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

Notification

Icon
Error

2 Pages<12
Share
Options
View
Go to last post in this topic Go to first unread post in this topic
Offline efel  
#51 Posted : 11 June 2015 04:32:20(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
Hi Fabrice / Ross / Fred

Thank you all for your support

Here are the results of the tests asked by Fred:

No resistance at all in any of the connections......

....


Hi all,

So there is probably a capacitor connected to the B wire of the bridge (why???). Esteban, could you check that, visually? If yes, what is its value (written on it)? where is connected its other leg?

In order to cancel the effect of the parasitic curent due to that capacitance, a capacitor should be connected between the B input of the 5233 (that goes to the B of the central) and its ouput that goes to the B of the bridge track. Its value will depend on the value of the capacitor connected to the bridge.

Of course, it would be interesting to know, in that eventuality, what is the purpose of the capacitor in the TT, and why Fabrice does not have one in its TT?

Regards,

Fred


thanks 1 user liked this useful post by efel
Offline ecastrog  
#52 Posted : 11 June 2015 05:13:05(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: efel Go to Quoted Post
Originally Posted by: ecastrog Go to Quoted Post
Hi Fabrice / Ross / Fred

Thank you all for your support

Here are the results of the tests asked by Fred:

No resistance at all in any of the connections......

....


Hi all,

So there is probably a capacitor connected to the B wire of the bridge (why???). Esteban, could you check that, visually? If yes, what is its value (written on it)? where is connected its other leg?

In order to cancel the effect of the parasitic curent due to that capacitance, a capacitor should be connected between the B input of the 5233 (that goes to the B of the central) and its ouput that goes to the B of the bridge track. Its value will depend on the value of the capacitor connected to the bridge.

Of course, it would be interesting to know, in that eventuality, what is the purpose of the capacitor in the TT, and why Fabrice does not have one in its TT?

Regards,

Fred




Hi Fred / all

So, I think here is the damn capacitor.......

IMG_1998.JPGIMG_1996.JPGIMG_1995.JPGIMG_1997.JPG

It is the only thing I see connected to "B" in the TT. Actually, as you can see is underneath the TT

Unfortunately, the other leg, I cannot determine where it goes. It is welded to a wire that goes under that plate, and I tried to disassemble it but can't and I don't want to damage the TT....

Regarding the number, it is really small, but it looks something like 318??. As you know I am not an electrician, so any number sounds the same to me. Maybe to you it means something else. Hope you can see it in the picture....

Thank you!! I think we might be getting closer....BigGrin

Esteban
Offline efel  
#53 Posted : 11 June 2015 09:19:34(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Hi,

In fact that component is likely a ceramic capacitor, but the marking, if we follow standard marking, would mean 0.31pF, which is a very very low value, and a nonsense in the frequency range used here. Furthermore,it is not a normalized value (0.27 or 0.33 pF). Then it's not a standard marking, and i don't know what could be the capacitance.

Then a possibility to cancel the parasitic current due to that capacitor is to connect a capacitor on the 5233, as said before, and find the good value by trial-and-error. The value should be less than 10 nF, in order not to degrade the sensitivity.

I understand that you probably don't have capacitors at hand, for you are not electronician. Resistance could be used instead, but with, for sure, a deterioration of the sensitivity, and you probably don't have resistors at hand either .

May be other members could tell more about that TT capacitor?

Regards,

fred
Offline ecastrog  
#54 Posted : 11 June 2015 18:35:43(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Hi Fred / all

Thank you Fred for your help

I will go today to the electronics shop and get some capacitors, in the specs you say (less than 10 nF), so I can do some testing

I could get some resistances also, if you tell me the values...

Now, the question I have is, in order to start the trial-and-error procedure, I need to understand how to connect (at the end weld) the capacitor...

I have this idea in my mind, options 1 or 2, (of course 1 is easier; in 2 I will need to open the 5233...).

Could you please give me some advise regarding this?'


Thank you again!!

Esteban

Capacitor connections.pdf (28kb) downloaded 46 time(s).
Offline French_Fabrice  
#55 Posted : 11 June 2015 19:25:41(UTC)
French_Fabrice

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

Just to provide a comparison, I've been able to take pictures of the below side of my TT.

UserPostedImage
UserPostedImage
UserPostedImage
UserPostedImage

-There is NO capacitor in my TT
-The below side side shown by Esteban is a bit different from mine

... Sorry to be not too helpful. I've forgot a lot from my electronic lessons in the past.

Cheers
fabrice
Offline ecastrog  
#56 Posted : 11 June 2015 19:33:22(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: French_Fabrice Go to Quoted Post
Hi all,

Just to provide a comparison, I've been able to take pictures of the below side of my TT.

.....

-There is NO capacitor in my TT
-The below side side shown by Esteban is a bit different from mine

... Sorry to be not too helpful. I've forgot a lot from my electronic lessons in the past.

Cheers
fabrice


Hi Fabrice

Thank you for sharing your pictures.

You have been more than helpful, and I hope you still follow with me the next steps.... this is far from over!! Crying Crying


Let you know later if I was able to get the capacitors..

Regards,
Esteban
Offline efel  
#57 Posted : 11 June 2015 20:31:51(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
Hi Fred / all

Thank you Fred for your help

I will go today to the electronics shop and get some capacitors, in the specs you say (less than 10 nF), so I can do some testing

I could get some resistances also, if you tell me the values...

Now, the question I have is, in order to start the trial-and-error procedure, I need to understand how to connect (at the end weld) the capacitor...

I have this idea in my mind, options 1 or 2, (of course 1 is easier; in 2 I will need to open the 5233...).

Could you please give me some advise regarding this?'


