Jump to content
 

Eastwood Town - A tribute to Gordon's modelling.


gordon s
 Share

Recommended Posts

Hi Guys

 

Gutted we lost our knockout match yesterday on the 73rd hole, so I need to follow my brother's advice "get over it". Sympathetic as always.....;)

 

Still been chasing around to understand Extended Route Control on the ECoS and one esteemed dealer has explained it will do AND, but isn't sophisticated enough to do OR, so drawn a blank.

 

Choices are manual control which will cost nothing, but seems a bit of a cop out or play around with a few relays and see what happens. I suspect the weather is starting to shift the balance between golf and ET, so things may well fire up again.

 

Just before I fire up my soldering iron, I need a couple of issues clarified...

 

1662747185_Screenshot2020-08-19at12_16_52.png.9434cf50498195897e37c4d6ba7b71cf.png

 

2108666620_Screenshot2020-08-19at12_16_38.png.1a9e5285b2077d52d945d6d3f48142a9.png

 

I assume the gate goes to the DCC Concepts LED driver board. Does it matter which way round the Source and Gate are connected? My logic says source to the relay coil and drain to system ground. Correct?

 

If the output from the LED driver board was 2.6v and that was sufficient to operate the coil of a 3v relay can I do away with the additional power supply and energise the coil directly? Just trying to keep the connections to the mimic panel to the bare minimum.

 

Anyway, I've ordered a few bits to play with just because it may fill a few hours without football/snooker/golf on the TV in the evening.....:D

 

Had an attack of the dreaded sh...ks yesterday, so off to the range to find the solution....

Link to post
Share on other sites

  • RMweb Gold
1 hour ago, gordon s said:

2108666620_Screenshot2020-08-19at12_16_38.png.1a9e5285b2077d52d945d6d3f48142a9.png

 

I assume the gate goes to the DCC Concepts LED driver board. Does it matter which way round the Source and Gate are connected? My logic says source to the relay coil and drain to system ground. Correct?

 

 

Hi Gordon,

 

Wrong. The intended connection is for the relay to provide the drain connection to the supply voltage. The source connection goes to ground:

 

gordon_mosfet.png.b7859b67db6f53ab8856402efaaeb9a8.png

 

Admittedly this is confusing for conventional electrical notation, but remember that in the real world electrons flow from negative to positive.

 

Quote

If the output from the LED driver board was 2.6v and that was sufficient to operate the coil of a 3v relay can I do away with the additional power supply and energise the coil directly? Just trying to keep the connections to the mimic panel to the bare minimum.

 

You can try it. The output may hold up at 2.6v under load, and the relay might pull in at 2.6v, or it might not. But to avoid damaging the LED driver board, you would need a quenching diode reverse-connected across the relay coil (or any other inductive load, i.e. anything containing a coil, such as a motor):

 

gordon_relay_quench.png.e047d7cad835e618daacfddb8e1156ac.png

 

p.s. the above assumes the DCC LED output is positive +2.6v. Make sure you get the diode the right way round, otherwise it will short-circuit the output.

 

The MOSFET driver has the advantage of having robust quenching diodes built-in, and the gate is insulated from any inductive effects.

 

cheers,

 

Martin.

Edited by martin_wynne
diagram added
  • Agree 1
  • Informative/Useful 2
Link to post
Share on other sites

Thanks Martin. You can guarantee if it's 50/50, I'll get it wrong......:D

 

Good job I asked.

 

I was planning to use the MOSFET with the relay as per your original diagram, but with the DCC module into the MOSFET and then to the relay coil. The other side of the coil going to system ground. Would that not work?

 

I wish I understood more about electronics.

 

Now ask me about the sh..ks and I'll talk for hours....

 

Hopefully 75 balls later I've understood a bit more. You can never say it's cured, simply kept at bay for a while.

 

 

 

Link to post
Share on other sites

  • RMweb Gold
27 minutes ago, gordon s said:

I was planning to use the MOSFET with the relay as per your original diagram, but with the DCC module into the MOSFET and then to the relay coil. The other side of the coil going to system ground. Would that not work?

 

 

You've lost me Gordon. With what voltage relay? And where would the MOSFET drain be connected? It needs to be connected somewhere, otherwise there would be no current through the relay. The MOSFET gate is in effect insulated from the source/drain through current.

 

Using MOSFETs it needs to be as in my diagram -- which corresponds to the MOSFET data sheet! It's only one extra wire from the power supply to the relays board for the drains, so I don't see the problem? The starting point is to find a regulated 5-volt or 12-volt power supply, so that you know what voltage relays and MOSFETS are needed. I'm sure you must have one somewhere amongst all that electronics stuff?

 

cheers,

 

Martin.

