Frogjuicer(Autofrog) or self-latching relay?

[leer76]

Hi All,

I have a DCC layout, all electrofrog with seep PM1 point motors. I find the "switching" can be unreliable on the seeps so I added autofrogs which work well with no problems, however I need to purchase some more.

While looking online I came across self latching relays which to me seem to do the same thing but at a sligtly lower price "

My question is what is the difference between an autofrog and a self latching relay?

I have looked online but can't seem to find a definative answer

[TT-Man]

I am not sure if the terms are used consistently but there are basically two different types of mechanism. The first is mechanical and so relies on a moving part to effect the switchover. The second is electronic. The latter having no moving parts is more reliable but tends to be much more expensive.

[dBerriff]

An autofrog senses any change in polarity as (usually) a locomotive wheel touches the frog (a momentary short, effectively) and quickly switches the feed polarity to match. For speed I assume it uses electronic switching rather than a mechanical relay.

A relay is simply an electromechanical or electronic device that uses a low-power signal to switch a high-power supply. It needs to be controlled by a switching voltage. This could be through a mechanical switch or from an electronic microcontroller.

Edit: some relays are self-latching in that they do not require a holding current once switched to the energised position. Simple electromechanical relays would require a small current to hold the switch blade against a mechanical spring.

https://uk.rs-online.com/web/content/discovery/ideas-and-advice/latching-relays-guide

[ColinB]

I think you will find that they are the same thing. As the loco passes over the point it momentarily creates a short if the frog is powered wrongly and triggers a self latching relay. The relay has to be self latching as once the frog has been powered correctly, if it were an ordinary one it would return to its original state. So I think it is just what people call them. I don't know if there is anything else special. I have them on my layout an easy solution that works even when you switch the points manually.

[ntpntpntp]

Frankly I refer using a changeover switch mechanically linked to the point motor and/or tiebar, I don't like the concept of detecting short circuits to trigger the changeover, no matter how rapidly an electronic circuit can handle it. It seems a poor engineering solution to the problem, and also a simple mechanical switch will work on DC as well as DCC. I'll admit this comes from my use of live frogs in G scale with a few amps flowing through the track :)

Of the two designs of auto frog polarity switch I would prefer the electronic circuit over the relay-based design.

Unfortunately I too have seen poor reliability of the SEEP changeover switches back in the 80s with N gauge pointwork. I ended up adding a microswitch, but for subsequent layouts I reverted to using Peco point motors and their accessory switches.

[dBerriff]

Cost-per-point is an issue for more complex layouts. The traditional solenoid with attached switch has some long-term reliability issues but it is (relatively) affordable and simple to install. Some motor-driven units include switching but are rather expensive.

On the large layout I am helping with we are moving over to servo-driven points. With a microcontroller in the loop (PIC, Arduino or Pi Pico) you can time the relay switching to the mid-point in the switch-over and avoid shorts. A single device can control, typically, 8 servos and 8 relays and so is quite cost-effective. But this is a whole other discussion so I will stop at this point, merely pointing out that there are options. Little Wicket railway on YouTube got me interested in the servo-driven option.

[P-Henny]

There are two types of latching relay products for model railways.

Guagemaster make a DCC80 which uses a latching relay, but detects short circuits to perform the switching. These operate in a similar way to the all electronic no moving parts frog juicer but use a very simple circuit design based on capacitors, diodes and resistors to switch the mechanical relay and operates completely independently from whatever solution is used to physically move the point blades. Being mechanical, the switching latency is slower than an all electronic frog juicer and could theoretically switch more slowly than the 'short circuit' detection in some DCC controllers, but this in my experience is rare.

Then there is what I call the generic latching relay, that uses a latching relay that is switched by the same current pulse that is sent to operate the solenoid point motor. These latching relays are wired across the solenoid point motor in parallel, the weakness of this solution is that with some scenarios, the latching relay saps all the current and prevents the solenoid motor from operating robustly. This latching relay product has the advantage that it will work just as well on DC powered track as on DCC powered track. Gaugemaster make these too, as GM500 and GM500D. The D version is supposed to be more DCC friendly than the non D version, but I have seen posts on this forum that reported that the D version in some instances didn't work reliably on DCC when the non D version meant for DC Analogue did.

Personally, I go with the "you get what you pay for" approach. My DCC Electrofrog points use the all electronic frog juicer product made by TamValley. As well as a switch for electofrogs, it can be configured by a link to function as a DCC Reverse Loop Module. I am using these TamValley juicers for both functions, frogs & RLM. The very fast switching of an all electronic no moving parts frog juicer is more likely to put less strain on the DCC controller output circuitry.

P.S. This is clearly a DCC question and should have been posted in the Hornby DCC forum.

[Brew Man]

I use the Gaugemaster DCC80 on all of my Electrofrog points and have never had a problem with them.

[ColinB]

It is a bit off subject but I have tried very different ways to power the frog. The Autofrog to me seems the best compromise for ease of fitting and cost as long as it is fitted properly as per instructions. I burnt one out when making the detected section too long. I have tried the Peco slider switches, I found that these put an undue load on the point motor and were very unreliable. The next idea I came up with was the computer microswitch that worked on the pin of the point I even had some 3D printed adapters made so they fitted better. Electrically very reliable but a pain to setup and they didn't always work properly due to the mechanical interface. Next one was to use the blade of the point to drive a relay, worked perfectly on DCC but because of the dummy pony truck on some Hornby models having super wide wheels, on some locos caused shorts going over the points. The effect of the dummy wheels was to create the same effect and switch the frog back. Next one I tried was the electronics to electronically latch on the switch, again it would have worked but if you switch the points manually the frog doesn't get energised. So as I said best compromise, yes it detects a short for a short period of time but if you properly insulate the frog as per instructions not an issue. Electrically for the milliseconds it takes to detect the short it is not going to be an issue. I have the best part of 60 points so the servo controlled ones was never a real option.

