Keyboard Shortcuts
Likes
Search
Disconnect between signal aspect showing on JMRI and ACTUAL signal aspect on layout
#nce
Hi JMRI user group, This little black duck in St. Clair, Sydney, finally got some signalling up & running, sort of... Is there any known issue[s] with JMRI and NCE that would cause JMRI to correctly show? the aspect? of a given [3 aspect] signal, BUT have the equivalent physical signal on my layout inconsistently update [or not] its aspect?? Note I said 'inconsistently', because some of the time a given signal will update correctly, another time not. [And it is not always the same signal that fails to update; it all appears rather random.] ? SOME DETAILS Simple loop running at present under test. Approximately 15 signals in the? large-ish loop. Currently running LOGIX programmed to have one signal RED, one signal AMBER, the remainder GREEN. Logix operates correctly 'on screen' or 'virtually'. I have tried various ways of programming the signal sequence, but always run into the same problem of inconsistent updating of aspects. ? HARDWARE - all NCE NCE PowerPro, NCE BD20 sensors [all track 'sensed'], NCE AIUs, their channels switched via the sensors. Each signal is connected to an NCE 'Light-it' stationary decoder acting as its driver. ? ADDITIONAL OBSERVATIONS Signals, when manually operated from the PC [JMRI signal table] update as directed 'on screen' & 'on layout', no failures. AIU on board channel indicators [LEDs] operate correctly and cause JMRI sensor table to shown sensors as 'active' or 'inactive' as expected. At one point I thought there may be data loss occurring through the 'bus' between the NCE system and the signal drivers, but this doesn't appear to be the case. I say this because I have one of the signals connected to the 'track bus' within 6 inches of the NCE box - and it still fails to update at times. I can't take the project any further until reliability of correct aspect be attained. ? I can supply more detailed info to anyone who thinks they can help. Regards, Nimble |
开云体育Nimble, ? Only think I’ve heard that might fit this is that if the commands to the Light-It come too fast, it will see the first command and miss the second command. I’ve not seen it myself but have heard it happen to so. There were comments, either here or on the NCE list, and I recall some solution was found. Around six months ago, I think. ? -Ken Cameron, Member JMRI Dev Team
? ? |
开云体育Hi Ken, Thanks for taking the time to reply to my JMRI user group request.? What you say appears to be the case: for when a train moves from one sensor [track block] to the next, often several 'commands' are issued by the JMRI logic almost simultaneously; for example, Signal A goes 'red', signal B goes amber, Sig C goes 'clear'.? Often it is the 'to amber' event that is missed. I've just sent a query to the NCE Information Station. I've spent a fair bit of $ and time setting all this up only to run into this issue.? On 23/04/2025 9:14 pm, Ken Cameron
wrote:
|
开云体育Say Ken, I'm not sure what you mean by 'NCE list'? Regards, Nimble
On 23/04/2025 9:14 pm, Ken Cameron
wrote:
|
On Wed, Apr 23, 2025 at 08:46 PM, Nimble Knumblebum wrote:
/g/NCE-DCC -- Peter Ulvestad Linux Mint 22.1, JMRI 5.11.5, Java 21.0.6 JMRI Users Group Moderator ( /g/jmriusers ) JMRI Developers Group Moderator ( ) Tam Valley Group Moderator ( ) Sprog-DCC Group Moderator ( ) Edmonton Model Railroad Association ( ) |