Thank you again!!

Esteban

Capacitor connections.pdf (28kb) downloaded 46 time(s).


Hi esteban,

Concerning the capacitor, 10nF is a value that doesnot degrade the sensitivity of the 5233. I hope it will be OK, but as the capacitors are not expensive, I suggest you buy also higher values, for the case....
Athough the voltage across that capacitor will be low, buy capacitors of at least 25V (it's practically always the case with those capacitor values). I suggest 1nF, 2.2nF, 4.7nF, 10nf, 22nF, 47nF.

For the resistors, buy 0.25watts type, and values 1.5k, 1k, 0.47k, 0.22k, 0.1k.

If it works, prefer the use of capacitor instead of resistor.

In both cases, it's option 1 of your schematic that should be used ( fortunately!).

Hope you'll succeed,

regards,

Fred





Offline French_Fabrice  
#58 Posted : 11 June 2015 21:29:52(UTC)
French_Fabrice

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

Something intrigues me:

a) When looking at post #35 from Esteban, he says that "Circle1"="brown wire (ribbon)"= "Pin#1" of white plug (leftmost one)= "0 left" (fake board)

b) When various tests have been performed with "0 left" to ground and "B" bridge to "B", and TT facing NO track, the detection was fine.

c) But, if you look at pictures taken by Esteban at post #52, the capacitor is clearly connected with one leg to "brown" (i.e. "0 left") and to an unknown destination with the other leg.


IF (AND ONLY IF) the unknown destination of the other leg of the capacitor is "B" (bridge), that would mean that "0 left" NEEDS this capacitor to provide correct detection... and maybe ? one capacitor is missing between "0 right" (orange wire) and "B" ???.

@Fred: What do you think:
1) of the fact that one leg of the capacitor is connected to "0 left"
2) of my reasoning (I know there are some "if" and "maybe", so you may drop it to the bin...)
Do you think the correction you want to make with an other capacitor is still valid ?

@Esteban: Would mind checking (once again, sorry !) with your multimeter (beep position) -no power, fake board, no bridge - if "0 left" goes to brown (TT ribbon), and "0 right" goes to orange (TT ribbon) at the base of the TT ? Put one test probe on "0 left" and the other one on "circle 1" (BEEP expected), and then one test probe between "0 right" and the other one on "circle 2" (BEEP also expected)...

I hope not adding too much confusion...

Cheers
fabrice
Offline efel  
#59 Posted : 11 June 2015 23:37:02(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: French_Fabrice Go to Quoted Post
Hi all,

Something intrigues me:

a) When looking at post #35 from Esteban, he says that "Circle1"="brown wire (ribbon)"= "Pin#1" of white plug (leftmost one)= "0 left" (fake board)

b) When various tests have been performed with "0 left" to ground and "B" bridge to "B", and TT facing NO track, the detection was fine.

c) But, if you look at pictures taken by Esteban at post #52, the capacitor is clearly connected with one leg to "brown" (i.e. "0 left") and to an unknown destination with the other leg.


IF (AND ONLY IF) the unknown destination of the other leg of the capacitor is "B" (bridge), that would mean that "0 left" NEEDS this capacitor to provide correct detection... and maybe ? one capacitor is missing between "0 right" (orange wire) and "B" ???.

@Fred: What do you think:
1) of the fact that one leg of the capacitor is connected to "0 left"
2) of my reasoning (I know there are some "if" and "maybe", so you may drop it to the bin...)
Do you think the correction you want to make with an other capacitor is still valid ?

@Esteban: Would mind checking (once again, sorry !) with your multimeter (beep position) -no power, fake board, no bridge - if "0 left" goes to brown (TT ribbon), and "0 right" goes to orange (TT ribbon) at the base of the TT ? Put one test probe on "0 left" and the other one on "circle 1" (BEEP expected), and then one test probe between "0 right" and the other one on "circle 2" (BEEP also expected)...

I hope not adding too much confusion...

Cheers
fabrice


Hi Fabrice,

You are perfectly right: the reason why there is a parasitic current in B bridge is not clear, and probably not the capacitor connected to "0 left" (I did not chek before and supposed the capacitor was connected to B bridge).
Nevertheless, the parasitic current flowing through the 5233, and leading to a detection of occupation is probably not due to a resistance ( I say "probably" for the maximum measurement of the Esteban ohmeter is only 20kOhm, and if, unluckily the resistance between B bridge and 0 was just, say, 21kOhm, the parasitic current would be a little larger than 1mA, and still be detected by the 5233. It's the reason why i wrote in a previous post "if resistance is > 100kOhm"). Conclusion: the parasitic current might be resistive, but it's very unlikely. Furthermore, the solution is the same: derivate that parasitic current with a capacitor (or resistor) connected on the 5233, as per option 1 of Esteban pic.

So I'm rather confident to that solution. It will not work only if there is a quite large capacitor between B bridge and "0 right" hidden somewhere.

Anyway, even if we can solve the problem that way, it will be interesting to know what is the exact cause of the problem, and why there are some components on Esteban bridge which are not on your bridge. In my opinion, there is NO need to have ANY component on the 3 wires: B bridge, 0 left and 0 right !

Let's wait the Esteban results..

Regards,

Fred
Offline ecastrog  
#60 Posted : 12 June 2015 02:37:40(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: efel Go to Quoted Post
Originally Posted by: French_Fabrice Go to Quoted Post
Hi all,

Something intrigues me:

a) When looking at post #35 from Esteban, he says that "Circle1"="brown wire (ribbon)"= "Pin#1" of white plug (leftmost one)= "0 left" (fake board)

b) When various tests have been performed with "0 left" to ground and "B" bridge to "B", and TT facing NO track, the detection was fine.

c) But, if you look at pictures taken by Esteban at post #52, the capacitor is clearly connected with one leg to "brown" (i.e. "0 left") and to an unknown destination with the other leg.


