Faulty events causes GHOST Train [SOLVED] Topic is solved

Post Reply
handihomann
Posts: 33
Joined: 27.01.2014, 17:20
Location: Marlboro, New Jersey US

Faulty events causes GHOST Train [SOLVED]

Post by handihomann » 07.06.2019, 14:25

Hi All,
Some sensors were triggered caused Ghost train on those blocks. I realized those faulty events were so quick unlike the real events! I know they were faulty events since the ECOS central station did not show these sensors were triggered. Now my question is:
Is there any option to ignore those events which are less than 1/10 second? I'm running out idea! I rewired all those sensors to make sure these wires are NOT parallel to the DCC wires but nothing works.
I'm using the ECOS and S88(reed switches)
Thanks in advance.


Regards
Handi
Attachments
issue.zip
(15.44 KiB) Downloaded 2 times
Last edited by handihomann on 10.06.2019, 20:35, edited 1 time in total.

rjversluis
Site Admin
Posts: 41593
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Faulty events causes GHOST Train

Post by rjversluis » 07.06.2019, 14:34

Hi Handi,
Is there any option to ignore those events which are less than 1/10 second?
No, that should be managed in the underlying hardware.
But IMHO is this caused by interference.
100ms events should be regarded as valid.

BTW: Update your Rocrail.

eroncelli
Posts: 348
Joined: 11.01.2009, 19:09
Location: Bergamo - Italy

Re: Faulty events causes GHOST Train

Post by eroncelli » 07.06.2019, 15:01

A simple calculation:

Real speed of 120 km/h means 33,3 m/s
In model railroad, recommended speed is about "real speed/model scale" (plus a 10-20%) : for H0 scale, it means about 400 mm/s.
The activation area for a reed contact with magnet is about 40 mm (probably less): that means a 100 msec pulse.

As suggested, it looks like an interference issue.

handihomann
Posts: 33
Joined: 27.01.2014, 17:20
Location: Marlboro, New Jersey US

Re: Faulty events causes GHOST Train

Post by handihomann » 10.06.2019, 15:40

rjversluis wrote:
07.06.2019, 14:34
Hi Handi,
Is there any option to ignore those events which are less than 1/10 second?
No, that should be managed in the underlying hardware.
But IMHO is this caused by interference.
100ms events should be regarded as valid.

BTW: Update your Rocrail.
Thank you Rob. Interesting thing I found after re-wiring and updated the software, my new Bemo Glacier Express with Loksound V4.0, is so noisy and interference to my S88 modules.
Thanks again.
Handi

rjversluis
Site Admin
Posts: 41593
Joined: 10.04.2006, 08:48
Location: Speyer, Germany
Contact:

Re: Faulty events causes GHOST Train

Post by rjversluis » 10.06.2019, 15:43

Hi Handi,

just turn RailCom off in the decoder of the Bemo Glacier Express.
S88 is very sensitive for interference, and not state of the art.
Use shielded CAT-5 cable.

handihomann
Posts: 33
Joined: 27.01.2014, 17:20
Location: Marlboro, New Jersey US

Re: Faulty events causes GHOST Train

Post by handihomann » 10.06.2019, 20:27

rjversluis wrote:
10.06.2019, 15:43
Hi Handi,

just turn RailCom off in the decoder of the Bemo Glacier Express.
S88 is very sensitive for interference, and not state of the art.
Use shielded CAT-5 cable.
:thumb_up:
Thank you Rob. It works great by turning the RailCom off.
Yes, I use CAT5

Regards
Handi

RainerK
Moderator
Posts: 3988
Joined: 29.04.2009, 09:31
Location: Sprockhövel (zwischen BO u. W)
Contact:

Re: Faulty events causes GHOST Train

Post by RainerK » 10.06.2019, 23:19

Hi Rob,
rjversluis wrote:
10.06.2019, 15:43
...Use shielded CAT-5 cable.
for S88 are unshielded cables also more then sufficient.
The reasons:
1. each wire of the twisted pairs has an best possible decoupled use in the electrical relation to the others in the S88 connecting scheme.
2. in most cases the CAT 5 sockets of the units have no contact or connection for the shield from the plugs of the cable.
3. the most offered assembled CAT 5 patch cables are from the UTP type (unshielded twisted pair).
4. the better CAT 6 or CAT 7 patch cables have mostly a good shield. Unfortunately they are many more expensive, but without some advantage, see 2.

Post Reply

Return to “Automatic mode”