Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
TVA/Corona Probe Suppliers 1
- Thread starter Addy71
- Start date
electricpete
Electrical
I didn't realize the probe had a voltage rating. The probe isn't energized at test voltage and should never be energized at that voltage unless a fault occurs where the coil breaks through ground insulation and arcs to the probe. Are they are voltage rated based on that arc scenario to protect safety of the operator?
Adwell is another manufacturer.
=====================================
(2B)+(2B)' ?
Adwell is another manufacturer.
=====================================
(2B)+(2B)' ?
- Thread starter
- #4
EPete,
I assume the voltage rating must be based on the insulation level provided in the probe rather the arc scenario; as arc would be nominal when generator is energized from a HV test kit. Iris never had its own probe; they got it when they took over Adwel.
I am looking at the Doble Lemke probe (LDP-5) as they have not mentioned any voltage limit (however; they have not yet confirmed the max. voltage rating yet). The instrument seems to have more flexibility & applications compared to the Corona Probe.
The Corona Probe is only sensitive to the negative half cycle PD. This seems to be a limitation; as it will then pick up only slot & end-winding PD.
Smallgreek,
I already have two PD Tech MICAMAXX & one HVPD Longshot PD Analyzers. However; I am not of any instrument (including these, Iris, Doble, etc.) that can pinpoint the PD to a specific stator bar. If you are aware of such instruments (other than TVA & Lemke), please let me know. I would truly appreciate it.
Regards,
Aditya
I assume the voltage rating must be based on the insulation level provided in the probe rather the arc scenario; as arc would be nominal when generator is energized from a HV test kit. Iris never had its own probe; they got it when they took over Adwel.
I am looking at the Doble Lemke probe (LDP-5) as they have not mentioned any voltage limit (however; they have not yet confirmed the max. voltage rating yet). The instrument seems to have more flexibility & applications compared to the Corona Probe.
The Corona Probe is only sensitive to the negative half cycle PD. This seems to be a limitation; as it will then pick up only slot & end-winding PD.
Smallgreek,
I already have two PD Tech MICAMAXX & one HVPD Longshot PD Analyzers. However; I am not of any instrument (including these, Iris, Doble, etc.) that can pinpoint the PD to a specific stator bar. If you are aware of such instruments (other than TVA & Lemke), please let me know. I would truly appreciate it.
Regards,
Aditya
electricpete
Electrical
Agreed. Monitoring at the terminals can’t tell you the same thing as tva probe scanning the stator bore.
An interesting pattern we have seen during TVA probe test of one family of 25+ year-old 13.2kv motor windings is that that the near-line-end coils show higher current (partial discharge indication) than the near-neutral-end coils, even though the entire winding is energized at the same voltage during the test. The correlation is very strong. It is proof positive that the particular pd pattern has evolved as a result of degradation from voltage stress during operation. Also may suggest a strategy to reverse line-end leads.
=====================================
(2B)+(2B)' ?
electricpete
Electrical
I left a few words out of my last paragraph. Added below in bold:
An interesting pattern we have seen during TVA probe test of one family of 25+ year-old 13.2kv motor windings is that that the near-line-end coils show higher current (partial discharge indication) than the near-neutral-end coils, even though the entire winding is energized at the same voltage during the test (neutral coils see same voltage as line end coils). The correlation is very strong. It is proof positive that the particular pd pattern has evolved as a result of degradation from voltage stress during operation. Also may suggest a strategy to swap line-end leads and neutral leads
=====================================
(2B)+(2B)' ?
An interesting pattern we have seen during TVA probe test of one family of 25+ year-old 13.2kv motor windings is that that the near-line-end coils show higher current (partial discharge indication) than the near-neutral-end coils, even though the entire winding is energized at the same voltage during the test (neutral coils see same voltage as line end coils). The correlation is very strong. It is proof positive that the particular pd pattern has evolved as a result of degradation from voltage stress during operation. Also may suggest a strategy to swap line-end leads and neutral leads
=====================================
(2B)+(2B)' ?
- Thread starter
- #7
EPete,
Thanks for the feedback. A few questions please:
1. I have been recommending swapping the line & neutral leads to many clients in India but no one takes it seriously. Is this a standard practice in USA?
2. For these particular 13.2 kV motors; do you have a phase-resolved PD pattern? That would tell whether the TVA senses negative PD only or not.
3. For other high voltage motors & generators; did you see any similar correlation? Or was this a one-off case?
Regards,
Aditya
Thanks for the feedback. A few questions please:
1. I have been recommending swapping the line & neutral leads to many clients in India but no one takes it seriously. Is this a standard practice in USA?
2. For these particular 13.2 kV motors; do you have a phase-resolved PD pattern? That would tell whether the TVA senses negative PD only or not.