IF (AND ONLY IF) the unknown destination of the other leg of the capacitor is "B" (bridge), that would mean that "0 left" NEEDS this capacitor to provide correct detection... and maybe ? one capacitor is missing between "0 right" (orange wire) and "B" ???.

@Fred: What do you think:
1) of the fact that one leg of the capacitor is connected to "0 left"
2) of my reasoning (I know there are some "if" and "maybe", so you may drop it to the bin...)
Do you think the correction you want to make with an other capacitor is still valid ?

@Esteban: Would mind checking (once again, sorry !) with your multimeter (beep position) -no power, fake board, no bridge - if "0 left" goes to brown (TT ribbon), and "0 right" goes to orange (TT ribbon) at the base of the TT ? Put one test probe on "0 left" and the other one on "circle 1" (BEEP expected), and then one test probe between "0 right" and the other one on "circle 2" (BEEP also expected)...

I hope not adding too much confusion...

Cheers
fabrice


Hi Fabrice,

You are perfectly right: the reason why there is a parasitic current in B bridge is not clear, and probably not the capacitor connected to "0 left" (I did not chek before and supposed the capacitor was connected to B bridge).
Nevertheless, the parasitic current flowing through the 5233, and leading to a detection of occupation is probably not due to a resistance ( I say "probably" for the maximum measurement of the Esteban ohmeter is only 20kOhm, and if, unluckily the resistance between B bridge and 0 was just, say, 21kOhm, the parasitic current would be a little larger than 1mA, and still be detected by the 5233. It's the reason why i wrote in a previous post "if resistance is > 100kOhm"). Conclusion: the parasitic current might be resistive, but it's very unlikely. Furthermore, the solution is the same: derivate that parasitic current with a capacitor (or resistor) connected on the 5233, as per option 1 of Esteban pic.

So I'm rather confident to that solution. It will not work only if there is a quite large capacitor between B bridge and "0 right" hidden somewhere.

Anyway, even if we can solve the problem that way, it will be interesting to know what is the exact cause of the problem, and why there are some components on Esteban bridge which are not on your bridge. In my opinion, there is NO need to have ANY component on the 3 wires: B bridge, 0 left and 0 right !

Let's wait the Esteban results..

Regards,

Fred


Hi Fabrice/Fred

Let's start with Fabrice's post.

What I stated in post #35 is correct:

"Circle1"="brown wire (ribbon)"= "Pin#1" of white plug (leftmost one)...... But that is NOT "0 Left" !, it is "0 Right"!! (Pin#1 is "0 Right", Pin#3 is "0 Left" and Pin#4 is "B".......

So I just tested again as asked, and it is the opposite as Fabrice says: "0 Left" goes to orange (Circle #2), and "0 Right" goes to brown (Circle #1)


Now regarding the test requested by Fred

Unfortunately it didn't work...

I tested with all of the capacitors, and the results are all the same:

LIT with and without Loco when facing tracks
Works fine when not facing tracks.

Crying Cursing Crying Cursing
Regards,
Esteban
Offline efel  
#61 Posted : 12 June 2015 05:20:07(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
...
Now regarding the test requested by Fred

Unfortunately it didn't work...

I tested with all of the capacitors, and the results are all the same:

LIT with and without Loco when facing tracks
Works fine when not facing tracks.

Crying Cursing Crying Cursing
Regards,
Esteban

Hi Esteban,

I must say I am lost....
For me, it was not working in the test 5 conditions, i.e. when the bridge is not facing any track and "0 left" and "0 right" connected together to 0. And all my proposals were aimed to correct that malfunctioning. Now, if test 5 was OK, it's another story.
Sorry if I misunderstood the problem.

regards

Fred

Offline ecastrog  
#62 Posted : 12 June 2015 05:57:30(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: efel Go to Quoted Post
Originally Posted by: ecastrog Go to Quoted Post
...
Now regarding the test requested by Fred

Unfortunately it didn't work...

I tested with all of the capacitors, and the results are all the same:

LIT with and without Loco when facing tracks
Works fine when not facing tracks.

Crying Cursing Crying Cursing
Regards,
Esteban

Hi Esteban,

I must say I am lost....
For me, it was not working in the test 5 conditions, i.e. when the bridge is not facing any track and "0 left" and "0 right" connected together to 0. And all my proposals were aimed to correct that malfunctioning. Now, if test 5 was OK, it's another story.
Sorry if I misunderstood the problem.

regards

Fred


Hi Fred

I am sorry I think I didn't make myself clear.

The results are like this (with all the capacitors...)

1. Bridge NOT facing any track:

1.1. Only "0 Left" connected:

NO LOCO on Bridge: NO LIT (OK)
LOCO on Bridge: LIT (OK)

1.2. "0 Left" AND "0 Right" connected:

NO LOCO on Bridge: LIT (Incorrect)
LOCO on Bridge: LIT (Correct)

2. Bridge FACING entry track:

2.1. Only "0 Left" connected:

NO LOCO on Bridge: LIT (Incorrect)
LOCO on Bridge: LIT (Correct)

2.2. "0 Left" and "0 Right" connected:

NO LOCO on Bridge: LIT (Incorrect)
LOCO on Bridge: LIT (Correct)

So you are right about what you thought: Not working when NOT facing any track, but both "0" connected

I've made so many tests that I am confused sometimes....sorry about that!!

