- Joined
- 24 Nov 2004
- Messages
- 428
- Reaction score
- 4
- Country
Anyone know how Bender LIM's work in some depth? this is what they give on their datasheets:
"A measuring current in the μA range is superimposed onto the system which is recorded and evaluated by a microprocessor-controlled measuring circuit. The measuring time is dependent on the selected measurement profiles, the system leakage capacitance, the insulation resistance and possible system-related disturbances"
the word "evaluated" is doing a lot of work here I think
if we take the example of a simple 500V DC power transmission line, 2 conductors, under LIM control;
As far as I can understand it, the LIM assumes that a perfectly insulated system will effectively be a closed current loop. Therefore any current that the LIM injects onto the + line should return on the - line. Any difference between outbound current and "received" current would imply insulation failure, the difference being inversely proportional to the insulation resistance.
Am I right so far?
In commisioning the LIM, I think the thing must be calibrated initially with a known-good circuit. I beleive the LIM will find things like capacitances in the line. Does this then imply that the LIM's measuring current is probably some sort of an AC signal which could be delayed (phase-shifted) by reactance within the power circuit. ?
Finally, and crucially, can the LIM see beyond a DC-DC power converter at the far end of the line it is looking after?
I don't know much about the DC-DC converter, other than it is likely to be some sort of switching power supply which takes the 500V DC supply and converts it to a stabilised 48VDC for lamps, motors etc.
It *appears* that faults on the 48VDC system can produce LIM trips with associated low insulation resistance measurements.
I can see that the DC-DC converter might well have one of its lines galvanically connected input-output, but not both.
If this is the case, then it would be possible for the LIM to detect insulation faults on the supply rail which is connected-through.
Let's assume that the negative supply rail is connected-through, so that the 0V of the 500VDC circuit is common to the 0V of the 48VDC circuit. The LIM injects a measuring current onto the 0V line, and owing to an earth fault in the 48V DC part, some of the measuring signal leaks away and the LIM trips.
If the LIM injects a measuring current onto the +500V line. this time, there is no path on the "hot" line through to the low volts side. But, on its way back towards the LIM, this current will flow onto the 0V line, and be subject to any leakage on it on the 48VDC side.
therefore, any leakage on the 0V side of the 48VDC circuit could be detected by the LIM, but the LIM could not detect any leakage on the +48V rail. ???
when I first saw the circuit diagrams, I immediately assumed that the 500V-48V DC DC converter would be an isolating inverter power supply; that is - no galvanic connection between input output. Therefore I assumed that the LIM was ONLY acting on the 500V DC circuit. and could not respond to faults on the 48V.
I have no schematic of the DC DC converter and it is in a potted bottle.
am I right in thinking that the LIM cannot see through the converter if it is of the isolating type? (but it could see through half of it if the converter is common-neg)
"A measuring current in the μA range is superimposed onto the system which is recorded and evaluated by a microprocessor-controlled measuring circuit. The measuring time is dependent on the selected measurement profiles, the system leakage capacitance, the insulation resistance and possible system-related disturbances"
the word "evaluated" is doing a lot of work here I think
if we take the example of a simple 500V DC power transmission line, 2 conductors, under LIM control;
As far as I can understand it, the LIM assumes that a perfectly insulated system will effectively be a closed current loop. Therefore any current that the LIM injects onto the + line should return on the - line. Any difference between outbound current and "received" current would imply insulation failure, the difference being inversely proportional to the insulation resistance.
Am I right so far?
In commisioning the LIM, I think the thing must be calibrated initially with a known-good circuit. I beleive the LIM will find things like capacitances in the line. Does this then imply that the LIM's measuring current is probably some sort of an AC signal which could be delayed (phase-shifted) by reactance within the power circuit. ?
Finally, and crucially, can the LIM see beyond a DC-DC power converter at the far end of the line it is looking after?
I don't know much about the DC-DC converter, other than it is likely to be some sort of switching power supply which takes the 500V DC supply and converts it to a stabilised 48VDC for lamps, motors etc.
It *appears* that faults on the 48VDC system can produce LIM trips with associated low insulation resistance measurements.
I can see that the DC-DC converter might well have one of its lines galvanically connected input-output, but not both.
If this is the case, then it would be possible for the LIM to detect insulation faults on the supply rail which is connected-through.
Let's assume that the negative supply rail is connected-through, so that the 0V of the 500VDC circuit is common to the 0V of the 48VDC circuit. The LIM injects a measuring current onto the 0V line, and owing to an earth fault in the 48V DC part, some of the measuring signal leaks away and the LIM trips.
If the LIM injects a measuring current onto the +500V line. this time, there is no path on the "hot" line through to the low volts side. But, on its way back towards the LIM, this current will flow onto the 0V line, and be subject to any leakage on it on the 48VDC side.
therefore, any leakage on the 0V side of the 48VDC circuit could be detected by the LIM, but the LIM could not detect any leakage on the +48V rail. ???
when I first saw the circuit diagrams, I immediately assumed that the 500V-48V DC DC converter would be an isolating inverter power supply; that is - no galvanic connection between input output. Therefore I assumed that the LIM was ONLY acting on the 500V DC circuit. and could not respond to faults on the 48V.
I have no schematic of the DC DC converter and it is in a potted bottle.
am I right in thinking that the LIM cannot see through the converter if it is of the isolating type? (but it could see through half of it if the converter is common-neg)
Attachments
Last edited:
