Welcome to the forum   
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

2 Pages12>
Share
Options
View
Go to last post in this topic Go to first unread post in this topic
Offline French_Fabrice  
#1 Posted : 02 October 2019 19:15:07(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello,

I'm wondering if a simple method exists, to add more feedback sections on the bridge of the TT ?

Today, if I want to get feedback from the TT bridge occupation, I can use either 'current' feedback (B) or 'ground' feedback (0).
This can be done rather simply, either by cutting the B tongue of spoke tracks (B feedback), or adding SL11 insulators on each spoke track (0 feedback).

However, this kind of detection allows only to detect "the bridge is occupied", but I cannot know where is located the loco, i.e. is the loco "entirely" on the bridge or not ? A possible solution is to use timers associated to the software, but a different timer should be associated to each loco, because they don't have same running characteristics...

The real solution would be to split the bridge detection in 2 (or 3) detection areas.
But... there are only 6 wires on the central axis... I need 2 more...

How to do it ?

Your experience is welcome.

Best
Fabrice
Offline DaleSchultz  
#2 Posted : 02 October 2019 19:33:48(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
I have also puzzled with this problem...

I can only think that the information must be broadcast out of the TT via wifi or other radio signal, as you say, there are not enough wires to do it.

I also do not want to try to using timers, one need real positive feedback.

Now that RemoteSign ESP modules have sensors, this could perhaps be done using one, but where do you want the data to go??? I suppose another device could relay it into the s88 system....
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline kiwiAlan  
#3 Posted : 02 October 2019 22:49:07(UTC)
kiwiAlan

United Kingdom   
Joined: 23/07/2014(UTC)
Posts: 8,067
Location: ENGLAND, Didcot
Originally Posted by: French_Fabrice Go to Quoted Post

The real solution would be to split the bridge detection in 2 (or 3) detection areas.
But... there are only 6 wires on the central axis... I need 2 more...

How to do it ?


Surely the 'two more' are the tracks around the rim of the turntable? OK, you cannot know that the loco is exactly centred, but all you need is to wait a second or so after the last axle has left the fixed track piece on the edge of the turntable pit. The loco is not moving at speed (or shouldn't be Scared Scared Scared ), so isn't going to travel far.

If you don't have a track on the opposite side of the turntable pit to where the loco is entering the turntable then you only get one extra contact - but you shouldn't be travelling at a speed that is going to overrun anyway. If you have a loco where the turntable is only just long enough to hold it then I suspect you should control it manually anyway.

thanks 3 users liked this useful post by kiwiAlan
Offline Purellum  
#4 Posted : 02 October 2019 23:20:28(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

Just an idea, I haven't tried this BigGrin

Using ground feedback, and dividing the feedback-rail i 3 sections - one in the middle and two short pieces at the end.

The two short pieces at the ends gets wired together, so you can't see if your not far enough on, or too far on to the turntable.

The two short pieces gives feedback via a diode "pointing to the rails" and
the middle section gives feedback via a diode "pointing away from the rail".

Then two DC-relays, also with diodes connected, will be able to "see" which track sections are occupied.

( Written fast, it's late here; I hope it makes sense BigGrin )

Per

P.S: We could go further, using zener diodes or resistors on the turntable, and the having more
advanced electronics to determine exactly where the loco is Cool

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 4 users liked this useful post by Purellum
Offline JohnjeanB  
#5 Posted : 03 October 2019 00:00:47(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Hi Fabrice

Re. 7286, cutting and installing more section can be made easily:
- cutting one of the rails with a rotating diamond blade (I use a Proxxon drill) after removal of the "planking"
- connecting the cutted portion of the rail either soldering from underneath on the existing contact blades or by using Phosphoric Acid to solder directly on the stainless steel rail
- connecting the rails to detectors: I use the DSD2010 decoder for turntable which allows 3 sections to be processed and fed back to the CS2 or CS3.
I have done this to process loco mouvements on the TT even very long ones (BR45 and SNCF 241 A)
I hope this answers your question.
Cheers
Jean
thanks 4 users liked this useful post by JohnjeanB
Offline DaleSchultz  
#6 Posted : 03 October 2019 02:39:28(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
I like Per's idea....
if the bridge is divided into three sections... A, B, & C
=A=|===B===|=C=

A and C are small lengths, essentially when the wheels get to the far side, it triggers A or C
If A and C produce half wave rectified current, then we can perhaps detect four states:

Nothing (loco not yet on TT)
Half wave positive (Section A and perhaps B)
Nothings (loco in middle section- B)
Half wave negative (section B and C)
Full wave current (A B and C - long locomotive)


We would need a circuit that takes a single input and triggers one or two of two outputs. Those outputs would go into an s88 so that the data can flow into whatever software is being used.
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
thanks 3 users liked this useful post by DaleSchultz
Offline French_Fabrice  
#7 Posted : 03 October 2019 08:11:40(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello Friends,

Thanks to all for your ideas/feedback !