Thank you again

Regards,
Esteban
Offline French_Fabrice  
#63 Posted : 12 June 2015 08:17:07(UTC)
French_Fabrice

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

This hardware is a real nightmare to understand...Cursing Cursing Cursing

OK, so the "318" capacitor is connected on one side to "0 right" and on the other side to an unknown destination.

The next idea, but I don't know if it will make progress us, is to measure the value of the current at the 5233 when "normal" detection occurs, AND when "NOT normal" detection occurs.

@Fred: Tell us if this test is useful or not, and if the multimeter will not interfere with the behavior of 5233

Step 15:
-bridge facing no track
-fake board
-multimeter set to measure "Amps -in the milli range -20mA"
-B & 0 of 5233 connected to B & 0 of CS2
-0 left connected to 0 CS2
-one probe of multimeter to "B" bridge
-Other probe of multimeter to line 1 of 5233 (use a female socket exiting from a wire connected to line 1 of 5233)

Test 15.1:
-power on
-measure current with no loco on bridge
-measure current with loco on on bridge
-power off

Test 15.2:
-remove 0 left and plug 0 right to ground
-power on
-measure current with no loco on bridge
-measure current with loco on on bridge
-power off

Cheers
Fabrice
Offline efel  
#64 Posted : 12 June 2015 09:48:06(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
...
The results are like this (with all the capacitors...)

1. Bridge NOT facing any track:

1.1. Only "0 Left" connected:

NO LOCO on Bridge: NO LIT (OK)
LOCO on Bridge: LIT (OK)

1.2. "0 Left" AND "0 Right" connected:

NO LOCO on Bridge: LIT (Incorrect)
LOCO on Bridge: LIT (Correct)

2....
So you are right about what you thought: Not working when NOT facing any track, but both "0" connected

....


Hi,
That means that the parasitic current is quite large!
If I understand well, the B bridge is the center screw.
Could you try to check (visually) if there are wires/components connected to that B wire on the TT (both in the mobile part and static part)?
Could you measure the resistance between the B bridge (yellow wire of the ribbon, I think) and the leg of the "318" "capa" which is not connected to the brown wire? (Ohmmeter on 20 kOhm, TT without any external connection)? Make 2 measurements, inverting black and red probes.

Fabrice,
The current is digital and too high frequency for the Esteban'multimeter intended for sinusoidal 50/60 Hz current: I guess the result will be difficult to interpret.

Regards,

fred
Offline French_Fabrice  
#65 Posted : 12 June 2015 14:08:58(UTC)
French_Fabrice

France   
Joined: 16/05/2011(UTC)
Posts: 1,474
Location: Lyon, France
Originally Posted by: efel Go to Quoted Post

...
Fabrice,
The current is digital and too high frequency for the Esteban'multimeter intended for sinusoidal 50/60 Hz current: I guess the result will be difficult to interpret.

Regards,

fred


ok Fred, Thanks !
@Esteban: don't do Step 15

Cheers
Fabrice
Offline ecastrog  
#66 Posted : 12 June 2015 17:41:39(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Hi Fred/Fabrice

I did the test requested.

The results are the same in both ways (probe red and black in one side and the other):

When I touch both wires with the probes, the multimeter gives a lecture (around 7.9 - 8.0) and then starts to "go down" until it reaches 0.01

I did it quite a few times, trying not to move the probes (I thought maybe I was doing a bad contact), but the results are the same.

Regarding your question of any connections to the "B", I cannot see anything. Maybe something hidden underneath that plate that can't be taken apart....

Actually, "B", as you say is the screw, it doesn't have any other wire connected to it (visible)


Hope it gives some clue

Regards,
Esteban
Offline efel  
#67 Posted : 12 June 2015 21:20:06(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: efel Go to Quoted Post

Could you measure the resistance between the B bridge (yellow wire of the ribbon, I think) and the leg of the "318" "capa" which is not connected to the brown wire? (Ohmmeter on 20 kOhm, TT without any external connection)? Make 2 measurements, inverting black and red probes.

Originally Posted by: ecastrog Go to Quoted Post
..
The results are the same in both ways (probe red and black in one side and the other):

When I touch both wires with the probes, the multimeter gives a lecture (around 7.9 - 8.0) and then starts to "go down" until it reaches 0.01

I did it quite a few times, trying not to move the probes (I thought maybe I was doing a bad contact), but the results are the same.

...

Hi,

Strange behavior indeed! It's the contrary that what happenned when connecting an Ohmmeter to a large capacitor: The reading is very low at the beginning, and increases up to overtake the range of the ohmmeter! The kind of behavior you describe may happen when there is a voltage somewhere in the circuit. You confirm you made the test with no external connection to the TT at all ?

Anyway, your test means there is an electrical path between B bridge and one leg of the "318 cap", the other leg of the "318 capa" being connected to the "0 right", i.e. a path between B and "0 right", leading to a current detected by the 5233.
I don't know what to do, for we don't know what is exactly that path, and what it is for???
For me, the central rail of the track (B) and the 2 lateral rails of the track should go to the corresponding plugs of the fake board without any electrical derivation (as on the Fabrice'TT)!

There is still a possibility; it's to disconnect the "318capa" from "0 right". But that should be avoided if the TT is still under guarantee!

Another possibility is to use the standard marklin occupation detection, between the 2 lateral rails, instead of a detection by current consumption.

It would be also interesting to create a new topic to ask members if their Marklin TT wiring looks like yours or like Fabrice' one .

And why not ask to Marklin, or your dealer, what is the purpose of the components connected on "0 right" or B bridge?

