Feedback from reverse loop

[Airbuspilot]

I am helping a friend with his extensive G scale garden layout and we are looking at the possibility of automating a section. The idea will be to use a circular area with a diagonal for direction change. We are using ZIMO MX 10 as the base station and iTrain for the automation. On my own N scale layout I am using Digikeijs DR 4108 as my current feedback, this is too small for G scale and I would be interested in opinions for a suitable equivalent for G Scale.

Using current feedback is not an issue for the circular portion of the track as it will have a common rail and insulated feedback block rails. However I am not sure how to deal with the diagonal. This section has a reverse loop sensor which is powered, but not necessarily drawing current. I am not sure if a current sensor will work on this section or if it might interfear with the reverse loop electronics? If anyone has dealt with this situation i would appreciate your views.

If current sensing is not possible can you suggest an alternative?

This diagram is only an illustration in reality there will be two diagonals.

Robin

 

[Rhinochugger]

I am helping a friend with his extensive G scale garden layout and we are looking at the possibility of automating a section. The idea will be to use a circular area with a diagonal for direction change. We are using ZIMO MX 10 as the base station and iTrain for the automation. On my own N scale layout I am using Digikeijs DR 4108 as my current feedback, this is too small for G scale and I would be interested in opinions for a suitable equivalent for G Scale.

Using current feedback is not an issue for the circular portion of the track as it will have a common rail and insulated feedback block rails. However I am not sure how to deal with the diagonal. This section has a reverse loop sensor which is powered, but not necessarily drawing current. I am not sure if a current sensor will work on this section or if it might interfear with the reverse loop electronics? If anyone has dealt with this situation i would appreciate your views.

If current sensing is not possible can you suggest an alternative?

This diagram is only an illustration in reality there will be two diagonals.

Robin

Personally, I'm not a lover of that sort of layout - operationally, the problem is that once you've gone down the diagonal, you've done it, and can only use it again in reverse.

It's just one of my pet bete-noirs

However, more importantly, there are some things that don't work well in G scale, and I'm not sure if feedback works OK because of some of the on-board electronics. I could be wrong on that one, the memory can play tricks.

 

[Airbuspilot]

Personally, I'm not a lover of that sort of layout - operationally, the problem is that once you've gone down the diagonal, you've done it, and can only use it again in reverse.

It's just one of my pet bete-noirs

However, more importantly, there are some things that don't work well in G scale, and I'm not sure if feedback works OK because of some of the on-board electronics. I could be wrong on that one, the memory can play tricks.

Thanks, in fact there are two diagonals to overcome that problem. The diagram is purely an illustration, my question is how to feedback from a reverse loop rather than a "straight" piece of track. ZIMO produce a device specifically to allow feedback so presumably it should automate reasonably well, but i can only find information on "conventional feedback wiring" nothing from a reverse loop.
Robin

 

[PhilP]

If it is not specifically mentioned in any of the documentation, then I believe it should 'just work'?

Normally, if there is a 'feature', then it is specifically mentioned, as a 'don't try this'.

Of course, I could be totally wrong about this, but the current detection should not be polarity sensitive, with DCC.

PhilP

 

[Airbuspilot]

Thanks Phil, I agree the current detection should not be polarity sensitive however I wonder if it is a current drain causing the reverse loop to operate, possibly even cycle. Frankly I have no idea which is why I thought someone on this forum might have an answer.
Robin

 

[phils2um]

Hi Robin,

You might consider what I did using Azatrax (IR model railroad train detectors, infrared proximity sensors) paired IR transmitter/receiver detectors. I'm using two 8-detector modules feeding standard Märklin s88 modules. I'm certain Zimo must make a feedback module that will also accept input from the Azatrax IR unit. It only closes (or opens) an isolated set of contacts when the IR beam is broken. Here's a link how I set up the IR detector pairs:IR Train Sensors.

The sensors have been working great so far. I may have some issues in the winter with snow but I don't think that will be a problem for you in Cyprus! Here are a few more pics. I don't think the sensor pairs look out of place at all. (I really need to touch up those scratches on that baggage wagon!)

 

[Greg Elmassian]

You are looking for a current sensor that can work with DCC?

Have you looked at Digitrax?

Have you looked at NCE?

I do realize you are not in the USA.

