RCBO nuisance tripping - need new ideas on isolating the issue

[MeltemiCaz]

We have been trying diagnose and troubleshoot a nuisance tripping of the RCBO on my British sailboat. We are location here in the U.S. (California). We completed an electrical upgrade earlier this year and have been commissioning the systems this past summer and fall.

We have an issue where the Shore 1 and Shore 2 power inlets are tripping the RCBO when power is routed through the Victron Energy Multiplus-IIs (3 x units in parallel). I'm told that the MP-IIs each leak 4-6mA and the RCOB is a 30mA device.

What is driving us crazy is the the RCBO first starting tripping at 5:14am, or at 7:04am. This happened religiously almost every day (4 out 5 days) per week for nearly 5 weeks. Then just recently the tripping has moved to predominantly 9:53am. There was one trip at 3:53pm. (notice the [:53]).

I am working with three electrical and marine engineers. We believe that we have eliminated any issues inside the boat and am not focused on what would cause the nuisance tripping from the dock power.

I am looking for new ideas on what to check and ideas for how to monitor the 240 dock mains to see if we can see any anomalous readings. I am enclosing a high schematic of the system. Essentially, power flows from Dock > Boat inlet > 40 amp breaker > V.E. Isolation transformer which converts to a 230 L and N > Blue Sea ELCI >>> to the AC panel > RCBO > 3 separate breakers to the MP-IIs > combined into a single 80A 30mA RCBO and distributed to the boat's AC breakers. Only the RCBOs on the main AC inputs are tripping. (However one of the ELCI breakers tripped once in the past 8 weeks.)

What is odd (to us) is that the RCOBs are tripping only at certain times of the day and only once per day. There have been some days when the breaker is NOT tripping.

The times that the RCOB began tripping in early November are:

5:14 AM
7:04 Am
This went on for 6 weeks.
In the past two weeks new times for tripping have taken over.

Most of them are now at 9:53 AM
We had one trip at 3:53 PM (notice the :53 and that there is a six hour offset)

I would like set up some test equipment to monitor the tripping as I've only witnessed it happening once. See this Youtube video at 1:30

I have a Hioki 3283 Clamp On Leak HiTester. The leakage between L and N is a nominal 1-1.2 mA.

I am wondering if I should set up an oscilloscope to measure the mains sinusoidal frequency for anomalies. (If this is a good idea, I need to figure how to do this using my small battery powered scope and 10x or 100x probe so that I don't fry the scope)

 

[ValhowellPower]

If you have an isolation transformer, chances are that it is not a dock power issue. Are the 3 Multiplus units in series or parallel? You are not trying to output 3 phase ac are you? A system schematic would be a lot more helpful.

 

[MeltemiCaz]

If you have an isolation transformer, chances are that it is not a dock power issue. Are the 3 Multiplus units in series or parallel? You are not trying to output 3 phase ac are you? A system schematic would be a lot more helpful.

The 3 MP-IIs are in parallel NOT 3-phase. We have had intense debates on the issue being boat side or shore side triggered. I have two electrical engineers helping to troubleshoot along with a marine chief engineer with a degree in mechanical engineering and 20+ years as chief engineer on super yachts with extensive experience with both 60Hz and 50Hz electrical systems.

Our bias is that the the issue is boat based but now they feel like we have eliminated all of the smoking-guns for it being an issue with something on the boat. We have not found anything on the boat that is time-based. It is the intermittent issues that are happening at prescribed times that have us puzzled.

I am the owner and an engineer but not a electrical power engineer. I am struggling with ways to set up to measuring and monitoring techniques to capture the tripping electrically.

Both of the electrical engineers are leaning towards the issue being dock side. One thinks that it's current leakage from another boat on my dock flowing back through my anodes into my ground system.

The other electrical engineer is speculating that its dirty power as we are at the end of the line after a cement plant and other industrial high energy consumers and wonders if it's a faulty reactive power management that is causing an issue for a fraction of a cycle causing an imbalance between Line and Neurtral.

At this point we are thinking about putting new RCBOs in the system as the ones tripping are original and my be overly sensitive tripping.

 

[Clark]

Do you have a galvanic isolator somewhere in the system with a self testing function? If so, turn off the self testing function and see if the problem is solved. Normally, there would not be one with an isolation transformer.

 

[ksanders]

You mentioned having an isolation transformer.

If that is not faulty, it isolates any issues within your boat from the shore power breaker system, except for overcurrent.

I would start looking dockside.

 

[MeltemiCaz]

Do you have a galvanic isolator somewhere in the system with a self testing function? If so, turn off the self testing function and see if the problem is solved. Normally, there would not be one with an isolation transformer.

No galvanic isolator in the system

 

[MeltemiCaz]

You mentioned having an isolation transformer.

If that is not faulty, it isolates any issues within your boat from the shore power breaker system, except for overcurrent.

I would start looking dockside.

That’s the direction we are going in. I’m looking for suggestions on how to monitor the power line and record the data for review.

 

[ValhowellPower]

The 3 MP-IIs are in parallel NOT 3-phase. We have had intense debates on the issue being boat side or shore side triggered. I have two electrical engineers helping to troubleshoot along with a marine chief engineer with a degree in mechanical engineering and 20+ years as chief engineer on super yachts with extensive experience with both 60Hz and 50Hz electrical systems.

Our bias is that the the issue is boat based but now they feel like we have eliminated all of the smoking-guns for it being an issue with something on the boat. We have not found anything on the boat that is time-based. It is the intermittent issues that are happening at prescribed times that have us puzzled.

I am the owner and an engineer but not a electrical power engineer. I am struggling with ways to set up to measuring and monitoring techniques to capture the tripping electrically.

Both of the electrical engineers are leaning towards the issue being dock side. One thinks that it's current leakage from another boat on my dock flowing back through my anodes into my ground system.

The other electrical engineer is speculating that its dirty power as we are at the end of the line after a cement plant and other industrial high energy consumers and wonders if it's a faulty reactive power management that is causing an issue for a fraction of a cycle causing an imbalance between Line and Neurtral.

At this point we are thinking about putting new RCBOs in the system as the ones tripping are original and my be overly sensitive tripping.

 

[ValhowellPower]

I did an install of an isolation transformer a while back. The Victron unit has a wet dry jumper that is used to set the ground to the shore ground when out of the water. We wired his as a switch.

I know this sounds odd, but he was having a similar problem and it solved it by throwing the switch to dry and back to wet. If your access is easy, you might give this a try. It may be a grounding anomaly in the isolation transformer.

 

[cafesport]

TF members love a good “puzzler” (google Click&Clack Magliozzi brothers).
And this sounds like a good one for the hive mind.

From the video there is apparently a greater than a 12ma (assuming the 3 victrons account for 18ma according to mfg spec) imbalance observed at a somewhat regular interval on a daily basis while connected to shore power. Let’s assume for the moment that all electrical mods are working as designed.

Before you go down a rabbit hole and spend more time and money trying to measure thing, I would first try to eliminate the hardest item to measure. Unplug from the dock and run the generator overnight. There are numerous suspects to investigate if this persists. Stray DC through the bonding system is interesting. Anyone around you monitoring the vendee globe via ssb?

 

[Greg QS]

These the original to the boat. So we are not talking about the new generation afci breakers, correct?
This breaker is the main breaker of the boat protecting the wire from the power post to the boat itself.
The power post protects the dock from the draw of the boat. If this is not tripping , most likely the problem is on you side. From the post to the main breaker inside.

If you are sure you are not overloading the boat system, Nothing new added? And A new problem. I would start by

1 check the connections on the breaker and the power supply cord.
2 change out the main breaker.

Breakers explained simplified. older style not new gen afci or gfi .

Heat from draw makes breakers trip. Breakers , when they are turned on and off, like any light switch, create a small connection spark. This is your main breaker for the boat and is commonly used. Over time the constant turning on and off corrodes or dirties the contact. This creates heat. So the breaker will trip at a lower amp draw over time until they quit working.

If nothing on the boat has changed and this is a new problem . My experience tells me you have a breaker going bad. I would start checking the connections on the breaker and cord. Any issues here create more draw to the boat and make breakers trip. If no issues change the breaker.