Regards,

Fred
Offline ecastrog  
#68 Posted : 13 June 2015 04:00:01(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: efel Go to Quoted Post

Hi,

Strange behavior indeed! It's the contrary that what happenned when connecting an Ohmmeter to a large capacitor: The reading is very low at the beginning, and increases up to overtake the range of the ohmmeter! The kind of behavior you describe may happen when there is a voltage somewhere in the circuit. You confirm you made the test with no external connection to the TT at all ?

Anyway, your test means there is an electrical path between B bridge and one leg of the "318 cap", the other leg of the "318 capa" being connected to the "0 right", i.e. a path between B and "0 right", leading to a current detected by the 5233.
I don't know what to do, for we don't know what is exactly that path, and what it is for???
For me, the central rail of the track (B) and the 2 lateral rails of the track should go to the corresponding plugs of the fake board without any electrical derivation (as on the Fabrice'TT)!

There is still a possibility; it's to disconnect the "318capa" from "0 right". But that should be avoided if the TT is still under guarantee!

Another possibility is to use the standard marklin occupation detection, between the 2 lateral rails, instead of a detection by current consumption.

It would be also interesting to create a new topic to ask members if their Marklin TT wiring looks like yours or like Fabrice' one .

And why not ask to Marklin, or your dealer, what is the purpose of the components connected on "0 right" or B bridge?

Regards,

Fred


Hi Fred

Yes, there was no external connection at all to the TT (Actually the TT was taken out of the layout!!)

I would like to follow in any of the two ways you suggest. Actually if I can go with standard occupation detection (I still have some addresses available in one s88), that would be the best for me, but I really don't know how to do it, and how to isolate the lateral rails in the bridge!! (I also want to use the 7687, because I am using automatic driving software, basically the beginning of all of this........)
If you can be so kind to give me advise on how to connect things to get the standard occupation detection, via s88, that would be more than appreciated!!Blink

I can also go ahead and remove that "318 capa", actually the TT is no longer under warranty, so it wouldn't matter. The thing is that I would like to know, that even if the detection via the 5233 doesn't work, the TT would still need to be operative for everything else (including the connection to 7687) after removing that component. What do you think?

I like your idea of creating a new topic so more people can discuss this issue. I think I will after this is over....

Best regards,
Esteban
Offline French_Fabrice  
#69 Posted : 13 June 2015 09:43:42(UTC)
French_Fabrice

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

Beware, I'm now speaking about S88 detection, no more 5233 detection !

To answer one of your question of your last post, it's very easy to insulate the 2 "0" tracks of the bridge, from the rest of the layout. The answer has been provided by Ross in the document "7686_turntable_programming": Use some insulating rail joiners Peco SL-11 on all the spoke tracks.

The 2 "0" rails of the bridge are not insulated, but the spoke tracks are (will be). In such a case, any position of the bridge (facing or not facing tracks) will NOT get the 0 from external tracks connected to the TT, but only from the "0 left & 0 right bridge" plugged into the 7687/fake board. You only have to put a S88 line on one of these "0" plugs and that's it.

Here is the schema of the first simple test you must do, before going further in the S88 detection way:
UserPostedImage

The purpose of this schema is to test that S88 detection runs well, on the bridge ONLY (you don't need at this time the SL11 insulators).

I prefer to be very cautious with your TT, because I'm still not able to determine if your TT is defective or not (maybe by design), about the bridge occupation detection. You must also remember that the main goal of Marklin is NOT to allow occupation detection on the bridge, but to provide a device that allows the bridge to rotate and been driven from the CS2 (or manually). So, if this new design forbids current occupation detection, it is still not proven that your TT allows standard S88 occupation detection on the bridge... This schema allows you to get the proof.

A few warnings:
-As previously stated by Ross, you must have only one "0" common line between all your power sources
UserPostedImage
-Please check your chain of S88 devices: The first one (and only the first one) must have the "T" plug connected to common ground.

Then you can do the test shown in my first picture.
-Do the test as shown. You don't need a loco on the bridge at first. A simple coach/wagon (with conducting axles ! i.e. designed for 3 rails) is enough. Wagon on bridge=LIT, nothing on bridge = NO LIT. Do also the same test with a loco...
-2nd test: first swap the 0 left and right, i.e. 0 left is connected to S88 line 1, and 0 right is connected to common ground. Second, do the previous test.

If all tests are successful, we may go further in the S88 way.

Cheers
Fabrice
PS: If you wish to open a new thread dedicated to detection occupation on bridge and/or using a TT, please do. Or, you may go on in this current thread, it's up to you.
thanks 1 user liked this useful post by French_Fabrice
Offline efel  
#70 Posted : 13 June 2015 10:45:32(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
..

I can also go ahead and remove that "318 capa", actually the TT is no longer under warranty, so it wouldn't matter. The thing is that I would like to know, that even if the detection via the 5233 doesn't work, the TT would still need to be operative for everything else (including the connection to 7687) after removing that component. What do you think?

..

Hi Esteban,

I can't tell what will happen without that "318 capa" for I don't know if there are other added components, where they are connected, and what is their purpose. In fact, I'm not even sure the "318" is a capacitor for it's not a standard marking!
I was just suggesting to try and see. If you were to do the trial, desolder only 1 leg of the capa (the one connected to the brown wire) and insulate that leg with electrical tape. That way, you can resolder it easily if needed.

Regards,

Fred






Offline ecastrog  
#71 Posted : 13 June 2015 19:54:55(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: French_Fabrice Go to Quoted Post
Hi Esteban,

Beware, I'm now speaking about S88 detection, no more 5233 detection !

To answer one of your question of your last post, it's very easy to insulate the 2 "0" tracks of the bridge, from the rest of the layout. The answer has been provided by Ross in the document "7686_turntable_programming": Use some insulating rail joiners Peco SL-11 on all the spoke tracks.