[ColinB]

Thank you for the information about TamValley I will have a look at those. From what you say they might be more useful than the ones I presently use.

[ntpntpntp]

I have tried the Peco slider switches, I found that these put an undue load on the point motor and were very unreliable.

 

 

@ColinB I have to say in my experience they've been extremely reliable on my N gauge exhibition layout over the past 25+ years and dozens of shows, despite the limited throw of N points. No problems with load on the motor, I've even run two sliders on some motors. I much prefer them to Peco's microswitches which apply "back pressure" to the point motor rod.

I did find a problem on my latest project with certain newer design N points having even less throw then the older ones, but I created a 3D printed mount with a lever to increase the throw to the switch.

[leer76]

Thanks everyone, some interesting reading here.

Apologies for posting in wrong forum, if someone wants to move it feel free.

It is actually DCC80's i already have and I was looking at the GM500's, I think I'll stick with what i know so I'll get some more DCC80's

Once again thanks everyone for your invaluable advice.

[P-Henny]

"Apologies for posting in wrong forum, if someone wants to move it feel free."

Leer, Only a Mod can move your thread, normal members cannot move a thread you own.

EDIT: Post edited to correct incorrect advice given.

[leer76]

Paul, I have followed your advice and tried to edit my very first post on this subject via the 3 dots, however it seems I am only able to change the title and i cannot see an option to locate the post in the correct forum, slightly annoying as like you say would it have been a learning exercise, maybe if a mod see's this they will move if for me

[P-Henny]

Leer,

Sorry, my mistake I thought you could. I must have been getting mixed up with the screen layout view you get when creating a new thread.

[leer76]

I think I messed up? I've bought a seep pm2 instead of a pm1. Will this be compatible with a dcc80 autofrog?

[leer76]

I think I messed up? I've bought a seep pm2 instead of a pm1. Will this be compatible with a dcc80 autofrog?

[P-Henny]

The DCC80 does not physically connect to any point motor that you use. The DCC80 operates completely independantly of ALL forms of point motor. Therefore it doesn't matter whether you have a PM1 or a PM2.

The PM1 has an integrated switch, the PM2 does not. But the PM1 switch is not used for frog polarity if you are using a DCC80.

The DCC80 has three wire connections. One of which connects to the point frog. The other two wire connections connect to the track rails. None of the three wires need to touch the point motor or its wiring. Therefore the point motor that is installed is immaterial.

If you have been wiring your existing PM1 integrated switches to your existing DCC80s, then you have wired them incorrectly.

[leer76]

So all my pm2's are wired thus:

Terminals A,B and C lead to a TTPC200 decoder

Terminals D and E lead to DCC bus

Terminal F to the frog wire

Some of these points/turnouts started to fail (as in short) and so i blamed it on the unreliability of the switching on PM1's.

To remedy this I purchased some DCC80's, I removed the wire to terminal F on the Seep and instead wired this to the DCC80 autofrog, the power for it comes from the DCC bus

This seemed to cure the problem.

Now I have found myself needing another point/turnout but ordered a PM2 point motor instead of a PM1, As i knew it wouldn't switch polarity I asked what is the difference between an autofrog and a self latching relay? and after reading the comments I went with what I knew and choose DCC80's but it dosen't look like i can wire to PM2 as not quite the same thing

[P-Henny]

On the PM1 you correctly disconnected the wire from terminal F and then redirected that wire to the frog terminal on the DCC80.

You left the wires on terminals D & E, fine that is not an issue, but you could have removed those wires too as they now serve absolutely no function since the F terminal is not now connected to anything.

On the PM2 there is no integrated switch so terminals D, E & F again serve no function (that is if the PM2 actually had those three terminals).

As I said before, the frog wire on the DCC80 goes to the point frog and the other two DCC80 wires go to the DCC rails, which is exactly the same electrically as going to your DCC BUS as the BUS connects to your rails.

The DCC80 works and is wired exactly the same whether you have a PM1 OR a PM2.

YOU DON'T NEED THE SEEP D, E & F TERMINALS when using the DCC80.

If you take a PM1 but don't use the D, E & F terminals for anything, then it is in effect a PM2. So all your points that have DCC80's for frog switching are in essence wired as PM2's even if you bought them as PM1's.

Trust me, if you wire the new DCC80's the same as your previous ones they WILL work with your new PM2's.

[leer76]

Paul, once again you have struck gold

I have always assumed that +V and -V (i think thats right) was required to power the point motor, however from your reply i now know that it powers the frog.

Can i take it therefore that the power for the Seeps come from the decoder?

[P-Henny]

Can I take it therefore that the power for the Seeps come from the decoder?

 

 

Yes absolutely.

The A, B & C terminals that connect to your decoder are connected to the solenoid operating coils on the PM1 & PM2. It is the current pulse that is generated by the decoder that energise the coils via the A, B & C terminals that operate the point motor.

The D, E & F terminals on the PM1 have absolutely no role to play in actually operating the point motor. They are just a completely separate SPDT (Single Pole Double Throw) changeover switch that can be used for any optional function.

You previously used it for frog switching, but it could just have easily been used for example to operate a Dual Aspect lamp signal OR to provide a point postion indicator on a layout plan mimic panel (which in your case is a function provided by your Railmaster track plan display).

[leer76]

Paul, your input on this thread has been most valuable. Thankyou

[P-Henny]

thumbsup