Link to post
Share on other sites

I was using a 3v relay, so that could be driven by the DCC Concepts output to the led's. I was going to take one output wire to the MOSFET and the other to system ground. One side of the relay coil would be connected to the MOSFET and the other side of the coil back to system ground.

 

From your post, that won't work....

 

If that is the case, I'll add the power supply....

Link to post
Share on other sites

  • RMweb Gold
12 minutes ago, gordon s said:

I was using a 3v relay, so that could be driven by the DCC Concepts output to the led's. I was going to take one output wire to the MOSFET and the other to system ground. One side of the relay coil would be connected to the MOSFET and the other side of the coil back to system ground.

 

From your post, that won't work....

 

If that is the case, I'll add the power supply....

 

Hi Gordon,

 

MOSFETs have 3 terminals -- you mentioned only 2?

 

Here's a 12V regulator for up to 2 amps. Any old DC input between 16 and 36 volts:

 

 https://uk.rs-online.com/web/p/switching-regulators/1934045/

 

2 amps should be enough for quite a few 12v relays. You might need one at each end of the mimic board.

 

cheers,

 

Martin.

Link to post
Share on other sites

  • RMweb Gold
3 minutes ago, gordon s said:

Sorry Martin. Output board to gate, drain to relay and source to ground. The other side of the coil back to ground....:D
 

 

er, no. That would provide no source of energy to drive the relay current round in circles.

 

The MOSFET senses the voltage on the gate, but can't take any current from it. That's what makes it so useful, because almost any old voltage on the gate can be used to switch the source-drain current on.

 

Martin.

  • Informative/Useful 1
Link to post
Share on other sites

Right, I have some MOSFETS and 5v relays plus a 5v power supply. Using this diagram with the 12v changed to 5v, I understand this will now trigger the relays from the route switching in the ECoS via the DCC Concepts mimic board.

 

Can you now show me on your diagram where the connection goes to the first relay changeover contact?  Am I right in saying it is tapped off between the DCC Concepts output and the MOSFET? 

 

Something like this? I know I've only shown one relay whereas they will be daisy chained and the last one will go to the LED.

 

DSCF1228.jpg.3b681a2ffdd2c31eddf17dbaa80ef598.jpg

 

I'm interested in doing this as a stand alone project to help me understand relay applications and get a better grip of electronics. We all have to start somewhere....

 

Edited by gordon s
Link to post
Share on other sites

  • RMweb Gold

Hi Gordon,

 

Is that the platform LED?

 

If yes the contact connections don't go anywhere near the DCC module output. If it's one of the routing LEDs, what are the contacts for?

 

Do your LEDs have an integral resistor? If yes, for what voltage? If no, what voltage are you planning for the platform LEDs?

 

I will draw up a diagram shortly. :)

 

Right, I have some MOSFETS and 5v relays plus a 5v power supply.

 

That was quick! Do you live next door to RS? Are they 5v MOSFETS?

 

cheers,

 

Martin.

Edited by martin_wynne
Link to post
Share on other sites

Yes that was supposed to be one of the platform led’s. As you probably guess I’m a bit confused with the led connections to both the designated turnout led’s driven by the DCC Concepts output board and the platform led’s which will sit at the end of the relay chain.

 

The led’s I’m using are here....

 

https://www.rapidonline.com/kingbright-l-7104pbc-a-3mm-blue-led-1000mcd-55-1476

 

They work perfectly directly from my Gaugemaster controller set at 2.6v to simulate the DCC Concepts output. Right now each DCC Concepts connector is attached to their larger led’s, so I planned to cut those off and resolder to the 3mm led’s once I’ve experimented a bit to see if the relays will do exactly what I want.

 

I’m not doubting what you have suggested, just want to build something and have the pleasure of setting routes and see it work in practice. I just want to play a bit and see what happens.

Link to post
Share on other sites

  • RMweb Gold

Hi Gordon,

 

I've scribbled on your diagram:

 

gordon_relay_circuit.png.851d6c73886e26426d8db305b8148f87.png

 

The connection at X assumes you want to use the same power supply for both relays and platform LEDs (you don't have to).

 

p.s. those are the relay contacts corresponding to each turnout.

 

I will draw up a better diagram. But it will take me a while -- as with the sketchboard upgrades, I'm struggling to keep up with your pace of building stuff. :)

 

cheers,

 

Martin.

Edited by martin_wynne
  • Funny 1
Link to post
Share on other sites

  • RMweb Gold

Hi Gordon,

 

I've been looking at the DCC Concepts data sheet for the Mimic LED module.

 

I can't see if it's safe to common the outputs to an external ground, and if so, which of the 2 wires is the common ground.

 

It probably is, but If not, you will also need an optocoupler on each output between the module and the MOSFET:

 

 https://uk.rs-online.com/web/p/optocoupler-ics/3956344/

 

I found you a bit of video:

 

 

It may be possible to use only an optocoupler driving the relay instead of the MOSFET, but not so rugged.