The 2 "0" rails of the bridge are not insulated, but the spoke tracks are (will be). In such a case, any position of the bridge (facing or not facing tracks) will NOT get the 0 from external tracks connected to the TT, but only from the "0 left & 0 right bridge" plugged into the 7687/fake board. You only have to put a S88 line on one of these "0" plugs and that's it.

....

The purpose of this schema is to test that S88 detection runs well, on the bridge ONLY (you don't need at this time the SL11 insulators).

I prefer to be very cautious with your TT, because I'm still not able to determine if your TT is defective or not (maybe by design), about the bridge occupation detection. You must also remember that the main goal of Marklin is NOT to allow occupation detection on the bridge, but to provide a device that allows the bridge to rotate and been driven from the CS2 (or manually). So, if this new design forbids current occupation detection, it is still not proven that your TT allows standard S88 occupation detection on the bridge... This schema allows you to get the proof.

....

....

If all tests are successful, we may go further in the S88 way.

Cheers
Fabrice
PS: If you wish to open a new thread dedicated to detection occupation on bridge and/or using a TT, please do. Or, you may go on in this current thread, it's up to you.


Hi Fabrice

I tried your schema and it works perfect!! BigGrin BigGrin

I do not have the rail joiner insulators, I will have to order them, so I can't connect everything now. But the test was successful!! (bridge facing tracks, not facing tracks, with loco, no loco, etc...) the occupancy detection worked fine!!

I also already connected the common ground as you showed me.

I will let you know when I get the insulators, and when everything is working fine.

Too bad we never found out the "mystery" of the parasitic current, but at the end, what I needed, that was the occupancy detection was achieved!

I really want to thank you, @Fred and @Ross for all of your kindness and helpfulness!!! You are really nice people.

I hope I can be helpful to you sometime also.

I will send pictures of my layout to you soon

Regards!!

Esteban
thanks 1 user liked this useful post by ecastrog
Offline French_Fabrice  
#72 Posted : 14 June 2015 00:30:46(UTC)
French_Fabrice

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

I'm really glad you succeed. Congratulations !

With the schema below:
UserPostedImage

...you'll be able to completely wire your TT for automatic processing.

Here are a few explanations and some suggestions:

-This schema uses S88 standard (towards ground) detection for all tracks (bridge included).

-The set of diodes is "the diode trick". It enhances ground conductivity when using locos with small number of axles. The use of diodes is not mandatory, it is only advised. Beware with the diodes, they have a "direction". This direction is shown on the electronic component with a grey or black circle on one side. I've shown it on the diagram (diode 1N400x). Use diodes from the family 1N400x with 1 <= x <= 7 (very common component). Download the doc about the diode trick from Ross web site ( http://members.ozemail.c.../pdf/the_diode_trick.pdf ) and read it, it is very well explained.

-About Peco SL-11 insulators: It's not very expensive (3.5 € for a small bag of a dozen pieces). As the shipping cost to Ecuador may be expensive (from US of Europe), order a few bags to have some spare parts at hand. Peco SL-11 are designed for "Code 100" tracks. It fits well in the spoke track of the TT (just did the test this afternoon). You have to remove the metallic rail joiner of a spoke track with a small flat screwdriver, and replace it with a SL-11 (beware not to hurt you).

-Some problem may arise with the second metallic rail joiner you have to replace with a SL11 (the one fixed on the C-K transition track, at the "K" side) : C-tracks are "Code 90" tracks and I don't have C-track nor any expertise on C-track (I use only K tracks). You have to decide, by first trying to insert a SL-11 on the K-side of the transition rail if it fits well or not. If it fits well, you have won, because you can insulate the 2 "0" rails. If not, you may have to order some K-tracks (2201 = 1/2 straight track = 9cm), and interpose a 2201 between the transition C-K track and the spoke track, then remove from the K-track the metallic rail joiner and replace it with a SL-11. Have also spare transition tracks, just in case...

-Finally, do not plug all the insulated tracks at the same time, then provide power. I strongly advise you to go step by step:
a) insulate the entry track, connect it to the TT (insulated) and S88 and test entry track detection and bridge (facing entry track and facing NO track)
b) add the 1st exit track, and repeat same test; all sections must be insulated from each other; also do a test with a rotation of 180° of the bridge, just to be sure there is no bad insulation.
c) then finally add remaining exit tracks and test again

I wish you good luck, and if needed, ask for help or clarifications.

Cheers
Fabrice
thanks 1 user liked this useful post by French_Fabrice
Offline efel  
#73 Posted : 14 June 2015 07:32:40(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Hi all,

The problem has already been solved: here.

The "318" component purpose would be to suppress radio interference! Removing it solve the problem.

Regards

Fred
thanks 2 users liked this useful post by efel
Offline ecastrog  
#74 Posted : 15 June 2015 04:22:31(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: French_Fabrice Go to Quoted Post
Well Esteban,

I'm really glad you succeed. Congratulations !

With the schema below:

....

...you'll be able to completely wire your TT for automatic processing.

....

I wish you good luck, and if needed, ask for help or clarifications.

Cheers
Fabrice



Hi Fabrice!!

Thank you very much for your (usual) clear, detailed explanation!

I hope I can get those insulators soon, so I can finish arranging that part of my layout. Meanwhile I will work in some other things

Just wanted to say thank you again for all your help and kindness!BigGrin BigGrin

Regards,
Esteban
thanks 1 user liked this useful post by ecastrog
Offline ecastrog  
#75 Posted : 15 June 2015 04:27:17(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Originally Posted by: efel Go to Quoted Post
Hi all,

The problem has already been solved: here.

The "318" component purpose would be to suppress radio interference! Removing it solve the problem.

