DC Question

after replacing the diode the voltage became normal

That proves the original item is f****ed, and needed replacement. It doesn't answer <...why this...happen?...>.

The leading causes of death for diodes are excessive reverse voltage and excessive forward current. Probably one or the other killed it.

As an aside issue, do they fail in open-circuit or short-circuit mode, or is it one or the other, depending on polarity causing the failure?

The <...negative side diode was damaged...> because either:

1. It was unable to sustain a forward current that may well have been in excess of its rating, or
2. It was unable to withstand a reverse voltage that may well have been in excess of its rating, or
3. Some sort of mechanical impingement that has had a diminishing effect on its original rating, or
4. Some combination of the above.

As to <...why this situation happen?...>, only the Original Poster is in any position to investigate and discover why, at the moment, however there are other options, one of which involves a quotation, an acceptance, a valid purchase order, and a site visit to carry out the investigation.

Voltage spike....

Perhaps you measure the ac component of the voltage because the diode was short-circuit?

Without a simple circuit diagram of your components and where you make measurements, it is very difficult to imagine even what you are experiencing...

we might guess a full wave rectifier 2 diodes and center tap transformer?

Dear Boss,

this diode is isolation between 2 DC source and 2 different rectifier

Voltage measured where? - relative to what ?

Such as ground/earth, or maybe your supply or supplies are floating?

The diode that failed is a steering diode or one of the rectifiers...?

the full sitation is

the entire power plant was tripped due to the trip in the MV switchgear from the protection relay ABB ref542 plus due to external trip received from field that was not correct because the external contact was open and in-same time when we remove the trip coil negative for the breaker (ABB vacuum breaker) the breaker is also receive trip signal from relay so we forced to remove the positive and negative contact

For your kind information

We found one diode (two diode one fix in positive and the other is connected to negative to make isolation between the other DC supply from the other busbar and this diodes fixed in the measurement cabinet) in one block in the way of the dc was short circuited and the –ve to ground was +154 V and +ve to ground was 288 V and the sum of voltage between +ve to –ve is 134 Vdc

Also in other block there was +ve ground fault +ve to ground was 0 V and –ve to ground was -134 Vdc from solid connection between the +ve to ground from the flap switch

In the same time plant blocks DC connected to gather so the DC voltage was shifted 154 V, that mean the DC in the plane was -ve to ground +154 and +ve to ground is 288 V and +_ve to -Ve is 134 V

For your kind information our DC system is not grounded

Floating DC system

If it is truly a <...Floating DC system...> then the voltage to earth at any place in the circuitry is somewhere in the range between [irrelevant] and [misleading].

In a floating (ungrounded) DC system, the measurement of voltage from + and - to ground really means nothing, except when you are looking for whether one or the other has become grounded. Usually you just want to be sure you have the full DC voltage from + to - (in your case, 134V). A single ground fault condition should cause an alarm, so that it can be investigated and corrected before a second ground fault actually causes your DC to be failed and unavailable.

Usually, when you measure + and - from an ungrounded DC system to ground, you tend to see the measurements sort of evenly split, but not exactly. It is minor leakage through the insulation that gives you a "measurement."

If it is true that one of your DC systems had a ground fault, and the two DC systems were interconnected via the diodes, that could have caused a failure of the diode in some way. However, it just could have been a random failure of the diode, which then showed itself through the change in relative "reference" voltages of your DC rails to ground. Was there some unusual loading or stress on the DC system at that time? You did not specify if the trip of your plant circuit breaker was due to an actual fault condition for which it should have tripped.

Either way, I am wondering why you have two DC systems tied together in this way?

Mr. Peter,

not only two busbar but it was 6 busbar each one have a battery charger and battery bank, 2 busbar for each block and also four bus bar for common service that interconnected with the 6 busbar, so it seems to be all system have the same DC voltage and as discuses the signal for external trip is came for 100 ms and then gone again and again the problem was initiated in two busbar and then transfer to the all system one busbar solid positive ground and the other one diode failure +ve to ground 288 and -ve to ground +154

So why not state all that in the original thread to give forum subscribers a better starting point? All that omitting it does, is waste time.

The cause of the failure cannot be determined from the information disclosed to the forum. A site investigation on the basis of historical data might reveal it. For example:

• there has been no information on the weather at the location of the installation, nor
• the ingress protection rating of the enclosure housing these <...diodes...>, nor
• detail of any planned or unplanned maintenance work taking place around the time the failure occurred that might have contributed to the failure.

All of these things are relevant, and in the absence of a purchase order to carry out investigation work on this Client's behalf, <unsubscribes>.