Pentair VSF pool pump trips GFI (continued)

Status
Not open for further replies.
D

ddbear

Member
Location
Southern California
Occupation
Designer
I came across this previous thread (now closed for replies) and would like to continue the discussion,

Pentair Pool Pump GFCI trips

Pentair Pool Pump GFCI trips I just completed a install of a 230V pentair vs pump and this thread had me seriously worried I'd be tripping the GFI. I ended up having to install a boost transformer to get from 208v to 232v. Could this have anything to do with why the QO gfi installed has...
forums.mikeholt.com

I am having the same problem with my Pentair IntelliFlo3 VSF pump, which is a new model just recently introduced on the market. It trips the SquareD Q0 GFI breaker intermittently once every few weeks.

Reading through all the prior comments, apparently the cause is RF noise introduced on the AC wires. Pentair's recommended solution is to use their Pentair GFI breaker which is identical to, basically a relabeled Siemens breaker. The Siemens breaker has better RF filtering to avoid nuisance trips.

Not wanting to tear out my beautiful Square D Q0 subpanel to replace it with a lesser Siemens panel, I placed a ferrite clamp on the wires leading to the pump.

Many people talk about trying out ferrite clamps and rings, but no one follows up on the thread to say if this very simple band-aid actually works.

So I would like to ask, has anyone tried a ferrite clamp/ring for this issue and what were your long-term results?

I will report back my own results but it will take a few months to confirm.
 
brantmacga said:
Never tried that. What I have done though is set a small Siemens 60A 2-space loadcenter next to the equipment pad and install a Siemens GFCI there.


Sent from my iPhone using Tapatalk
Click to expand...

Nice solution! I wonder why I didn't think of that. 👀 So now we have two possible very easy solutions, 1. clamp on a ferrite filter (testing now), or 2. add on a small Siemens loadcenter near the equipment. Could it possibly be this simple to address after hundreds of posts on the internet with all sorts of electrical engineering discussions about how to solve this issue.
 
Some more observations - a small ferrite bead did nothing, still trips. Next I will try another one with a different filtering value or a ferrite ring with wires wrapped around it a few times (now I'm getting curious).
There is variation between different SquareD Q0 breakers. I tried a brand new one and it trips every night. The older breaker from 15 years ago trips once every few weeks.
 
I’m one of those who suggested ferrite beads, but as I said I had heard of people planning to try it, but never heard back on any success. So it’s a bit disappointing to hear that it didn’t work. I was going to try it on my own hot tub, but it developed other problems and the “boss” said we had to get rid of it.

The issue with different breakers and vintages has to do with the algorithm embedded in the microprocessor of the GFCI circuit board. Different mfrs have different “secret sauce” and don’t share it. Then within their own engineering depts., they periodically tweak that algorithm to address issues that come up. Sounds as though Schneider has not yet found a tweak that allows the use of an inverter. It may not be high on their list of priorities.
 
Jraef said:
I’m one of those who suggested ferrite beads, but as I said I had heard of people planning to try it, but never heard back on any success. So it’s a bit disappointing to hear that it didn’t work. I was going to try it on my own hot tub, but it developed other problems and the “boss” said we had to get rid of it.

The issue with different breakers and vintages has to do with the algorithm embedded in the microprocessor of the GFCI circuit board. Different mfrs have different “secret sauce” and don’t share it. Then within their own engineering depts., they periodically tweak that algorithm to address issues that come up. Sounds as though Schneider has not yet found a tweak that allows the use of an inverter. It may not be high on their list of priorities.
Click to expand...
The ferrite bead I used is a spare one I had from Radio Shack for small cords (DC cords?). I also ordered a couple Fair-Rite ferrite cores that are more commonly used for applications like this. Since they have different frequency filtering values I haven't given up on the ferrite idea yet, the experiment continues.
 
There is a difference in performance of GFCI when exposed to high frequency leakage current that can occur when the protected equipment uses power conversion equipment. A presentation to the members of UL 943 talked about testing with a refrigerator using 10 different GFCI devices. Some tripped and some did not trip under the exact same conditions.

Pentair uses power conversion equipment in some, if not all of their pool pump motors, and has found that the Siemens breaker does not trip with their equipment, but other brands do, so they specify the one that they know works.

UL 943 is working on proposals for new high frequency immunity standards for GFCIs, but the published standard is probably a couple of years away, and production a year or two after that.
 
With the new SquareD Q0 breaker (which trips every night so far instead of every few weeks) the 3 trips have been late at night/morning between 2am to 7am. During these hours the pump is set to a constant low speed. This makes me suspicious if there is a pattern that is affected by the pool bonding (hours of the night when there is more moisture in the landscape, or some device emitting RF on this schedule?). Do we know for sure that it's because of the power conversion inside the Pentair pump?