3. For other high voltage motors & generators; did you see any similar correlation? Or was this a one-off case?
Regards,
Aditya
Addy71
I would recommend you talk to Omicron about what they have (MPD System). Their units can look at multiple sensors simultaneously, which helps to pinpoint the location. I am not a PD expert, but have attended quite a bit of seminars on this. They do have quite a few of the world experts in this subject.
I would recommend you talk to Omicron about what they have (MPD System). Their units can look at multiple sensors simultaneously, which helps to pinpoint the location. I am not a PD expert, but have attended quite a bit of seminars on this. They do have quite a few of the world experts in this subject.
electricpete
Electrical
From googling, I see the omicron system measures the signal from coupling caps, just like iris does.
It's not logical imo to think you can tell the difference between pd in individual coils better by measuring a signal derived from machine terminals (coupling cap) than by measuring the pd related emissions physically adjacent to the individual coil (tva probe)
=====================================
(2B)+(2B)' ?
It's not logical imo to think you can tell the difference between pd in individual coils better by measuring a signal derived from machine terminals (coupling cap) than by measuring the pd related emissions physically adjacent to the individual coil (tva probe)
=====================================
(2B)+(2B)' ?
electricpete
Electrical
I will dig up our pd results for those machines
=====================================
(2B)+(2B)' ?
=====================================
(2B)+(2B)' ?
electricpete
Electrical
Hi Aditya,
The particular motors of interest is a family of 8. We have seen visual indications of partial discharge on the last 4 that we have pulled for refurbishment. We did TVA probe maps on the first two of those (that’s the only time we have ever done TVA probe tests on any motors). The first was rewound, 2nd had leads reversed, 3rd and 4th were rewound.
We did not particularly see elevated partial discharge during our on-line pd testing. Note we have surge caps installed on these motors, which may render the pd caps less sensitive. An overview of pd data shows generally equal positive and negative predominance.
Attached is data for the first of these motors. 1st slide suggests slight positive predominance, particularly on C phase. 2nd slide shows more complicated pattern including some interphase pd. 3rd slide shows tva probe test results. I’ve provided some info about the winding configuration at top of slide. I have more details somewhere, can’t locate it at the moment.
I do remember from an Iris training that for classic pd (no interphase), the positive predominance is supposed to represent pd at the outer surface of insulation, negative predominance is supposed to represent pd at voids between conductor and insulation, equal is supposed to represent voids spread through the volume of insulation. I never understood why that was.
I’m not familiar with limitations of TVA probe with regard to pulse polarity. Can you explain some more about that.
Reversing leads was recommended by two different respected repair shops for the 2nd motor, which was not as severe as the first (all tva probe readings were less than 20 on the 2nd motor). It actually was tougher for this motor than most because this is only a 3-lead motor... required doing a little bit of rewiring in the connection end to break the neutral there and bring those leads out. I’m prettty sure it is described in Greg Stone’s “Electrical Insulation for Rotating Machinery”. It is a sound option to consider when motor is in shop and evidence of aging is found on the line-end coils. Of course rewind is better, but swapping is far better than sending the motor back for operation with the voltage-aged coils still on the line end. Perhaps a caveat to mention is that there is a possibility that a “weak” coil may be hiding on the neutral end... so you wouldn’t want to do this procedure unless you also do a hi-pot to check for that condition to find it during test rather than operation. And if you do a hi-pot, you have to be prepared for possible rewind depending on results anyway.
smallgeek - Going back to the question of tva probe vs pd testing by coupling capacitor... I didn’t mean to be argumentative. Each certainly has their strengths. Strength of pd test is that you can do it on-line whereas tva probe requires motor offline AND disassembled. Strength of TVA probe method is that you can better discriminating which coils have pd and what is the pattern around the winding.... as you can see from slide 3 of attachment we get a reading for each slot. When you talked about multiple sensors, I think you were referring to the use of two separate couplers on each phase to discrimate between pulses from machine and pulses from system by "time of flight". That is a useful enhancement to pd monitoring in a noisy power system.
=====================================
(2B)+(2B)' ?
The particular motors of interest is a family of 8. We have seen visual indications of partial discharge on the last 4 that we have pulled for refurbishment. We did TVA probe maps on the first two of those (that’s the only time we have ever done TVA probe tests on any motors). The first was rewound, 2nd had leads reversed, 3rd and 4th were rewound.
We did not particularly see elevated partial discharge during our on-line pd testing. Note we have surge caps installed on these motors, which may render the pd caps less sensitive. An overview of pd data shows generally equal positive and negative predominance.
Attached is data for the first of these motors. 1st slide suggests slight positive predominance, particularly on C phase. 2nd slide shows more complicated pattern including some interphase pd. 3rd slide shows tva probe test results. I’ve provided some info about the winding configuration at top of slide. I have more details somewhere, can’t locate it at the moment.
I do remember from an Iris training that for classic pd (no interphase), the positive predominance is supposed to represent pd at the outer surface of insulation, negative predominance is supposed to represent pd at voids between conductor and insulation, equal is supposed to represent voids spread through the volume of insulation. I never understood why that was.
