Source for DIN rail circuit breaker

[Electromatic]

I had to look at a control panel for a municipal water mixer today that is protected by a din rail ground fault circuit breaker. It has been intermittently tripping. I didn't measure any substantial current on the ground. The customer said that there was some utility work in the area around the time the problem started, so I somewhat suspect the ground fault circuitry may have been damaged. The exact replacement for the breaker (Schneider M9R84240) is about $400, and other brands I am familiar with are about the same. However, I see some online that are only $50-100. I would normally never trust something like that, but I believe these types of breakers (RCD) are reasonably common in Europe and other parts of the world. Does anyone have a reliable source or brand for this kind of device? The trip threshold for the existing breaker is 300mA.
Thanks in advance.

 

[petersonra]

Ab makes that are in the $75 range

 

[Jraef]

Just out of curiosity, what is this being used for in a mixer control panel? Is it something that the equipment mfr required?

Also, this is not an “RCD”, it is an RCCB, which is different. An RCD is the ground fault ONLY, it is not a “circuit breaker”. An RCCB is a full blown circuit breaker PLUS the RCD functionality. So when looking for a replacement, make sure it is equivalent. RCCBs are more expensive.

Also worth noting: if you are here in North America, these are not NRTL listed, it is considered a “supplemental protector”. So you will still need another listed OCPD ahead of this in the circuit.

 

[Electromatic]

There is a regular 20A,120V circuit breaker feeding the control panel. In the control panel, the power goes to a din rail rotary disconnect switch then to this device which then feeds two other 1P din rail breakers. One of those breakers feeds an enclosure fan and a couple of relays for SCADA (not being used). The other breaker feeds a Powerflex 525 drive which then feeds this mixer that sits inside the water tank. The control panel was pre-assembled and came with the mixer. I assume the ground fault is because the mixer sits submersed in water.

I think functionally this device only needs to serve as ground fault protection. The existing one is rated at 40A, but again, the entire panel is fed at 20A.

I looked at the AB line (thinking no way AB is cheaper than something else!), and sure enough, with the residual current feature, it's also around $400. There are some kind of sketchy ones on Amazon for around $70. Everything else seems to be on international websites. I'm just not sure how to source this.

 

[jim dungar]

There is a regular 20A,120V circuit breaker feeding the control panel. In the control panel, the power goes to a din rail rotary disconnect switch then to this device which then feeds two other 1P din rail breakers. One of those breakers feeds an enclosure fan and a couple of relays for SCADA (not being used). The other breaker feeds a Powerflex 525 drive which then feeds this mixer that sits inside the water tank. The control panel was pre-assembled and came with the mixer. I assume the ground fault is because the mixer sits submersed in water.

I think functionally this device only needs to serve as ground fault protection. The existing one is rated at 40A, but again, the entire panel is fed at 20A.

I looked at the AB line (thinking no way AB is cheaper than something else!), and sure enough, with the residual current feature, it's also around $400. There are some kind of sketchy ones on Amazon for around $70. Everything else seems to be on international websites. I'm just not sure how to source this.

Is your 'ground fault breaker' feeding the drive or is it after the drive?
Is the mixer hard wired or connected by flexible cord?

 

[Jraef]

1) There is no requirement for a GFEP (Ground Fault Equipment Protection) in this situation. GFEP is what you have with a 300mA RCCB (at best).
If the issue is that this is in a residential application and there is water involved, it needs to be a GFCI, which is NOT the same as what you have there. A GFCI is required to trip at ~5mA (4-6mA), so a 300mA RCD is pointless because it would NOT satisfy the requirement. If the OEM did this thinking it did, they were woefully uninformed. Besides, the NEC requires it to be a "Class A GFCI" which is established by the UL listing of it, so anything that is not UL listed cannot by definition be a Class A GFCI.

2) If they had some OTHER reason for adding an RCD, using an expensive RCCB rated for 40A downstream of a 20A BCPD was a total waste of money, they could have just bought a less expensive RCD. But again, doing so would NOT be based on the NEC, because those devices would NOT qualify if that were the case.