You are most likely getting a small spike in consumption on the boat or variations in dock power. The variation in dock power can spike the load of the boat. This draw is now over the capability of the dirty breaker. Not because you have a over draw. Because the breaker can no longer hold the load. So change the breaker.

Whenever I am diagnosing nuisance tripping in electrical system I generally start at the breaker it self.

 

[MeltemiCaz]

These the original to the boat. So we are not talking about the new generation afci breakers, correct?
This breaker is the main breaker of the boat protecting the wire from the power post to the boat itself.
The power post protects the dock from the draw of the boat. If this is not tripping , most likely the problem is on you side. From the post to the main breaker inside.

If you are sure you are not overloading the boat system, Nothing new added? And A new problem. I would start by

1 check the connections on the breaker and the power supply cord.
2 change out the main breaker.

Breakers explained simplified. older style not new gen afci or gfi .

Heat from draw makes breakers trip. Breakers , when they are turned on and off, like any light switch, create a small connection spark. This is your main breaker for the boat and is commonly used. Over time the constant turning on and off corrodes or dirties the contact. This creates heat. So the breaker will trip at a lower amp draw over time until they quit working.

If nothing on the boat has changed and this is a new problem . My experience tells me you have a breaker going bad. I would start checking the connections on the breaker and cord. Any issues here create more draw to the boat and make breakers trip. If no issues change the breaker.

You are most likely getting a small spike in consumption on the boat or variations in dock power. The variation in dock power can spike the load of the boat. This draw is now over the capability of the dirty breaker. Not because you have a over draw. Because the breaker can no longer hold the load. So change the breaker.

Whenever I am diagnosing nuisance tripping in electrical system I generally start at the breaker it self.

Changing the RCBO (MCB+RCD) is the direction I am leaning as well. 30 years as an engineer makes me reluctant to start swapping out parts. Both Shore 1 and Shore 2 and therefore both RCBOs are tripping when the MP-IIs are connected. Both RCBOs are of the same age. And it’s clear that they have been used as switches for 22 years. I’m inclined to do some “A” “B” testing by installing a new RCBO on Shore 1 and a regular breaker on Shore 2 and alternate running power through each input. I still have the ELCIs on both circuits to provide protection.

 

[Greg QS]

If there are ELCIs on both circuits and not tripping 99% chance its the RCBO breaker/ breaker connections.

 

[MeltemiCaz]

I agree with your assessment. The ELCI has tripped once in eight weeks. I really want to find the source of the spike in the current leakage imbalance between L and N. I’m looking for suggestions on how to monitor and try and capture the temporary increase in leakage to ground.

 

[Greg QS]

Leakage from N to G .
Make sure all N + G are Isolated in the panel.
Check your 220 loads. Common for the N in 220 loads is crossing grounds or bars. Meaning the 220 load draws from L1 and L2 but the G or N in the circuit is only tied to 1 N or G bar causing the imbalance on the opposite Line. Some times it the 220 device. Depending on how its internally wired.

I had a problem on my 42 with a Seakeeper 220 volt version. I had twin 30 amp cords. The ELCI would never hold for long when I had the Seakeeper plugged into dock power either powered on or off. The solution was a triple line break breaker. Breaking L1 L2 and N to the Seakeeper . Solved the issue but had to be in the off position.

Trying to find it.
Simply turn everything on. Turn off 1 or 220 breaker off at a time. then test till you find it. At some point It will disappear . The check everything on that line. Then go through that line.

Check AC/DC units. Like bat chargers. Where ever ac and dc could potentially share a ground. The shared ground ac to dc throws off the balance of the N&G

 

[cafesport]

I did a quick search for data logging current transformers. You could install up to four split core hall effect sensors using this display (4-Channel Analog Data Logger MX1105 | Onset's HOBO Data Loggers)
if you want to build a system to monitor a few items simultaneously. A small ct around the ground lead from the neutral to ground tie in is probably all you need, but you can easily add a few more.

 

[MeltemiCaz]

This device was recommended in one of the Victron Energy Community Forums.