@Jean: I know you have used DSD2010. I spent yesterday's evening to inspect the DSD2010 concept and details : https://www.digital-bahn...2010/dsd2010_konzept.htm . However, It doesn't seem to be a simple affair (bridge, motor, signals, & decoder), and I need to entirely translate the content, as I don't read German.

@Per & Dale: Interesting idea. When you have time, would you mind to elaborate a bit more, and possibly provide some electronics schema ? The final goal is to have 3 S88 sensors...

On my side, I'm still thinking. The idea currently in my mind would be to add some adhesive copper bands (2) on the side of the central cone, in order to provide 2 more contacts + build some additional contacts with brass, connected to each new section at the ends of the bridge...

@Jean: You said "after removal of the planking". Is the planking clipsed or glued ?

Cheers
Fabrice
Offline JohnjeanB  
#8 Posted : 03 October 2019 12:53:06(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Hi Fabrice

DSD2010 installation: yes you are right. Installation and programming / configuration is tricky. For me the hardest part was the wiring under the bridge with signals, lighting and sound module.
The configuration and diagnostic software (free) is a piece of art. See here for latest version Digital Bahn download page
But the result is very rewarding.
What I call the Planking are those 2 metal parts on the bridge between the rails to bring power to the slider.
Cheers

Jean
Offline French_Fabrice  
#9 Posted : 03 October 2019 14:17:43(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Thanks Jean !

I misunderstodd the "planking" word you used. Yes, the central part is easy to remove, only one screw.
My 'planking' word was refering to the brown decoration on the outer side of each rail, with looks like a set of planks...

My feeling is, when you cut the rails with a disc, you may damage a bit the edges...

Cheers
Fabrice
Offline DaleSchultz  
#10 Posted : 03 October 2019 20:20:35(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
Originally Posted by: French_Fabrice Go to Quoted Post

@Per & Dale: Interesting idea. When you have time, would you mind to elaborate a bit more, and possibly provide some electronics schema ? The final goal is to have 3 S88 sensors...


Why three sensors?

You only need two, one at each end of the bridge. The section between them would be 'on' if it had a sensor most of the time anyway...

I am not sure how to design the circuit part, I am guessing it would have two diodes in parallel, and then some transistors beyond those that switched ground to the s88 inputs....
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline Purellum  
#11 Posted : 03 October 2019 22:21:34(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

If you make =A=|===B===|=C= feedback, like Dale suggested, I would like to know:

Are you planning to make the "B"-section long enough to hold ALL your locos,
so they can be centered without touching the "A" or "C" sections?

Or, do you want to make it as Dale suggested, so that short locos can stay centered, only touching "B"
and long locos will touch both "A", "B" and "C" ?

And ( I have to ask, since I don't have a turntable myself BigGrin ) would it be possible to modify the turntable even more,
installing a simple ( Old DELTA ) decoder to control the turntable motor, and thus freeing up more wires for feedback?

I also don't know how the S88 reacts to only getting "half wave" feedback signals, so this is a simple schematic
of how I originally thought it should be done, using two relays to give signals to the S88.

This is based on making the "B"-section long enough to hold ALL your locos,
so they can be centered without touching the "A" or "C" sections.

You will not be able to tell from the feedback if the loco is not yet far enough on to the turntable or too far;
this has to be done in the programming of your control system.

UserPostedImage

I love challenges like this BigGrin

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 1 user liked this useful post by Purellum
Offline Purellum  
#12 Posted : 03 October 2019 22:47:28(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

This is how the schematic should look, if we follow Dale's suggestion, with no feedback on the middle part
of the turntable ( "B" ) and different feedback depending if the loco is on "A" or "C", or both "A" and "C".

I can see the advantages BigGrin

However, your controlling software should be able to handle enter and exit both from the "A"-side and the "C"-side Cool

UserPostedImage

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 3 users liked this useful post by Purellum
Offline DaleSchultz  
#13 Posted : 03 October 2019 23:10:22(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
great job, that matches my understanding.

When driving a loco onto the TT...

if S88 port K1 goes on first stop when s88 post K2 goes high
if S88 port K2 goes on first stop when s88 post K1 goes high


Length of one short section plus the length of middle section = longest loco that can be automatically positioned

if the software can handle the the conditional stuff, just always drive on from A side?

Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline DaleSchultz  
#14 Posted : 03 October 2019 23:13:43(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
BTW since steam locos tend to be as long as the TT, sensing the middle section as one input and treating the two outer sections as one, you will not be able to detect when a long loco has reached the far end. I think that is why Per asked about loco length.
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline DaleSchultz  
#15 Posted : 03 October 2019 23:19:41(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
adding a loco decoder is also a great idea, then two wires are used for digital track plus ground, and the rest can be used for 4 sensors.

It does mean that the TT turning then has to be controlled as a loco address and not as an accessory.

I have been thinking about getting proper feedback on the position of the bridge, by placing a magnet on the underside of the bridge and the magnetic switches on the underside of the TT 'basin' and connect those to an s88 module.

The fact that Märklin produces and sells an important item such as the TT without feedback of position of bridge and loco position indicate that either truly do not understand digital control, or just don't care.
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline Purellum  
#16 Posted : 03 October 2019 23:45:56(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

Originally Posted by: DaleSchultz Go to Quoted Post
.....and the rest can be used for 4 sensors.


Which could be 8, using the diode trick Cool

Originally Posted by: DaleSchultz Go to Quoted Post
I have been thinking about getting proper feedback on the position of the bridge, by placing a magnet on the underside of the bridge and the magnetic switches on the underside of the TT 'basin' and connect those to an s88 module.


If you could rebuild the turntable mechanically, so that the center axis extends down through the base, you could add a rotary encoder Cool

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 2 users liked this useful post by Purellum
Offline Purellum  
#17 Posted : 04 October 2019 00:02:49(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

P.S: Having a decoder to control the turntable motor and at the same time an exact positioning system,
could give you acceleration and deceleration of the turntable bridge.......... Cool

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 1 user liked this useful post by Purellum
Offline French_Fabrice  
#18 Posted : 04 October 2019 08:08:28(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello Per & Dale,

Many thanks for the schema and ideas. Great job ThumpUp

My answer and remarks to the questions:
-My longest loco is around 29 cm (BR45), when the bridge is 31cm length, so the 2nd schema would be the good base to start.
-However, "centering" a shortest loco on the bridge without a middle sensor can be difficult... That would lead to have maybe 4 sensors, 2 at both ends with short length, and nothing at the center
-I'm not sure adding a decoder to the TT is a good idea -> All programming of the hardware may disappear. I'm actually using a TT-DEC decoder from LDT and I'm pleased with this device. I have currently 2 feedback : bridge occupied (1 section only = all bridge length) and "position reached" i.e. the bridge when moving has reached the expected position.

I don"t clearly understand the meaning of the S88 and associated switches in the schema ??? Can you please provide a bit more explanation for a not so good electronic newbie Blushing ?

Many thanks again

Fabrice
Offline JohnjeanB  
#19 Posted : 04 October 2019 08:49:09(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Originally Posted by: French_Fabrice Go to Quoted Post
Thanks Jean !

I misunderstodd the "planking" word you used. Yes, the central part is easy to remove, only one screw.
My 'planking' word was refering to the brown decoration on the outer side of each rail, with looks like a set of planks...

My feeling is, when you cut the rails with a disc, you may damage a bit the edges...

Cheers
Fabrice


Hi Fabrice
With a small diameter diamond disk (approx 1.5cm) you can manage to leave the plastic visible planking undamaged
Cheers
Jean
Offline JohnjeanB  
#20 Posted : 04 October 2019 08:56:15(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Originally Posted by: DaleSchultz Go to Quoted Post
Originally Posted by: French_Fabrice Go to Quoted Post

@Per & Dale: Interesting idea. When you have time, would you mind to elaborate a bit more, and possibly provide some electronics schema ? The final goal is to have 3 S88 sensors...


Why three sensors?

You only need two, one at each end of the bridge. The section between them would be 'on' if it had a sensor most of the time anyway...

I am not sure how to design the circuit part, I am guessing it would have two diodes in parallel, and then some transistors beyond those that switched ground to the s88 inputs....

Hi
I started using the TT with 2 sensors and indeed it works BUT with long locomotives 29 cm or more the stopping precision is not good enough now with 3 sensors and Rocrail all locos stop with less than 10 mm precision
Cheers
Jean
Offline DaleSchultz  
#21 Posted : 04 October 2019 16:25:14(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
the s88 modules are the normal sensor detection devices that send information back to the controlling software. Just like normal track with isolated sections.
By s88 we also mean a generic feedback, it may be from a different manufacturer or the newer Märklin ones...
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline French_Fabrice  
#22 Posted : 04 October 2019 17:31:04(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello,

To make it short: I don't understand the need of relays.
What are they supposed to allow: multiplexing 2 wires on one output, latching, other ?

Sorry for being ignorant in this area.

Cheers
Fabrice
Offline DaleSchultz  
#23 Posted : 04 October 2019 19:20:29(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
The relays ensure that the grounding of the s88 port is solid and not semi digital.
I am not sure how much the ground from the track on the bridge will be oscillating, in fact it may not be at all. In that case, polarizing it with diodes wont do much...

Per, can this work at all?
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
thanks 1 user liked this useful post by DaleSchultz
Offline Purellum  
#24 Posted : 04 October 2019 19:32:04(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

I'm not sure the relays are needed at all; I just put them in the drawing to be sure it would work. Cool

If a S88 can detect on a half-wave signal, and I mean "both" half-waves, then the relays are not needed.

I might be able to Google the answer for you BigGrin

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 1 user liked this useful post by Purellum
Offline Purellum  
#25 Posted : 04 October 2019 19:43:04(UTC)
Purellum

Denmark   
Joined: 08/11/2005(UTC)
Posts: 3,496
Location: Mullerup, 4200 Slagelse
Cool

Since this trick works, it means that you need the relays Cool

http://members.ozemail.c.../pdf/the_diode_trick.pdf

On my diagram, you can't send current directly from the S88 to ground on the green wire, so that will not work.

On the orange wire you can send current directly from the S88 to ground, so that will work. ( It always work with orange wire LOL )

This means that you could do this only using K1, and let the orange wire go directly to the S88 and forget about K2.

But I would still use both relays. BigGrin

Per.

Cool
If you can dream it, you can do it!

I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide.

In case this is not legally possible:
I grant anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law.

UserPostedImage
thanks 3 users liked this useful post by Purellum
Offline Ross  
#26 Posted : 05 October 2019 05:49:53(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hello Fabrice/All,

It is possible to add stop sensor contacts to each end of the turntable bridge but there are a few caveats.
First the Marklin/Fleishmann bridge end of rail spring contacts must exist. If they have been cut off this method won't work.
My friend Vic users this method with TrainController to provide a stop contact at each end of the TT bridge with a brake contact between the two stop contacts and it works very well.
Before going down this path I would exhaust all options to improve the speed profile of the locomotives before making permanent physical changes to the turntable/bridge. I haven't found a compelling need to add the extra stop contacts to my turntable but on my transfer table I have added the extra stop contacts.

TT bridge 3 contacts

The photo above shows the TT bridge cabin end at the 9:0 o'clock position. On the Bridge you will notice both rails at each end of the bridge have been cut 55mm from the end of the bridge these small segments are shown in blue and have no electrical connection to them. The cutting of the rails is best done with a very small end mill to avoid cutting the plastic on either side of the rail(s).
The brown rail on the bridge is ground (O) shown on the cabin side and the brake contact is shown in green.

The stub tracks around the turntable is where the s88 contacts are joined up to form two groups (two s88 contacts) (you have to ensure that the stubs rails for each end of the TT bridge is in a different group) I have shown this to be the yellow rail. This yellow rail must be insulated from the connecting stall rail. The brown rail doesn't need to be insulate but must be connected to the ground (O).

As the turntable bridge rotates the bridge spring contacts complete the circuit for the ground (O) and the s88 contact for each end of the bridge.

This method allows 360 degree rotation of the TT bridge.

I hope you have found this useful.

Edit: I have spoken with Vic and the TT bridge extra contact length is 55mm not 30mm as I mentioned.

Turntable rotating demonstration


Originally Posted by: French_Fabrice Go to Quoted Post
Hello,

I'm wondering if a simple method exists, to add more feedback sections on the bridge of the TT ?

How to do it ?

Your experience is welcome.

Best
Fabrice

Edited by user 15 October 2019 02:04:47(UTC)  | Reason: Not specified

Ross
thanks 4 users liked this useful post by Ross
Offline DaleSchultz  
#27 Posted : 05 October 2019 15:58:37(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
a nice approach Ross, (and great animated images too)

but I don't recall there being any running rail contacts between the bridge and spoke rails... can you show a picture of what you mean by "rail spring contacts" ?

I only know of the center rail contact, (which would cause short circuits when it touched running rails).
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline French_Fabrice  
#28 Posted : 05 October 2019 23:47:24(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello Ross, All,

This is a nice concept. You have moved the problem of additional bridge wires outside of the bridge, i.e. on the spoke tracks.

However, I'm suspecting possible problems with automation... I must say I haven't experimented this kind of configuration until so far, and it may depends also of the software used, but this alerts me a bit.

For the following explanations, I've reused part of your drawings, with some addition.

UserPostedImage
Note: Not all wires are shown

Let's call the 3 sensors/areas of the bridge "enter", "brake", "in".
The "brake" can be ignored for the rest of the explanation.

If I understand correctly, one rail of half of spoke track set are connected together to the same S88 input (Enter) -I have marked them E0, E46, E44... to E26, and the other rail is grounded for the same set.
For the remaining half spoke tracks set, tracks are connected in the same manner, to another same S88 input : In. I have marked them I2 to I24.
If this is not like that, then I've missed something. Please provide more explanations.

Here is an example of questions I have in mind:
-Q1: A loco is entirely on the bridge (0-24), i.e. All sensors are up (enter, brake & in). Now you move the bridge to another position, for instance 46-22. When the bridge moves (without having reached the final position), enter & in disappear because grounded sections disappear, even if brake is still high. What is the behavior of the software when some sensors of bridge vanish while rotating ?
-Q2: same kind of question, when the bridge reaches the new position, i.e. the sensors enter & in are back.
-Q3: now a variant of Q2. The new position reached is such that enter & in sensors are swapped compared to the initial position, i.e. 2-26... What is the behavior ?

Maybe the answer is TC doesn't care to lose some sensors/ignore them while bridge is moving... Let's suppose this is the case also with Rocrail.

-Q4: More annoying, at least with Rocrail: A loco wants to enter the bridge coming from I2 side. It enters the spoke track and fires I2, thus "In" sensor, without having fired previously the "Enter" sensor -> Unexpected detection = Ghost Train = Emergency power off !

I'm sure TrainControler is extremely sophisticated and is (possibly?) able to deal with such stuff, but I'm not sure with Rocrail.

These considerations are theoretical and I would need to simulate a complete behavior in order to be sure it's OK with Rocrail, before going that way.

However, many thanks for your suggestion, but at the present time it raises many questions in my mind.

Cheers
Fabrice
Offline Ross  
#29 Posted : 06 October 2019 02:46:48(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi Dale,

Thanks for your comments.As requested a photo showing the spring rail contacts which are at each end of the turntable bridge.

TT spring rail contacts

Originally Posted by: DaleSchultz Go to Quoted Post
a nice approach Ross, (and great animated images too)

but I don't recall there being any running rail contacts between the bridge and spoke rails... can you show a picture of what you mean by "rail spring contacts" ?

I only know of the center rail contact, (which would cause short circuits when it touched running rails).


Ross
thanks 2 users liked this useful post by Ross
Offline DaleSchultz  
#30 Posted : 06 October 2019 03:08:48(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
thanks!

I have discovered that since I landscaped my TT in, I can no longer get the bridge out! It demands at least one of the spokes to be removed....
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline Ross  
#31 Posted : 06 October 2019 03:52:05(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi Fabrice,

I'll try and answer your questions below. As I mentioned my friend Vic uses this method and it works well. Cutting tracks on the TT bridge is the point of no return. You must be certain this is the path you want to go down. Experiment with some test track the same length as the bridge. Make the 3 contacts and wire to s88 contacts then use your software Rocrail to drive the profiled loco on to the test track. For the two stop sensors add a switch in line with the wires to the s88 contacts so you can switch the sensors off to simulate the TT bridge movement then after a few seconds switch the sensors back on and see if this has adverse problems for the software. The block should remain highlighted with the loco because the middle sensor isn't switched off. Please provide feedback on your testing.

TT bridge block stop and brake markers

In TrainController software the Stop markers are shown as red directional markers and the orange directional markers is the brake with a distance over which the loco slows to threshold speed until it hits the stop marker. The markers are associated to the 3 TT bridge sensors/s88 contacts.


Originally Posted by: French_Fabrice Go to Quoted Post
Hello Ross, All,

This is a nice concept. You have moved the problem of additional bridge wires outside of the bridge, i.e. on the spoke tracks.

However, I'm suspecting possible problems with automation... I must say I haven't experimented this kind of configuration until so far, and it may depends also of the software used, but this alerts me a bit.

For the following explanations, I've reused part of your drawings, with some addition.
I have added to you wiring diagram to show the ground(O) connection and the s88 sensor connections. It is important to have an equal number of wires to the 2 stop s88 contacts that represent both ends of the turntable bridge. I would think of the of the bridge ends as position 1 and 2 rather than "enter" and "in".

tt wiring update

Note: Not all wires are shown

Let's call the 3 sensors/areas of the bridge "enter", "brake", "in".
The "brake" can be ignored for the rest of the explanation.

If I understand correctly, one rail of half of spoke track set are connected together to the same S88 input (Enter) -I have marked them E0, E46, E44... to E26, and the other rail is grounded for the same set.
For the remaining half spoke tracks set, tracks are connected in the same manner, to another same S88 input : In. I have marked them I2 to I24.
If this is not like that, then I've missed something. Please provide more explanations.
Remember for both ends of the bridge the number of wires must be equal for each s88 contact.

Here is an example of questions I have in mind:
-Q1: A loco is entirely on the bridge (0-24), i.e. All sensors are up (enter, brake & in). Now you move the bridge to another position, for instance 46-22. When the bridge moves (without having reached the final position), enter & in disappear because grounded sections disappear, even if brake is still high. What is the behavior of the software when some sensors of bridge vanish while rotating ?
-Q2: same kind of question, when the bridge reaches the new position, i.e. the sensors enter & in are back.
-Q3: now a variant of Q2. The new position reached is such that enter & in sensors are swapped compared to the initial position, i.e. 2-26... What is the behavior ?

Maybe the answer is TC doesn't care to lose some sensors/ignore them while bridge is moving... Let's suppose this is the case also with Rocrail.
I think as I stated above the block should remain highlighted with the loco because the middle sensor isn't switched off. I hope this is the case.

-Q4: More annoying, at least with Rocrail: A loco wants to enter the bridge coming from I2 side. It enters the spoke track and fires I2, thus "In" sensor, without having fired previously the "Enter" sensor -> Unexpected detection = Ghost Train = Emergency power off !
I don't understand this statement. If I2 is fired first this I would expect the brake sensor to fire next and finally stop ("in') by firing E26 as shown in your diagram.

I'm sure TrainControler is extremely sophisticated and is (possibly?) able to deal with such stuff, but I'm not sure with Rocrail.

These considerations are theoretical and I would need to simulate a complete behavior in order to be sure it's OK with Rocrail, before going that way.

However, many thanks for your suggestion, but at the present time it raises many questions in my mind.

Fabrice, I hope I have answered you questions with the extra diagrams etc.

Cheers
Fabrice
Ross
thanks 5 users liked this useful post by Ross
Offline DaleSchultz  
#32 Posted : 06 October 2019 04:10:26(UTC)
DaleSchultz

United States   
Joined: 10/02/2006(UTC)
Posts: 3,997
Great stuff Ross!

also to test, does the software handle the fact that the track which is the TT bridge sometimes connects to track X and at other times track Y....
Dale
Intellibox + own software, K-Track
My current layout: https://cabin-layout.mixmox.com
Arrival and Departure signs: https://remotesign.mixmox.com
Offline Ross  
#33 Posted : 06 October 2019 05:42:05(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi Dale,

10 demerits for lack of serviceability Blushing

Originally Posted by: DaleSchultz Go to Quoted Post
thanks!

I have discovered that since I landscaped my TT in, I can no longer get the bridge out! It demands at least one of the spokes to be removed....


Ross
Offline Ross  
#34 Posted : 06 October 2019 06:03:26(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi Dale,

I'm pleased you are finding the topic interesting.

X and Y directions isn't a problem. Note the directional markers.

tt fwd and rev props

In the form shown you can specify how the loco exits the turntable by pointing the TT bridge cabin end to the track stubs required then ticking the forward or reverse boxes. In the example the loco can enter the TT bridge in a forward or reverse direct but on the stubs marked orange the loco may only exit the bridge in a reverse direction. This can be overridden at the schedule level for automatic operation see below.

TT direction overide

Originally Posted by: DaleSchultz Go to Quoted Post
Great stuff Ross!

also to test, does the software handle the fact that the track which is the TT bridge sometimes connects to track X and at other times track Y....
Ross
thanks 4 users liked this useful post by Ross
Offline French_Fabrice  
#35 Posted : 06 October 2019 09:16:58(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Thank you, Ross, for your additional explanations.

That means I had correctly understood the wiring.

I'm going to do a complete manual simulation, in order to check if Rocrail is able to swap the meaning of "Enter" and "In" sensors depending on which semi-circle it is fired.

As you stated, cutting the track bridge is the point of no return, so I must be sure...Confused
It'll take a bit of time, but I report my experimentation in this thread.

Thanks again
Fabrice
thanks 4 users liked this useful post by French_Fabrice
Offline kiwiAlan  
#36 Posted : 06 October 2019 13:56:28(UTC)
kiwiAlan

United Kingdom   
Joined: 23/07/2014(UTC)
Posts: 8,067
Location: ENGLAND, Didcot
Originally Posted by: French_Fabrice Go to Quoted Post
Hello Ross, All,

This is a nice concept. You have moved the problem of additional bridge wires outside of the bridge, i.e. on the spoke tracks.