Greg

 

[dunnyrail]

I am helping a friend with his extensive G scale garden layout and we are looking at the possibility of automating a section. The idea will be to use a circular area with a diagonal for direction change. We are using ZIMO MX 10 as the base station and iTrain for the automation. On my own N scale layout I am using Digikeijs DR 4108 as my current feedback, this is too small for G scale and I would be interested in opinions for a suitable equivalent for G Scale.

Using current feedback is not an issue for the circular portion of the track as it will have a common rail and insulated feedback block rails. However I am not sure how to deal with the diagonal. This section has a reverse loop sensor which is powered, but not necessarily drawing current. I am not sure if a current sensor will work on this section or if it might interfear with the reverse loop electronics? If anyone has dealt with this situation i would appreciate your views.

If current sensing is not possible can you suggest an alternative?

This diagram is only an illustration in reality there will be two diagonals.

Robin

Another option would be to mount a micro switch that the treadle can be activated by a train passing over, this can then operate a relay to do the necessary. Alternatively an LGB epl reed switch contact could be used to throw an LGB point motor with an accessory switch achieving the required result but in this case each loco would need a magnet underneath to throw the epl reed.

 

[Airbuspilot]

Hi Robin,

You might consider what I did using Azatrax (IR model railroad train detectors, infrared proximity sensors) paired IR transmitter/receiver detectors. I'm using two 8-detector modules feeding standard Märklin s88 modules. I'm certain Zimo must make a feedback module that will also accept input from the Azatrax IR unit. It only closes (or opens) an isolated set of contacts when the IR beam is broken. Here's a link how I set up the IR detector pairs:IR Train Sensors.

The sensors have been working great so far. I may have some issues in the winter with snow but I don't think that will be a problem for you in Cyprus! Here are a few more pics. I don't think the sensor pairs look out of place at all. (I really need to touch up those scratches on that baggage wagon!)

View attachment 307220
View attachment 307221


View attachment 307222

Thanks Phil.
I looked at IR sensors some time ago when I tried to automate my own layout using Arduino's, I found then to be too sensitive to light changes and finally settled for Hall sensors and magnets. You'r set up looks to be much better thought out than my initial try, If I understand you are breaking an IR beam which would solve the light variation problem somewhat. I would prefer to go the current sensing way if possible as it is essentially plug and play.
Robin

 

[Airbuspilot]

You are looking for a current sensor that can work with DCC?

Have you looked at Digitrax?

Have you looked at NCE?

I do realize you are not in the USA.

Greg

Hi Greg
I am aware of both but not familiar with them, my layout is N scale so my sensor has low amperage I need something for G scale hence my question. However if I cannot make a current sensor work with a reverse unit then the project becomes more complicated and will probably be dropped for the medium term. I will look at both to see what, if anything, they say bout current sensing from the reverse unit (PSX-AR) section of track.
Robin

 

[Diesel2000]

Current sensing feedbacks on an outdoor layout become problematic due to the wiring, and lots of it. I set up my layout for iTrain automation using Massoth digital track contacts and loco magnets into both Marklin S88 feedback modules and Massoth feedback modules. The track contacts draw power from the rails and will need to be wired directly to the feedback modules, but this was far easier to accomplish than to use current sensing modules in my case.

I have a reverse section thats controlled by a PSX-AR module and has the track contacts on it, and it works very well. You'll need the feedback module from whatever brand CS you are using, but the track contacts will work with any feedback of the 'generator' type. They are hall sensors, so same concept as a reed switch.

 

[idlemarvel]

...I would be interested in opinions for a suitable equivalent for G Scale.

Using current feedback is not an issue for the circular portion of the track as it will have a common rail and insulated feedback block rails. However I am not sure how to deal with the diagonal. This section has a reverse loop sensor which is powered, but not necessarily drawing current. I am not sure if a current sensor will work on this section or if it might interfear with the reverse loop electronics? If anyone has dealt with this situation i would appreciate your views.

I have used Lenz current detection sensor on a section of track controlled by a Lenz reverse loop module with no issues. The reverse loop manual has a diagram of how to wire it for such an eventuality. This may be Lenz specific but thinking about how such devices work in general I can't see why they wouldn't. I guess it would depend on the timing sensitivity of the current detection compared to the speed of the reverse loop current switchover. I'll see if I can dig out that manual with the wiring diagram.

 

[idlemarvel]

Here is the diagram or diagrams I was referring to. The first shows how the reverse loop module is connected, the second how to connect it if you want to monitor the loop for occupancy.



The text "To the amplifier" means the command station.



In fact I used an LDT RS-8 current based occupancy detector rather than the Lenz LB100/LB101 but it connected the same way.