3) Putting a GFCI ahead of a VFD is not going to work, or at least not for long. Most of the time the VFD will cause the GFCI to nuisance trip. It likely didn't trip immediately because it was a 300mA RCD, but eventually the common mode current leakage of the VFD / motor circuit increased to above 300mA, so now it is tripping. I would expect that replacing the RCD will not fix this. Anecdotally I have heard that Siemens GFCI breakers don't nuisance trip on the internal VFDs that now come built into some pool and spa pumps, but I don't know of anyone that has tried that on a VFD driven load in a residential application. I have had a couple of customers try to use PowerFlex 525s and other brands of 120V input VFDs behind GFCIs and could never get them to hold in.

I know this may not be what you want to hear, but it sounds as though this needs a full reevaluation.

 

[Electromatic]

Thanks for the feedback and information, @Jraef

This system is at a large municipal water tank, not residential. I put a DMM between the branch circuit EGC and the grounds to the drive and other control panel components and read about 16mA cumulative among everything. Would the VFD/motor leakage be intermittent for any reason?

The town(customer) approved getting an exact replacement of the breaker, but I'm going to reach out to the equipment manufacturer and see if there is a reason to have this device.

 

[Jraef]

… I put a DMM between the branch circuit EGC and the grounds to the drive and other control panel components and read about 16mA cumulative among everything. Would the VFD/motor leakage be intermittent for any reason? …

Yes, all manner of reasons because it’s not really “leakage”, it is common mode current (noise) caused by how a VFD emulates a sine wave output to the motor using high speed DC pulses. In doing so, you introduce capacitive coupling in the output leads referenced to ground so there is no way to absolutely ensure that all three simulated phases sum to zero, resulting in a slight (as in mA) amount of current not returning to the source, which is exactly what a GFCI/RCD is looking for. That capacitive coupling is what can vary with site conditions etc. There are things you can try however.

1. If you did not use VFD cable between the drive and motor, change it out to that, even if you are in steel conduit. Capacitive coupling is one of the issues that VFD cable addresses. If you did or do use VFD cable, make sure to ground BOTH ends of the shield and on the VFD end, only use the extra ground terminal provided on the drive itself.
2. Another is the carrier frequency of the VFD (in the PF 525 it is called the “PWM frequency”, param. A440). That drive can be set between 2 and 16 kHz, the factory default is 4 kHz. Set that to 2 kHz. Increasing the carrier frequency increases the capacitive coupling. If someone didn’t like the whiny sound made by the VFD and increased the CF to get rid of it, that may have triggered this issue.
3. Lastly, you can try adding a Common Mode Choke to the output of the drive. The choke (usually a ferrite ring) slows down the rise time of the CM noise peaks, helping to lower any additive effects.

There’s no guarantee any or even all of these things will help, but given that it had worked for a while, it might get you back to that.

 

[Electromatic]

Thanks again, @Jraef

The installation is with THWN in GRC for about 100' from the control panel & drive to a disconnect at the top of the water tank. There is then about 100' of factory cord from the disconnect down into the tank and submersed mixer. There is also, as part of the factory control panel, a LC sine wave filter on the output of the drive.

I'm waiting to hear back from the equipment supplier with information and/or their recommendations.

 

[Jraef]

Ah, the LC filter on the output of the drive can actually INCREASE the CM current… It’s good for other reasons, especially the long distance involved, but not that one (because of the “C”). This is why VFDs and ground fault devices are so finicky. There are too many variables involved.

 

[Electromatic]

The equipment manufacturer sent a replacement, non-ground fault breaker for this system. Apparently, their engineering department and panel building shop concluded that the ground fault feature was both unnecessary and had issues tripping. The system ran fine after installing the replacement breaker for around 30min. while I was on site. I did, however, find a Fault 13--Ground Fault code in the Powerflex history. I told the customer to keep an eye on things, and we'll see if anything happens.

Thanks again for the help and advice.