B-stop (to be installed) - TEYPRA

Fixes the unwanted tripping of the residual current device (RCD).

www.teypra.com

B-Stop

I can't tell if this is snake-oil or a legitimate device.

 

[MeltemiCaz]

Also, on the same discussion thread this device was mentioned for data logging:

Zeus Din Home Standard Kit - TEYPRA

 

[twistedtree]

Some questions:

- Do you have a generator, and does this happen when running on generator power?

- If would be helpful to see exactly how your neutral to ground bonding is set up, both on the shore inlet side of the iso transformers, and on the boat side. In particular, how is n-g bonding handled for the inverters, both when in pass through mode, and when in invert mode?

- Have you tried trimming down the stack of MP-2 inverters? It might be informative to know whether the problem persists with a single inverter, and with a stack of two inverters.

I agree that the regularity of the tripping is a clue - I just don't know what it means. But it makes me wonder if the MP-2 might do a periodic quick cycle of the transfer relays to both verify they are working, and the cycle/exercise them. Such an exercise cycle coupled with your bonding strategy is might be the cause.

 

[MeltemiCaz]

Very interesting thoughts. I will speak with the elect. engineers today when I get back to the boat and get more details today regarding neutral-ground bonding and post them here. I too am wondering about about the time component of this puzzle and also wonder if it’s tied to the multiplus IIs.

 

[SIBERNUT]

I am an electrcal idiot. But due to the times, is it possible that the electric company brings more power online for the start of the workday causing a small spike, enough to trip ?
Please dont laugh..........

 

[twistedtree]

I think a spike or surge is an unlikely cause. An RCD looks for any difference in current in the line and neutral wires. They should always be the same, and if they differ by more than 30ma, the RCD trips. So somehow, somewhere, current is finding an alternate path other than through the intended line and neutral wires. There also is a time component to an RCD tripping. I don't know the duration off hand, but the 30ma limit needs to be exceeded for some period of time before it trips.

There are two ways to do the N-G bonding in a situation like your, specifically with isolation transformers.

Some more questions:

- In your isolation transformers (7kva, right?) is J12 connected to anything? It should not be.

- Are J5 and J7 jumpered together?

- How are you MP-2s configured with respect to N-G bonding?

 

[MeltemiCaz]

I think a spike or surge is an unlikely cause. An RCD looks for any difference in current in the line and neutral wires. They should always be the same, and if they differ by more than 30ma, the RCD trips. So somehow, somewhere, current is finding an alternate path other than through the intended line and neutral wires. There also is a time component to an RCD tripping. I don't know the duration off hand, but the 30ma limit needs to be exceeded for some period of time before it trips.

There are two ways to do the N-G bonding in a situation like your, specifically with isolation transformers.

Some more questions:

- In your isolation transformers (7kva, right?) is J12 connected to anything? It should not be.

- Are J5 and J7 jumpered together?

- How are you MP-2s configured with respect to N-G bonding?

- J12 is not connected to anything. I only use it to jumper to J9 when on the hard.
- J5 IS jumpered to J7 to establish an AC output neutral to earth connection link which according to the documentation is required for the RCBO to work correctly.

 

[MeltemiCaz]

Some questions:

- Do you have a generator, and does this happen when running on generator power?

- If would be helpful to see exactly how your neutral to ground bonding is set up, both on the shore inlet side of the iso transformers, and on the boat side. In particular, how is n-g bonding handled for the inverters, both when in pass through mode, and when in invert mode?

- Have you tried trimming down the stack of MP-2 inverters? It might be informative to know whether the problem persists with a single inverter, and with a stack of two inverters.

I agree that the regularity of the tripping is a clue - I just don't know what it means. But it makes me wonder if the MP-2 might do a periodic quick cycle of the transfer relays to both verify they are working, and the cycle/exercise them. Such an exercise cycle coupled with your bonding strategy is might be the cause.