Regards

Fred


Hi Fred

As you may have seen, I also found this post, and wrote to the author because the pictures are no longer available. So I was hoping to get an answer from him.

As you know, I decided for standard (via s88) detection, and already worked. But I was really curious and today I removed (desoldered one leg...the one going to the brown wire "0 Right"), and guess what..... it didn't work!!

There was no power in the TT when I removed this leg....

Maybe I did it wrong? Did I have to remove both legs? What may have I missed?

Anyway, I soldered it again, so at the end the TT is functional again

If I can solve this, maybe I will be using this method (via 5233), because as you know, I don't have the insulators for doing it the other way..(yet..)

Thank you for all of your nice help!!!BigGrin

Best regards

Esteban

thanks 1 user liked this useful post by ecastrog
Offline efel  
#76 Posted : 15 June 2015 07:28:23(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Originally Posted by: ecastrog Go to Quoted Post
Originally Posted by: efel Go to Quoted Post
Hi all,

The problem has already been solved: here.

The "318" component purpose would be to suppress radio interference! Removing it solve the problem.

Regards

Fred


Hi Fred

As you may have seen, I also found this post, and wrote to the author because the pictures are no longer available. So I was hoping to get an answer from him.

As you know, I decided for standard (via s88) detection, and already worked. But I was really curious and today I removed (desoldered one leg...the one going to the brown wire "0 Right"), and guess what..... it didn't work!!

There was no power in the TT when I removed this leg....

Maybe I did it wrong? Did I have to remove both legs? What may have I missed?

Anyway, I soldered it again, so at the end the TT is functional again

If I can solve this, maybe I will be using this method (via 5233), because as you know, I don't have the insulators for doing it the other way..(yet..)

Thank you for all of your nice help!!!BigGrin

Best regards

Esteban



Hi Esteban,

So, as there is no power after removing the "318" component, that means that component is not a capacitor.
It seems that Jonas-sthlm had succeeded in doing so, but may be he was no talking about the same component.
I think I can't help more, for we don't know the whole electrical diagram, and trials "at random" lead nowhere. Fortunately, you can solve your problem using lateral rails detection.
Anyway, I thank you for your trial: even if the result is not positive, it increases the global knowledge of the forum.

Regards

Fred
Offline French_Fabrice  
#77 Posted : 20 June 2015 16:04:02(UTC)
French_Fabrice

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

I think I've found the reason why the 5233 doesn't run well with new TTs: See there (1st picture): http://www.sb-modellbau....20Lokpilot%20Decoder.pdf

@Fred: You were right, there is an other device connected to yellow wire -"B bridge" (looks like a capacitor, but I'm not sure as connectors are not radial -may be a coil) -Yes, after passing the german sentence to the translator, it's a coil !

So, capacitor and coil must be removed for the 5233 to operate correctly!

Cheers
Fabrice
Offline efel  
#78 Posted : 21 June 2015 11:56:37(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Thanks for your follow-up, Fabrice.
I must say I am not yet understanding why it did not work when deconnecting one leg of the capa.

After your link, keeping that coil and capa may lead also to problems with the loco decoder !

Regards,

fred
Offline ecastrog  
#79 Posted : 22 June 2015 18:16:56(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Hi Fabrice/ Fred

Thank you Fabrice for the information, very interesting.....

In my case, I think I will go first with the s88 detection (after I get the insulators). Only if this doesn't work (why shouldn't??), then I would move to remove those components....

I think this will be useful for other forum members anyway, and for "solving the mystery"!!Blink

Best regards,
Esteban
Offline ecastrog  
#80 Posted : 05 July 2015 05:27:42(UTC)
ecastrog

Ecuador   
Joined: 05/06/2014(UTC)
Posts: 44
Location: Quito
Hi Fabrice / Fred / All

Just wanted to let you know that I got the insulated rail joiners, and after doing all the necessary connections, I can tell you that:

1. The TT is working fine. With the 7687
2. The loco detection is working!! (Both on the bridge and on the side tracks)
3. I use iTrain and everything is working perfect between the layout (TT) / CS2 / software !

BigGrin BigGrin BigGrin

So, after all this time, everything is working fine!!

Thank you again for all your help. I will send you some pictures (or maybe a video) of my layout soon!

Best,

Esteban
thanks 1 user liked this useful post by ecastrog
Offline efel  
#81 Posted : 07 July 2015 13:25:49(UTC)
efel

France   
Joined: 23/02/2005(UTC)
Posts: 800
Glad to hear that!
Thanks for the feedback.
Fred
Offline French_Fabrice  
#82 Posted : 14 May 2017 19:36:56(UTC)
French_Fabrice

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

Time has passed since my last writing on this topic.
A few things have changed, even if the principles of use are still the same as the ones described in post #1.

1. My layout Trossingen has been replaced by Trossingen II, which is larger. Regarding the TT area, there are still one entry track and six spoke tracks.
2. I've replaced the TT decoder (old Marklin ref:7686) by a LDT TT-DEC decoder. The reason was due to a malfunction when the bridge entered a half-turn (180° rotation). Approximately one time over 2, the bridge didn't rotate entirely successfully, and this ruined the automatic sequences and I had also to reprogram the decoder... Does the LDT TT-DEC solved that issue ? The answer is YES ! BigGrin
3. I've progressed in automating the TT area. As I have only one sensor for each block involved (entry block, bridge, spoke tracks), I dug in the Rocrail docs and done many experimentation with the various timers, to find an acceptable solution whatever the steamer entering this area... There are still a bit of tweaking to be done, but the principles are now known.

Reminder and Warning: What is described below involves a turntable dated from 1990-1995 (I don't remember exactly the precise date of purchase). With such a turntable, CURRENT CONSUMPTION DETECTION works for all tracks by only removing the "B" tongue of the spoke tracks, as described in post #1. If you have a more recent TT, current consumption detection may NOT work (See Esteban's long story in previous posts) !!! due to additional electronic components (capacitor and coil) in the TT base.

Here is a shot of the real area:
UserPostedImage

Replacing the old Marklin 7686 decoder by a LDT TT-DEC decoder has raised an issue, regarding the bridge occupation feedback when using the current detection method. I'm going to provide you the solution if you wish to do such a thing.

Reminder: the LDT TT-DEC provides 2 feedback information: "bridge occupied", and "target position of the bridge reached". The "(target) position (of the bridge) reached" feedback uses a standard ground detection, i.e. standard S88 or LDT RM-88-N or equivalent. I'm going to discuss only the "bridge occupied" feedback, using current consumption detection.

Reminder: These 2 feedback information are mandatory if you wish to achieve automation in the TT area (+ of course feedback from the entry track and the various spoke tracks)

LDT site is very well documented and provide many examples for wiring their various devices. Unfortunately, my case was not (completely) described. After some investigations on various schema, the schema "page 1673" http://www.ldt-infocente...sbeispiele/page_1673.pdf has led me to the solution:
The brown and orange wires of the TT ribbon cable are respectively rail 1 and rail 2 (0 line). The yellow wire is the B line.

As a result, I only need to pick the yellow wire from the TT ribbon and plug it into the current consumption feedback detector.

The wiring schema for such a thing is finally very easy:

UserPostedImage

The settings in real life is a bit more complicated Crying . The yellow wire of the TT ribbon cable must be left intact, and plugged into the current feedback detector (LDT RM-GB-8-N or Viessmann 5233 or other equivalent). Then, the 6 pin plug of the LDT TT-DEC must be connected to the TT ribbon, except for the yellow wire. To solve that, I've used:
a) a 0,5m extension cable for S88 bus from LDT
b) a hand made interface, done with a small piece of Veroboard, 2 X 6-pin connectors and 5 straps over 6 connected between these 2 connectors. The remaining wire goes to the current consumption detector. Then, plug the TT ribbon cable on the TT input and the extension cable on the DEC output.

UserPostedImage

Some additional information about MY turntable with the LDT TT-DEC:
After the initial setup of the TT, it worked perfectly !!! except I though the bridge was rotating a bit fast in my opinion (faster compared to the old 7686). I've tried to slow down the rotation speed using the adequate potentiometer. The result was perfect in one direction (clockwise), but in the other direction it was a disaster (not all clicks heared, and the bridge stopped anywhere except in the expected position Cursing ). Finally, I went back to the initial settings and left this higher (but acceptable) rotation speed and now the TT works perfectly 100% of the time.BigGrin

I'll discuss later the various tweaking of Rocrail and provide more examples for automating the TT area. To summarize, the most critical part involved in automatic sequences in the TT area is the ENTIRE placement of the steamer ON the bridge, before the rotation starts. To achieve a good placement, a loco must have set its own timer for this block (the TTBridge block). The various parameters of Rocrail for a loco object provides such a timer. This timer is used by default if it has a non 0 value, instead of the block timer.
Each loco going in this area will have a different timer value, depending of the mechanic, the loco decoder tweaking, the "slow speed" value, and the position of the slider when using current consumption detection !

Stay tuned for more information...
Cheers
Fabrice



thanks 4 users liked this useful post by French_Fabrice
Offline applor  
#83 Posted : 06 April 2018 08:08:17(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,652
Location: Brisbane, Queensland
Rather than isolating every stub track, could you not remove the rail contact strips on the turntable - or are they used for position detection?

edit: never mind, I found my answer! Yes you can cut/remove the contact tab for the rail used for detection as it not used for position detection.
This thread lead me to this excellent analysis of how the turntable operates.
It also shows how to dis-assemble the turntable, which I realise would be best practice when painting/weathering.

edit2: reading that guide it is advised to bend/remove the rail contact tabs from the turntable anyways, to avoid possible short circuits with the centre conductor as the turntable rotates through the positions - since we manually run power feeds to the spur tracks in a digital environment.
This method saves isolating every spur track from the turntable by just isolating the two turntable rails.
modelling era IIIa (1951-1955) Germany
thanks 1 user liked this useful post by applor
Offline applor  
#84 Posted : 24 September 2018 04:28:23(UTC)
applor

Australia   
Joined: 21/05/2004(UTC)
Posts: 1,652
Location: Brisbane, Queensland
When testing my BR64 I found it makes a slight buzzing sound which becomes louder when on the turntable.
I decided to remove the capacitor and inductor from my turntable.

It was a royal pain in the ass to do since my turntable is installed, even with the centre being accessible.

To access the inductor you need to remove the capacitor, de-solder all the wires and then remove the plastic cover.
Here is that work performed with the inductor now exposed:

IMG_2740.JPG

After removing the inductor and re-soldering the wires (now with yellow on the centre pin which was previously soldered to the inductor leg) here is what it looks like:

IMG_2742.JPG

After re-testing I found no change with the buzzing noise of the BR64 so I am not sure if this work has helped me at all.
I will be using S88 detection for my turntable, rather than current detection which is what other users had problems with.

Edit: I also cut off the centre strip 'tabs' that stick out from the turntable that provide power to the spur tracks in analogue operation.
Since I run digital it was not required and also caused problems as it could occasionally make contact with the rails during rotation, resulting in a short.
modelling era IIIa (1951-1955) Germany
Users browsing this topic
2 Pages<12
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 1.429 seconds.