Speedtronic (MKV and MKVI) gives two ways to check integrity of the shaft voltage monitoring system.
By pressing 'Alarm Test' push button on HMI screen it is injected AC signal (5 volts and 1000Hz) into Speedtronic card what should provoke appearance of the two alarms: 'High shaft voltage' and High shaft current'.
Result of this test is positive if both alarms are displayed.
Result of this test is not affected by the state of the grounding brushes. It verifies only Speedtronic internal monitoring system.
By pressing the other push button located on the HMI screen labelled: 'Sensor Test" Speedtronic injects DC voltage (5V) into rotor grounding brushes. In this test Speedtronic serves as 'Ohmmeter' measuring the resistance of the loop consisting of: cabling from Speedtronic to grounding brushes, resistance of contact between grounding brushes and the surface of the rotor and finally connection of the grounding system to ground. Results of the resistance measurement are displayed on the same screen on the two display windows typically labelled: 'Brush resistance' and 'Shunt Resistance'.
Result of this test is positive if none of the previously mentioned alarms appeared. If it appeared, this indicates the defect in the external (external to Speedtronic) rotor grounding system. Measured resistance below 10 Ohms for each of the two circuits ('brush' and 'shunt') is considered by GE spec as satisfactory: if both conditions are fulfilled it should appear result of the test: 'Test OK' in the next display window on the HMI screen.
Below I collected some remarks.
Rotor grounding brushes rig on 9FA turbine consists of four brushes fixed on the bearing No1 - opposite side to exciter bearing. Exciter side bearing is insulated from ground. Every two of four brushes are connected in parallel thus giving possibility to replace them with turbine online and reducing the resistance of contact between the brush and rotor's surface.
The conductor that goes from the two brushes (connected in parallel) to solid grounding system goes through the shunt resistor located on the brush rig. With properly maintained generator/turbine unit there is only one point where generator's rotor is connected to ground: it is grounded through these two brushes. In case that exciter's bearing gets the contact to ground, the loop for electrical current circulation will be created and current will pass through the grounding brushes and through the shunt resistor to ground. Drop of voltage from this shunt is directed to Speedtronic, converted to current and presented on HMI as alarm: 'High shaft current' when result of the conversion exceeds 4A.
Typical resistance of the shunt is 0.005 Ohm so 20mV voltage detected across the shunt will suffice to provoke the alarm.
Such alarms (High shaft current but also High shaft voltage) are often observed during the turbine start up. Most probably as result of voltage transient spikes provoked by LCI starting system. It was observed that alarms appear during turbine start up also on the well maintained rotor grounding system.
Expected readings for 'shunt resistance' for well maintained rotor grounding system are very small, practically close to 1 Ohm (depends mainly on the distance from Speedtronic to rotor) as in this part of the circuit are present only solid electrical connections - no passage through the brushes is involved during this test.
The other two brushes (also connected in parallel) lead to Speedtronic and in properly maintained grounding system they should bring the voltage detected on the surface of the rotor. Path for electric current during the 'brush resistance' measurement at the 'Sensor Test' consists of: cabling from Speedtronic to brush rig, the resistance of the passage from brushes to shaft and back from shaft to brushes (plus resistance of the small part of the shaft located between the brushes) and remaining part of the circuit is the same as during the 'shunt resistance' measurement. For brand new brushes resistance measured by Speedtronic gives app 1..2 Ohms and the most part of this resistance comes from the resistance of contact between the brushes and the shaft. The carbon brushes of the model recommended by GE and with standard springs installed on the brush holders will survive for apr 2...3 months period of turbine operation. During this time, result of the 'brush resistance' test will give practically always the same value.
Rapid increase of the 'brush resistance' readings indicates that brushes became too short and barely touch the surface of the shaft. I assume that the surface of the shaft is clean (please consider possible presence of the oil traces from the hydrogen seal oil system located nearby) and interconnecting wiring is in good shape.
Frequent appearance of the alarm: 'High shaft voltage' is also indication of the deteriorated contact between grounding brushes and shaft. The other alarm (High shaft current) also may appear sporadically at this bad state of the brushes but this second alarm is prevalently false - see above explanation how milivolt readings are converted to current in Speedtronic.
Rotor grounding system installed on GE produced steam turbines/generators uses the same components as on the gas turbines (i.e. brush rig with four brushes and shunt and the same monitoring system in Speedtronic) but experience shows that combating the shaft voltage issue on steam turbines is more complex than on gas turbines.
One doubt..In the diagrams, the two 5V (P5) test relays contacts are shown, while sensor test, which will give power 5Vdc to the brush, as well as for the shunt..which may be wrong. As you said, I believe the 5 Vdc is applied only to the measurement brushes, and the contact resistance of brush+ shaft resistance + cable resistance is measured and indicated as brush resistance. interestingly the return path for this is taken for brush resistance is shown in diagram as after the ground brush and shunt resistor at the the grounded point. so will it measure the voltage drop across brush + shunt??? and show it as brush resistance? During the same sensor test the shunt resistor circuit will consider the shunt resistor voltage drop to find the shunt resistance. However it is not clear whether the P5 Volt will be directly given across the shunt, as from the diagram. (I think this P5 Vdc will not be given as the test activation buttons for both shunt and brush resistances are same.)
Also when we do the alarm test, no voltage is activated on the brushes. I think, and it is given internally to the circuits. Is this correct? No diagram is there for explaining how the alarm test is done...
I am not aware which exactly voltage readings and which current readings are used by Speedtronic to make the calculation (using Ohm's law) of the resistance during the "Sensor Test". In the 'Brush resistance' result is included resistance of all brushes (i.e. 'measuring' brushes as well as 'grounding' brushes).
As you noted, during the "Alarm Test", the testing voltage (AC signal) is connected only to internal Speedtronic part of the circuit, external circuit is disconnected - as indicated on the diagram that can be found on CSP.prn
Facebook
Google
GitHub
Linkedin