I’m not familiar with limitations of TVA probe with regard to pulse polarity. Can you explain some more about that.
Reversing leads was recommended by two different respected repair shops for the 2nd motor, which was not as severe as the first (all tva probe readings were less than 20 on the 2nd motor). It actually was tougher for this motor than most because this is only a 3-lead motor... required doing a little bit of rewiring in the connection end to break the neutral there and bring those leads out. I’m prettty sure it is described in Greg Stone’s “Electrical Insulation for Rotating Machinery”. It is a sound option to consider when motor is in shop and evidence of aging is found on the line-end coils. Of course rewind is better, but swapping is far better than sending the motor back for operation with the voltage-aged coils still on the line end. Perhaps a caveat to mention is that there is a possibility that a “weak” coil may be hiding on the neutral end... so you wouldn’t want to do this procedure unless you also do a hi-pot to check for that condition to find it during test rather than operation. And if you do a hi-pot, you have to be prepared for possible rewind depending on results anyway.
smallgeek - Going back to the question of tva probe vs pd testing by coupling capacitor... I didn’t mean to be argumentative. Each certainly has their strengths. Strength of pd test is that you can do it on-line whereas tva probe requires motor offline AND disassembled. Strength of TVA probe method is that you can better discriminating which coils have pd and what is the pattern around the winding.... as you can see from slide 3 of attachment we get a reading for each slot. When you talked about multiple sensors, I think you were referring to the use of two separate couplers on each phase to discrimate between pulses from machine and pulses from system by "time of flight". That is a useful enhancement to pd monitoring in a noisy power system.
=====================================
(2B)+(2B)' ?
electricpete
Electrical
To clarify, the pd data attached to previous message was taken at the plant with motor energized from 3-phase power. We didn't take any off-line pd test. But tva probe test was done in "off-line" test configuration where entire winding is energized to 13kv/sqrt(3)~ 8kv.
=====================================
(2B)+(2B)' ?
=====================================
(2B)+(2B)' ?
electricpete
Electrical
... and also surge test once winding is reversed. Again the purpose is to prevent "infant mortality" failures during operation from any weak coils that may previously have been hiding/protected on the neutral end of the winding.
=====================================
(2B)+(2B)' ?
- Thread starter
- #14
Hi EPete,
Thanks for the slides. So, I understand that you do TVA probe checks only if the Iris system shows high values. Is that correct?
I have attached a presentation explaining the cause of +ve & -ve pulses, by Cutler-Hammer. It is most logical. I get confused at times as PD Tech uses different nomenclature. I will attach their representation in a separate file.
Regarding the TVA probe limitation; check out specification on pulse polarity. This is from the Adwel days; I've asked Iris but they are silent on this matter.
Clients often show interest in the the line & neutral connection swapping but no one has actually got it done till date.
Smallgreek,
I have seen the Omicron MPD 600. You can use multiple MPD units to triangulate the PD on transformers; but I've never seen them trying it on a generator.
Regards,
Aditya
Thanks for the slides. So, I understand that you do TVA probe checks only if the Iris system shows high values. Is that correct?
I have attached a presentation explaining the cause of +ve & -ve pulses, by Cutler-Hammer. It is most logical. I get confused at times as PD Tech uses different nomenclature. I will attach their representation in a separate file.
Regarding the TVA probe limitation; check out specification on pulse polarity. This is from the Adwel days; I've asked Iris but they are silent on this matter.
Clients often show interest in the the line & neutral connection swapping but no one has actually got it done till date.
Smallgreek,
I have seen the Omicron MPD 600. You can use multiple MPD units to triangulate the PD on transformers; but I've never seen them trying it on a generator.
Regards,
Aditya
-
1
- Thread starter
- #15
- Thread starter
- #16
Hello All,
There is a new paper available from Omicron (Cigre) that discusses PD on rotating machines. Again, I'm not an expert, but from what I understand, the 3PARD evaluation technique which Omicron uses appears to be the next generation of PD detection.
There is a new paper available from Omicron (Cigre) that discusses PD on rotating machines. Again, I'm not an expert, but from what I understand, the 3PARD evaluation technique which Omicron uses appears to be the next generation of PD detection.
electricpete
Electrical
Thanks Aditya. I have never seen that. Looks like a good practical empirical chart "diagnostic chart" to try to help us identify significance of various pd patterns. I will file that away in my references.
Thanks also smallgreek. It looks like they are trying to compare the characteristics/timing of a single discharge pulse launched from the winding as it appears at all three terminals. That is more than Iris does. It also looks like the paper falls in the category of research.
=====================================
(2B)+(2B)' ?
Thanks also smallgreek. It looks like they are trying to compare the characteristics/timing of a single discharge pulse launched from the winding as it appears at all three terminals. That is more than Iris does. It also looks like the paper falls in the category of research.
=====================================
(2B)+(2B)' ?
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question