Facebook
Google
GitHub
Linkedin
C
It's presumed you're referring to Diagnostic Alarm 'P125-N125 Overvoltage'. Diagnostic Alarms are usually associated with a single processor; is it <S> that's annunciating this alarm? Is it being annunciated by more than one processor?
What is the DC voltage split with respect to ground of the 125 VDC supplying the Mark V? In other words, is the voltage split fairly equal or is there a slight or even a "hard" ground?
One quick check would be to replace the card; I detest that troubleshooting method but sometimes it's effective. If the alarm is associated with <S> processor, then it would seem plausible that the card is the source of the problem.
The high cranking motor amps alarm and high temp alarms/trips (it's presumed the high temp alarm/trip is from the cranking motor) are kind of odd. How are these signals "communicated" to the Mark V? Via contact outputs from the LodTrak? If so, which contact input module(s) are the alarms connected to: <CD>, or <QD1>, or both? The same for the generator ground alarm: which contact input module is the alarm contact connected to?
For a process alarm which is driven by a contact input to <QD1> to be annunciated, two of the three control processors must detect the change of state of the contact input through each processor's TCDA card. If only two of three detect a change of state, the process alarm will be annunciated and a Voting Mismatch Diag. Alarm will also be annunciated for the single processor which isn't detecting the change of state to annunciate the alarm.
For a process alarm which is driven by a contact input to <CD> to be annunciated, <C>'s TCDA card must send the signal over the DENET to the three control processors to annunciate the change of state.
So, for a process alarm to be annunciated falsely by a contact input connected to <QD1>, it would take two of three TCDA cards to "detect" a change of state. For a process alarm to be annunciated falsely by a contact input connected to <CD>, it would take only the TCDA card in <CD> to "detect" a change of state.
Have you tried using the Contact Input Change Detect feature to log changes of state of the contact inputs in question?
Have you checked for the presence of AC on the contact input wiring or the 125 VDC supply wiring? Strange things can happen when induced AC voltages above approximately 40 VAC are present on the contact input wiring. Specifically, check the wiring associated with the alarms you've mentioned.
Induced AC voltages are the hardest things to resolve; they are usually the legacy of poor design and/or construction practices, routing 125 VDC contact input cabling in close proximity to AC current-carrying conductors. When the cranking motor-related alarms (presuming they come into the panel via contact inputs) or the generator ground alarm (presuming it comes in as a contact input) are annunciated, is there some AC motor or space heater which starts at the same time?
What is the DC voltage split with respect to ground of the 125 VDC supplying the Mark V? In other words, is the voltage split fairly equal or is there a slight or even a "hard" ground?
One quick check would be to replace the card; I detest that troubleshooting method but sometimes it's effective. If the alarm is associated with <S> processor, then it would seem plausible that the card is the source of the problem.
The high cranking motor amps alarm and high temp alarms/trips (it's presumed the high temp alarm/trip is from the cranking motor) are kind of odd. How are these signals "communicated" to the Mark V? Via contact outputs from the LodTrak? If so, which contact input module(s) are the alarms connected to: <CD>, or <QD1>, or both? The same for the generator ground alarm: which contact input module is the alarm contact connected to?
For a process alarm which is driven by a contact input to <QD1> to be annunciated, two of the three control processors must detect the change of state of the contact input through each processor's TCDA card. If only two of three detect a change of state, the process alarm will be annunciated and a Voting Mismatch Diag. Alarm will also be annunciated for the single processor which isn't detecting the change of state to annunciate the alarm.
For a process alarm which is driven by a contact input to <CD> to be annunciated, <C>'s TCDA card must send the signal over the DENET to the three control processors to annunciate the change of state.
So, for a process alarm to be annunciated falsely by a contact input connected to <QD1>, it would take two of three TCDA cards to "detect" a change of state. For a process alarm to be annunciated falsely by a contact input connected to <CD>, it would take only the TCDA card in <CD> to "detect" a change of state.
Have you tried using the Contact Input Change Detect feature to log changes of state of the contact inputs in question?
Have you checked for the presence of AC on the contact input wiring or the 125 VDC supply wiring? Strange things can happen when induced AC voltages above approximately 40 VAC are present on the contact input wiring. Specifically, check the wiring associated with the alarms you've mentioned.
Induced AC voltages are the hardest things to resolve; they are usually the legacy of poor design and/or construction practices, routing 125 VDC contact input cabling in close proximity to AC current-carrying conductors. When the cranking motor-related alarms (presuming they come into the panel via contact inputs) or the generator ground alarm (presuming it comes in as a contact input) are annunciated, is there some AC motor or space heater which starts at the same time?
