servo option, but even so it's less than the cost of a peco solenoid. And of course, with the servo, you avoid the horrible "thump" of solenoids and impart far less wear and tear on the tiebar.

[kevinlms]

32 minutes ago, WIMorrison said:

How do you bc avoid the shirt that appears when the switch pole has changed but the servo delay hasn’t yet moved the turnout blades?

Shirt? Isn't autocorrect wonderful!

[WIMorrison]

1 hour ago, Stubby47 said:

Using DC , so there is no power to the track.

Really? How do the locos move then?

[Stubby47]

8 minutes ago, WIMorrison said:

Really? How do the locos move then?

 

Not when the points are being switched.

 

Also, the track plan is an inglenook, so there are no isolating breaks required at the toe-end, and it's a one-loco-in-steam operation.

 

Edited July 24, 2020 by Stubby47

[WIMorrison]

A rather serious limit to operation if you have to stop the trains every time you want to change the direction of a turnout.

 

The solution of a DPDT switch doesn’t seem a viable operating method to me given this severe limitation in use.

[Stubby47]

52 minutes ago, WIMorrison said:

A rather serious limit to operation if you have to stop the trains every time you want to change the direction of a turnout.

 

The solution of a DPDT switch doesn’t seem a viable operating method to me given this severe limitation in use.

 

As I said, it's an inglenook, so the trains will stop before each point change as they will also change direction.

 

The DPDT switches do work, I'm surprised you seem so negative.

[WIMorrison]

I am disparaging if your solution as it will only work in your specific modus operandi and it will lead to faults for other people who do not stop the layout when throwing turnouts.

 

These threads are mostly read by beginners looking for advice and the solution you suggested does not sufficiently clarify the highly specific use case required to ensure that it does not present operational problems.

[Izzy]

4 hours ago, WIMorrison said:

How do you bc avoid the shirt that appears when the switch pole has changed but the servo delay hasn’t yet moved the turnout blades?

Why will there be a short? I’m afraid I don’t understand how that would occur whether it’s DC or DCC. Or is this specifically to do with peco RTL pointwork rather than generally?
 

Izzy

[WIMorrison]

The short happens because when you throe the DPDT switch the blade is still in contact with the stock rail thus at one polarity, but the frog polarity changes when the DPDT switch has been changed but the delay in the servo moving the point blades causes a track short until the blade has left the stock rail.