- I have not run the generator long enough for to see if it trips.
- re neutral-ground bonding set up, I will speak to the engineers to get the exact details. What is interesting is that we spent a lot of time planning the install of the new equipment. I listened to several lengthy calls re making sure the n-g was done correctly. We have checked and re-checked our wiring over the past few weeks, but as soon as I can get a detailed explanation, I am happy to post here for all to read. I have seen many, many discussions on the various forums where DIY's and even some experienced installers have struggled. I think shinning a big light on this subject is warranted.
- I will speak to the main electrical engineer re the idea of trimming back to only one MP-II. He has said that in his experience from other marine, RV, and off-grid installations, that the MP-IIs leak 4-8mA by design.

- RE you comment "But it makes me wonder if the MP-2 might do a periodic quick cycle of the transfer relays to both verify they are working, and the cycle/exercise them.", I have wondered the same. I've posted questions on the Victron Community forum board, but have not had much luck getting answers.

I am also using these documents as references:

- V.E. Wiring Unlimited

- Multiplus-II 230 volt manual

- LEAKAGE AND GROUND CURRENTS: MEASUREMENT TECHNIQUES

- FLUKE: Leakage Current Measurement Basics

 

[twistedtree]

It sounds like the isolation transformers are wired correctly, so that's good. I'm sure you have checked all this, but it's often good to go back over it with different eyes, so to speak. Some of the trickiness here might relate to having a boat wired to EU standards, and shore power built to US standards.

Does your Shore1/Shore2/Gen selector switch break both line and neutral? Or does it only switch Line? I should switch both, otherwise you will have multiple bonding points active at the same time.

I'm definitely curious about how bonding is handled in the MP-2s, especially with three of them. I expect they are similar to the other MPs where there is an internal boding relay that can be enabled/disabled, and also a control output that can be used to drive an external bonding relay. One thing I wonder about is how those internal relays are used with stacked inverters like you have. I think to work reliably you would need to use only one of the relays. Or maybe it doesn't matter. I'd have to think about that one some more.

The other way to handle bonding when you have isolation transformers is to do it centrally, and have a common neutral throughout the boat. This is my own preferred way because it's simple and nearly impossible to screw up or break over time, and there is no switching on/off of bonding.

 

[Bmarler]

The rcbo’s are fairly old right? Trips have occurred many times over their life. Is it possible that the contact themselves have resistance built in to them from arcing over long time periods? Add in the ma loss at the multiplus units. To top it off, there’s some outside influence of voltage variation at the supply side either from industrial or power company usage. A voltage variation may influence the amp draw on a leg feeding a transformer by a small amount. Multiple things that are adding up to take the rcbo over the edge.

 

[MeltemiCaz]

1) It sounds like the isolation transformers are wired correctly, so that's good. I'm sure you have checked all this, but it's often good to go back over it with different eyes, so to speak. Some of the trickiness here might relate to having a boat wired to EU standards, and shore power built to US standards.

Agreed. I've asked another electrical engineer a few boats over to provide another set of eyes. He does have some questions about the possibility of a ground loop in the system but does not think that if so, is not contributing to the issue.

2) Does your Shore1/Shore2/Gen selector switch break both line and neutral? Or does it only switch Line? I should switch both, otherwise you will have multiple bonding points active at the same time.

Yes, we are using the Kraus-Namier rotary switches to switch both Line and Neutral. I am also digging into and trying to fully understand how the bonding is done when we are using the Shore 1 and Shore 2 Isolation Transformers. This includes Shore 1 only, Shore 2 only and BOTH Shore 1 + Shore 2. The team spent considerable time working through this but it does not hurt to have another set of eyes.

3) I'm definitely curious about how bonding is handled in the MP-2s, especially with three of them. I expect they are similar to the other MPs where there is an internal boding relay that can be enabled/disabled, and also a control output that can be used to drive an external bonding relay.

Yes, that is my understanding from the Victron Energy documentation that I am reading.

One thing I wonder about is how those internal relays are used with stacked inverters like you have. I think to work reliably you would need to use only one of the relays. Or maybe it doesn't matter. I'd have to think about that one some more.

I will ask the electrical engineer that designed the system to comment on this.

4) The other way to handle bonding when you have isolation transformers is to do it centrally, and have a common neutral throughout the boat. This is my own preferred way because it's simple and nearly impossible to screw up or break over time, and there is no switching on/off of bonding.