This is what I was suggesting up in post 3.

Offline French_Fabrice  
#37 Posted : 06 October 2019 14:07:42(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello kiwiAlan,

I'm sorry, I may have read your post too quickly.Blushing

Cheers
fabrice
Offline kiwiAlan  
#38 Posted : 06 October 2019 14:45:25(UTC)
kiwiAlan

United Kingdom   
Joined: 23/07/2014(UTC)
Posts: 8,067
Location: ENGLAND, Didcot
Originally Posted by: French_Fabrice Go to Quoted Post
Hello kiwiAlan,

I'm sorry, I may have read your post too quickly.Blushing

Cheers
fabrice


Don't worry, you are not the only one guilty of that Blushing , and I may not have explained it clearly either Blushing .

Certainly the pictures made it a lot clearer. ThumpUp

thanks 1 user liked this useful post by kiwiAlan
Offline JohnjeanB  
#39 Posted : 06 October 2019 15:00:04(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Hi Fabrice
With Rocrail it is possible to use virtual sensors based on real sensors associated with conditions so almost all is possible
With Rocrail during the TT rotation nothing happens
Cheers
Jean
thanks 1 user liked this useful post by JohnjeanB
Offline French_Fabrice  
#40 Posted : 06 October 2019 15:14:23(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Thank you Jean,

This solves Q1 and Q2.
I haven't thought about virtual sensors. Good idea... And probably some XML to write conditions...

Cheers
Fabrice
Offline Ross  
#41 Posted : 15 October 2019 02:07:58(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi All,

I have corrected the TT bridge contact length in the original text see colour text above.
Ross
thanks 2 users liked this useful post by Ross
Offline French_Fabrice  
#42 Posted : 22 January 2020 13:59:02(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello,

I promised to provide some feedback...

Unfortunately, one of the spring ensuring ground contact between bridge and spoke tracks has unwelded Crying.
This prevents to test the method provided by Ross.

As a result, I've cut off the 3 remaining springs, as they are no more useful.

I think I'm going to test the bridge occupation with only one feedback for whole bridge, and speed curve associated to each loco, as I did previously.

Cheers
Fabrice
Offline Ross  
#43 Posted : 22 January 2020 23:09:05(UTC)
Ross

Australia   
Joined: 25/09/2006(UTC)
Posts: 867
Location: Sydney, NSW
Hi Fabrice,

Shame you had a faulty contact and decided to cut the other contacts off. I would of first taken the bridge and loose contact to a metal fabrication shop to see if it could be spot welded back on. With the contacts cut off you could use the TT stub tracks as an extra contact (one side only of each stub) that as the locomotive enters the TT bridge the stub contact remains on until the locomotive has fully entered the bridge then the stub track contact turns off and the locomotive stops. Food for thought.

Ross
thanks 2 users liked this useful post by Ross
Offline French_Fabrice  
#44 Posted : 23 January 2020 08:10:33(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hi Ross,

Thanks for this interesting ideaThumpUp .
I'll try it. Probably needs a bit of an additional timer to make sure the back of the locomotive doesn't spill off the deck.

Cheers
Fabrice
thanks 1 user liked this useful post by French_Fabrice
Offline mcs51  
#45 Posted : 25 May 2020 16:50:48(UTC)
mcs51

Belgium   
Joined: 25/05/2020(UTC)
Posts: 11
Location: West-Vlaanderen, Kortrijk
Hello,
May I come back on this ??

How did you finally solve the 2 sensor problem in Rocrail Fabrice?
I face the same issue as you did and before cutting things I would like to be sure to end up with a working solution.
One can find a lot of things on the internet but no complete “How to” document on this issue.
I presume all users of this TT, with Rocrail are facing this problem… … no?


Bonjour Jean,
At one time in the discussion (#20) you talk about 3 sensors on the bridge?!?
How did you manage that because there are no spare sliding contacts under the TT.
So how did you bring 3 sensors to the outside world ? Confused ?


Thanks for you inputs guys!
Regards
Alain
Offline French_Fabrice  
#46 Posted : 25 May 2020 19:24:54(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Hello Alain,

Unfortunately I didn't solved it (yet), and I don't know if I could ever find a solution, due to the cutting of contacts Blushing .
At the present time, I have only one contact for the whole bridge, so I must use timers with Rocrail, as I did previously.

Jean has used a very special decoder (DSD2010), and this decoder has the ability to manage up to 3 sensors + signals on the bridge. The change is not obvious as it implies to add some electronics below the deck + program this decoder. Jean will probably provide more info if you are interested.

Cheers
Fabrice
Offline JohnjeanB  
#47 Posted : 25 May 2020 21:58:35(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Hi Alain and you all

I used a DSD 2010 from DigitalBahn.
It is a sophisticated set of decoders for turntables (TT) containing one for the bridge, one for the entire TT, plus an opto sensor for the position reporting.
It is available in semi assembled version (all surface-mounted components are already mounted) but you have to solder the connectors, regular components. for 129€

The processors are already programmed but you have to configure the system (where you have your tracks, motorola or DCC control, test the system and adjust speeds, fault detection). For this a free software can be downloaded and all 3 books documentation alas in German.
The set includes a 64 bit sensor system (compatible with S88) to report on bridge location (48), bridge occupancy (3 sensors), general purpose signals.
The set drives also 4 signals (not included), and a SUSI interface for a sound generator, a flash lamp and cabin illumination
The TT needs to be adapted:
- removal of the TT latching system (solenoid, feeler)
- wiring
- possible addition of the sound system, cabin illumination, signals, warning flash.
This is quite some work but I believe it is worth it so let me know if you are interested.
Here is the result on my layout


So to answer your question, the decoder set includes on the bridge decoder, 3 entries for track sensors (Märklin system), the info, mixed with others is multiplexed and passed to the main decoder which is connected to the system's S88 sensor system

Cheers
Jean
Offline dickinsonj  
#48 Posted : 26 May 2020 01:52:58(UTC)
dickinsonj

United States   
Joined: 05/12/2008(UTC)
Posts: 1,675
Location: Crozet, Virginia
Originally Posted by: JohnjeanB Go to Quoted Post

So to answer your question, the decoder set includes on the bridge decoder, 3 entries for track sensors (Märklin system), the info, mixed with others is multiplexed and passed to the main decoder which is connected to the system's S88 sensor system

Cheers
Jean


Ah Jean - that video is so amazing and I have watched it many times. Your layout is exactly what I aspire to one of these days. ThumpUp

The provisioning of the 241 is a joy to behold. Your layout and control of it are exemplary and a worthy goal for me as I move closer to my forever layout.

Thanks for your inspiration and also from all of the great MRR clan here on this forum. BigGrin

Cheers,

Jim
Regards,
Jim

I have almost all Märklin and mostly HO, although I do have a small number of Z gauge trains!
So many trains and so little time.
Offline JohnjeanB  
#49 Posted : 26 May 2020 11:24:13(UTC)
JohnjeanB

France   
Joined: 04/02/2011(UTC)
Posts: 3,078
Location: Paris, France
Hi Jim
Thank you for your compliments.

Yes I really enjoy the 241 and I was almost sure not to have it but my son found one Trix 9 month after the normal deliveries.
I had multiple layouts but it is the first one where I have a turntable and multiple garage tracks I enjoyed controlling by PC.

It is so much fun that if I may help one of my fellow train lovers I will.
Cheers
Jean
thanks 1 user liked this useful post by JohnjeanB
Offline kiwiAlan  
#50 Posted : 26 May 2020 12:47:03(UTC)
kiwiAlan

United Kingdom   
Joined: 23/07/2014(UTC)
Posts: 8,067
Location: ENGLAND, Didcot
Originally Posted by: mcs51 Go to Quoted Post

Bonjour Jean,
At one time in the discussion (#20) you talk about 3 sensors on the bridge?!?
How did you manage that because there are no spare sliding contacts under the TT.
So how did you bring 3 sensors to the outside world ? Confused ?


Thanks for you inputs guys!
Regards
Alain


hello Alain,

I don't believe you need three sensors on the bridge itself. I think the it should be possible to use the radial tracks of the turntable as the entry and exit sensors (positions 1 and 3) with the rotating turntable being the central sensor ( sensor 2). After all a loco will (should) not be entering the turntable at more than 'idle' speed so once whichever radial sensor is no longer energised by the loco then speed step 0 should be commanded, and whatever momentum the loco has will be enough to stop it in the middle of the rotating section. This may require some slight fine tuning if time delays in software for different sizes of loco with different amounts of momentum set, but for the most part should be correct.

This is one of the things I want to play with once I get back home.

thanks 3 users liked this useful post by kiwiAlan
Users browsing this topic
2 Pages12>
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

| Powered by YAF.NET | YAF.NET © 2003-2024, Yet Another Forum.NET
This page was generated in 1.267 seconds.