Hope this helps, even if you don't use Lenz.

 

[phils2um]

become problematic due to the wiring, and lots of it.

This along with the need to break into existing trackwork to install current sensing or braking modules is the reason I went the route I did. Like Diesel2000, my entire feedback system is layered on top of my existing RR. I did need to run a Cat6 s88 bus and local sensor cables. But this was a whole lot simpler than trying to break existing trackwork into a bunch of hardwired and isolated track sections. i set up my IR sensor pairs using inexpensive 4-solid conductor 22 awg Indoor/outdoor "security" cable. I prepped all IR sensor pairs on my workbench. The IR emitter and receiver connections to the security cable were soldered, covered with heat shrink then "painted" with a couple layers of black "Liquid Electrical Tape". Only one cable per sensor pair needed to run back to my detection modules when it came time to install the IR sensors.

 

[Airbuspilot]

This along with the need to break into existing trackwork to install current sensing or braking modules is the reason I went the route I did. Like Diesel2000, my entire feedback system is layered on top of my existing RR. I did need to run a Cat6 s88 bus and local sensor cables. But this was a whole lot simpler than trying to break existing trackwork into a bunch of hardwired and isolated track sections. i set up my IR sensor pairs using inexpensive 4-solid conductor 22 awg Indoor/outdoor "security" cable. I prepped all IR sensor pairs on my workbench. The IR emitter and receiver connections to the security cable were soldered, covered with heat shrink then "painted" with a couple layers of black "Liquid Electrical Tape". Only one cable per sensor pair needed to run back to my detection modules when it came time to install the IR sensors.

Thanks Diesel 2000 and Phil
I tried photo resistors, I/R and Hall sensors with my first try at automation using Arduino micro controllers. I found the hall sensors to be the easiest and most reliable, my preference is for current sensing which is what I am using today but if I must look for an alternative then I think I will go the hall sensor way. Today I am looking to the future so not urgent, winter is coming so plenty to do in preparation.
Robin

 

[Diesel2000]

Thanks Diesel 2000 and Phil
I tried photo resistors, I/R and Hall sensors with my first try at automation using Arduino micro controllers. I found the hall sensors to be the easiest and most reliable, my preference is for current sensing which is what I am using today but if I must look for an alternative then I think I will go the hall sensor way. Today I am looking to the future so not urgent, winter is coming so plenty to do in preparation.
Robin

I think current sensing feedback in general only really has the advantage if you are 1) building it into a layout from the start, 2) have the ability to locate modules on the layout (or a well defined wiring bus), and 3) have decoders that support Railcomm in order to know which loco is in each block directly. For an indoor G scale layout on tables/benchwork then this would work just like traditional N/HO layouts.

For outdoors the hall sensor approach is rock solid and is really straightforward to set up. If you locate the hall sensor track contacts at all entry/exit points of blocks then iTrain can track very well if all the block lengths are properly input and the locos have accurate speed measurements. I've been extremely impressed with how well this works on my outdoor layout. Since iTrain has no knowledge of reverse sections or polarity requirements unless specified they are just treated like any other block.

 

[Airbuspilot]

I think current sensing feedback in general only really has the advantage if you are 1) building it into a layout from the start, 2) have the ability to locate modules on the layout (or a well defined wiring bus), and 3) have decoders that support Railcomm in order to know which loco is in each block directly. For an indoor G scale layout on tables/benchwork then this would work just like traditional N/HO layouts.

For outdoors the hall sensor approach is rock solid and is really straightforward to set up. If you locate the hall sensor track contacts at all entry/exit points of blocks then iTrain can track very well if all the block lengths are properly input and the locos have accurate speed measurements. I've been extremely impressed with how well this works on my outdoor layout. Since iTrain has no knowledge of reverse sections or polarity requirements unless specified they are just treated like any other block.

Interesting points, thanks. Our layout is not at the starting point but it is not close to being mature so whatever solution we ultimately use I guess the work will be much the same. My experience of current sensing, admittedly on a small N gague layout, is that a single wire to a simple unit is all that is required. To use a Hall sensor needs something between it and the sensing unit, for me it was an Arduino micro processor which adds to the complication. However I agree the Hall sensors are excellent and you are correct a reverse loop would be just another piece of track.

My little layout is run automatically with iTrain but I am still discovering its capabilities, do you really need a sensor at both entry and exit points of each block? I would have thought the entry point would be enough as iTrain calculates the position of the train in the block and in the case of a station stops reliably at a precise point.