Sam the marine engineer says that the ground-neutral bonding is as follows:

Isolation transformer casing is grounded via the shore connection.
The ship's ground is isolated from the shore power connection.
Neutral ground connection is in the isolation transformer when shore power is connected.
When the ship is inverting neutral ground bonding is in the inverter.
AC ground and DC negative are bonded.

I was measuring the mA at the input to the Shore 1 RCBO across L-N this morning using the min-max function on the Hioki 3283 with varying loads on the system hoping to capture a trip at 9:53 AM with luck. The Hioki was giving a reading range of 18.5-18.9 mA. I will double check the Fluke tutorial literature, but believe that I was taking the measurements correctly .

I hate replacing parts without knowing the cause in the hopes of finding and fixing the problem. But I think at this point, replacing the RCBO on Shore 1 with a new one is worth the effort. We will also replace the other RCBO of similar vintage with a regular 2-pole breaker to see if the tripping when testing on Shore 2 is pushed back to the Blue Sea ELCI.

However, this will do nothing to discover the source of the time-based tripping.

 

[twistedtree]

I don't know what control panel you have for your MP-2s, but some provide a way to tell the MPs to "Ignore" the AC input, thereby forcing it to disconnect from AC and switch to invert mode. As an experiment, you might try toggling that back and forth a bunch of times to see if you can cause the RCD to trip. If there is something funky about the ground bonding relays, it might stimulate the problem.

 

[more]

​

I hate replacing parts without knowing the cause in the hopes of finding and fixing the problem. But I think at this point, replacing the RCBO with a new one is worth the effort.

RCBO is broken or you neutral-ground bonding set up have leeking.
but do you know tunder can trip from 10-20 km, solar panel usually also trip, solar storm can trip electric network all continent. also 30ma usually use in bath tube to sensitive. but really protective life. try new 100 or 300ma

https://www.aliexpress.com/item/4001032275000.html?spm=a2g0o.productlist.main.5.2b22KkemKkemxZ&algo_pvid=69f1c48b-956e-4634-988a-10ee68f502bf&algo_exp_id=69f1c48b-956e-4634-988a-10ee68f502bf-2&pdp_npi=4%40dis%21EUR%2124.25%2122.80%21%21%2124.79%2123.31%21%40210388c917342940210764219eb927%2112000027318658985%21sea%21HR%216274884%21X&curPageLogUid=G0aM2xrs7SFb&utparam-url=scene%3Asearch%7Cquery_from%3A

1
Those 100-300mA are selective, they are intended to supply multiple <10-30mA, they are not meant for direct touch protection RCD. For direct touch, a RCD of less or equal than 30mA is suitable.
10mA and 30mA are for human protection, though the IEC 60364 standards only ask for 30mA. 30mA are mostly used for socket outlets and places of human access.10mA are for Sauna, Swimming pool equipment, etc.
For light fittings and others 100mA is selected and others 300 or 500mA which is called equipment protection or fire protection.
.
but if you need install this 300 ma master RCBO and instal socket rcd 10-30ma where need.
also buy Chinese good brand because is made in centuries more advance factory vs USA factory. Europa have couple new factory , but big brands I see inside made in china

 

[MeltemiCaz]

The rcbo’s are fairly old right? Trips have occurred many times over their life. Is it possible that the contact themselves have resistance built in to them from arcing over long time periods? Add in the ma loss at the multiplus units. To top it off, there’s some outside influence of voltage variation at the supply side either from industrial or power company usage. A voltage variation may influence the amp draw on a leg feeding a transformer by a small amount. Multiple things that are adding up to take the rcbo over the edge.

I hate replacing parts without knowing the cause in the hopes of finding and fixing the problem.

But I think at this point, replacing the RCBO on Shore-1 with a new one is worth the effort. We will also replace the other RCBO of similar vintage with a regular 2-pole breaker to see if the tripping when testing on Shore-2 is pushed back to the Blue Sea ELCI.

This will provide some comparisons that might prove useful.

However, this will do nothing to discover the source of the time-based tripping.