 

Martin.

Edited by martin_wynne
Link to post
Share on other sites

  • RMweb Premium

From a Control Engineers point of view this is getting far too complicated..relays? MOSFETS?

Rule 1

 Keep it Simple, Stupid or KISS..

 

There must be too much wood to see the trees..

 

Sorry Gordon. No need for relays or MOSFETS....

 

Baz

  • Agree 2
Link to post
Share on other sites

  • RMweb Gold
29 minutes ago, Barry O said:

There must be too much wood to see the trees..

Sorry Gordon. No need for relays or MOSFETS....

 

 

I agree, but what is your suggested alternative?

 

It could all be done with some dedicated logic chips, or by programming a Raspberry Pi, or with levers and rods, or pneumatic valves, or some lengths of string and falling weights.

 

It could also be done very easily by using the extra contacts on the Tortoise motors, but that involves making the correct logical connections upside down under the baseboards, and some long wires.

 

But Gordon has invested in a lot of expensive DCC stuff which he wants to use, and as it stands it is not giving him what he wants.

 

Martin.

Edited by martin_wynne
Link to post
Share on other sites

This was just going to be a bit of fun over the weekend, but it would appear it’s got more complicated again. The ECoS is doing what I want perfectly in terms of route selection, so no problem there.  I’m probably asking it do something outside it’s capabilities, so I have no problem if it can’t be done.

 

I have a manual option that works perfectly well and I’m positive that once ET is running well and I start moving into signalling, it will be easy to incorporate.
 

Thanks for your help Martin. None of it is a problem as it was only curiosity got the better if me in the absence of a programmed solution.

 

A decent weather forecast tomorrow and 18 holes await at Minchinhampton for the first leg of an inter club challenge between good friends. I will forget all about led’s and relays and will enjoy a decent walk in the Cotswold countryside.
 

There may even be a few beers and a curry afterwards.....My sort of day...:drink_mini:

 

 

  • Friendly/supportive 1
Link to post
Share on other sites

  • RMweb Gold
1 minute ago, gordon s said:

This was just going to be a bit of fun over the weekend, but it would appear it’s got more complicated again. The ECoS is doing what I want perfectly in terms of route selection, so no problem there.  I’m probably asking it do something outside it’s capabilities, so I have no problem if it can’t be done.

 

 

Hi Gordon,

 

I'm pretty sure that if I had the actual gear in front of me, I could rig up a system which would be simple to build and inexpensive. But explaining everything at arm's length and not having all the information makes for lots of questions and makes it look far more complex than it really is.

 

Enjoy your golf.

 

Martin.

  • Thanks 1
Link to post
Share on other sites

  • RMweb Gold

Hi Gordon,

 

Do you think you could manage to get just one wire back to the panel from each turnout?

 

If you could, it would be possible to operate a relay simply by connecting it directly to the V-crossing, like this:

 

gordon_simple.png.66a3f9c5ee5b5b00513127fd693455c2.png

 

No electronics, no MOSFETs, no DCC module connections, no power supply needed, nothing to go wrong.

 

No need to work out the logic connections under the baseboard, just add a wire to the existing vee dropper.

 

This assumes 2 things:

 

1. you are switching the v-crossing polarity directly, using the Tortoise motor contacts or some other means (i.e. not frog juicers), and

 

2. the DCC power connection to the track is normally always on, and not switched off to isolate sections during running sessions.

 

DCC power is a 18v high frequency square-wave AC, so a diode will be needed for a cheap DC relay. The relays might make a bit of a buzz when on, or might not, or you could put a cardboard cover over each one to quieten it. It doesn't matter which DCC bus wire you use. The relays can all be on the same one. It's just a case of identifying with your tester which way the relay contacts are lying with respect to each turnout, i.e. relay normally off or on.

 

p.s. I'm fairly confident a 12v DC relay will pull in reliably on half-wave DCC power (check with several trains running or other load on the system), but if not you could use 5v DC relays instead. Check that they don't get too warm, might need to add a series resistor. Sorry I don't have any DCC stuff here to test or measure.

 

cheers,

 

Martin.

 

Edited by martin_wynne
Link to post
Share on other sites

  • RMweb Gold
15 minutes ago, hayfield said:

 

I thought it was just me, still Gordon will start playing trains soon, might even get some track building going

 

For those mystified by electrons shuffling to and fro in wires, you can if you prefer use string instead:

 

 https://templot.com/info/wooden_point_motors_1956.pdf

 

Sorry if I have posted this before.

 

p.s. this is a serious post, it is actually a very clever idea. The modular lever frame is an obvious candidate for 3D filament printing.

 

Martin.

  • Like 1
  • Informative/Useful 1
  • Funny 1
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...