Did you make your own sensors or purchase one? Can you tell me which one you use if purchased?

Robin

 

[Airbuspilot]

Interesting points, thanks. Our layout is not at the starting point but it is not close to being mature so whatever solution we ultimately use I guess the work will be much the same. My experience of current sensing, admittedly on a small N gague layout, is that a single wire to a simple unit is all that is required. To use a Hall sensor needs something between it and the sensing unit, for me it was an Arduino micro processor which adds to the complication. However I agree the Hall sensors are excellent and you are correct a reverse loop would be just another piece of track.

My little layout is run automatically with iTrain but I am still discovering its capabilities, do you really need a sensor at both entry and exit points of each block? I would have thought the entry point would be enough as iTrain calculates the position of the train in the block and in the case of a station stops reliably at a precise point.

Did you make your own sensors or purchase one? Can you tell me which one you use if purchased?

Robin

Disregard my sensor at each end remark, the loop is bi directional of course As will be the rest of the layout as a result.

 

[Diesel2000]

Disregard my sensor at each end remark, the loop is bi directional of course As will be the rest of the layout as a result.

Exactly, you need a sensor at each end of the block if its bidirectional, which yours will be if including a reversing section. On some shorter blocks I've only used 1 sensor in the middle of the block just to get going with the layout, but I've found after 6 months of operating that ideally those should have the sensor at each end as well as braking can become too abrupt if there is traffic. I plan to add those remaining sensors in the spring.

I'm using Massoth digital track contacts for the hall sensors. They can collect power directly from the rails. I'm soldering ring terminals to the power leads and connecting them to rail joiner screws. The signal leads need to be connected to a feedback module. This is the biggest hurdle with the hall sensors approach, but in my estimation far less wiring than if using current sensing modules where you need to bring track power into the current module from each block and then isolate every block in your layout.

I have 23 blocks on my layout, but only have 4 isolated power districts because iTrain uses 'logical' blocks as defined by measurements between switches. By using the hall sensors I dont have to have 23 physically isolated blocks as I would with current sensors.

I took this approach because I wanted to automate the whole layout, however, there is no reason you couldnt mix current sensing blocks with hall sensor-monitored blocks. It just really depends on how your layout is configured and how difficult it will be to wire. In my case I already had 24 wire pairs installed on the layout for 4-channel switch decoders. I converted these all to single channel decoders at the switches and then reused the wires for the hall sensor connections to the feedback modules. I basically just swapped the 4-channel Massoth switch decoders for Massoth feedback modules on my main control board in the house, so for me the hall sensors were the easiest approach.

 

[Airbuspilot]

Exactly, you need a sensor at each end of the block if its bidirectional, which yours will be if including a reversing section. On some shorter blocks I've only used 1 sensor in the middle of the block just to get going with the layout, but I've found after 6 months of operating that ideally those should have the sensor at each end as well as braking can become too abrupt if there is traffic. I plan to add those remaining sensors in the spring.

I'm using Massoth digital track contacts for the hall sensors. They can collect power directly from the rails. I'm soldering ring terminals to the power leads and connecting them to rail joiner screws. The signal leads need to be connected to a feedback module. This is the biggest hurdle with the hall sensors approach, but in my estimation far less wiring than if using current sensing modules where you need to bring track power into the current module from each block and then isolate every block in your layout.

I have 23 blocks on my layout, but only have 4 isolated power districts because iTrain uses 'logical' blocks as defined by measurements between switches. By using the hall sensors I dont have to have 23 physically isolated blocks as I would with current sensors.

I took this approach because I wanted to automate the whole layout, however, there is no reason you couldnt mix current sensing blocks with hall sensor-monitored blocks. It just really depends on how your layout is configured and how difficult it will be to wire. In my case I already had 24 wire pairs installed on the layout for 4-channel switch decoders. I converted these all to single channel decoders at the switches and then reused the wires for the hall sensor connections to the feedback modules. I basically just swapped the 4-channel Massoth switch decoders for Massoth feedback modules on my main control board in the house, so for me the hall sensors were the easiest approach.

Thanks again for your answers, I have looked at the Massoth detectors and feedback modules. You are correct they could be the solution, either for the reverse loops in isolation or for the complete layout for standardisation.

One question, how do you tell iTrain to ignore the block exit sensor? I use two sensor tracks for stations to fix the stop point but the track is uni directional. I assume your entry and exit sensor is at the very start of the block and your blocks are bi directional?