I found this on an Amazon review of the Pentair branded breaker: "I was constantly having ghost tripping of my 50 Amp GFCI (non-Siemens) breaker feeding the control panel. My electrician found a .22 amp current on the ground lead bonding the pumps to the pool. The suspicion is the pool is a better ground than my house main ground, and it was feed back through my VS pump. Tried putting a GFCI on the pump itself, both would trip, ended up buying this breaker for the VS pump, put the other 15 am GFCI on the single-speed pump, and voila no more ghost tripping (removed the feeder GFCI)."

Is it possible that the Pentair VS pumps have poor isolation between the bonding and other circuitry? Or the root cause is already well known as what don describes in the reply above?
 
ddbear said:
With the new SquareD Q0 breaker (which trips every night so far instead of every few weeks) the 3 trips have been late at night/morning between 2am to 7am. During these hours the pump is set to a constant low speed. This makes me suspicious if there is a pattern that is affected by the pool bonding (hours of the night when there is more moisture in the landscape, or some device emitting RF on this schedule?). Do we know for sure that it's because of the power conversion inside the Pentair pump?

I found this on an Amazon review of the Pentair branded breaker: "I was constantly having ghost tripping of my 50 Amp GFCI (non-Siemens) breaker feeding the control panel. My electrician found a .22 amp current on the ground lead bonding the pumps to the pool. The suspicion is the pool is a better ground than my house main ground, and it was feed back through my VS pump. Tried putting a GFCI on the pump itself, both would trip, ended up buying this breaker for the VS pump, put the other 15 am GFCI on the single-speed pump, and voila no more ghost tripping (removed the feeder GFCI)."

Is it possible that the Pentair VS pumps have poor isolation between the bonding and other circuitry? Or the root cause is already well known as what don describes in the reply above?
Click to expand...

The reviewer fixed his tripping problem by installing the Pentair (Siemens) breaker. The other information in the review is superfluous, especially without more details. If the source of the current on the bonding wire was the pump motor, any GFCI breaker would trip instantly - changing to the Pentair breaker wouldn’t fix it.
 
As an update - I decided not to try out the AC-optimized frequency ferrite rings I was planning to install (someday someone else can run that experiment), instead I'm adding a small Siemens 2 space loadcenter (E0204ML1060SU) with the Siemens QF120A breaker next to the equipment.

For the main subpanel I could use a regular breaker, but I decided to go redundant and I'm using a Square D Q0 EPD breaker (30mA GFI) in instead. I tested the EPD breaker and there is no tripping so far unlike the 5mA GFI SquareD Q0 breaker.

As a side note, I like safety redundancy even though I know it's unnecessary, which is why I will use the EPD breaker + Siemens GFI breaker for the pool pump. It gives a little extra peace of mind in case one of the GFIs malfunctions many years later when it's not on anyone's mind anymore. I did something similar for the 120V pool lights (one GFI in the subpanel, another one closer to the pool lights, in series).
 
petersonra said:
I think all current GFCIs are self testing so there should be no worry about failure. If they fail, they will just shut off.
Click to expand...
They are not required to stop providing power. They are permitted have an audible or visual indication that the device is no longer providing GFCI protection. From UL 943:
5.16.4 The consequence of the auto-monitoring test detection of a problem shall be one or more of the following:
a) Power denial (trip with the inability to reset);
b) Trip with the ability to reset, subject to the next auto-monitoring test cycle or repeatedly trip;
c) Visual and/or audible indication as permitted by 6.31.8.
Click to expand...
 
As another interesting data point, today the EPD breaker (30mA) tripped but not the Siemens 5mA GFI. I will have the electrician use a regular breaker at the subpanel and the Siemens GFI at the equipment room will be enough. This is interesting because apparently it has nothing to do with the mA but other noise phenomenon interfering with non-Siemens GFIs.
 
ddbear said:
As another interesting data point, today the EPD breaker (30mA) tripped but not the Siemens 5mA GFI. I will have the electrician use a regular breaker at the subpanel and the Siemens GFI at the equipment room will be enough. This is interesting because apparently it has nothing to do with the mA but other noise phenomenon interfering with non-Siemens GFIs.
Click to expand...
You are never permitted to use an EDP breaker where the code requires GFCI protection.
 
don_resqcapt19 said:
You are never permitted to use an EDP breaker where the code requires GFCI protection.
Click to expand...
Read above. The EPD was not the one for protection but put there as a second one in series upstream in lieu of a regular breaker that has zero ground fault protection. The protection is from the Siemens 5mA GFI that is next to the equipment. The 2nd EPD was not necessary, just more of an experiment. This led to an interesting result finding that the 30mA EPD is still tripped by some kind of noise phenomenon in the Pentair variable speed pump while the 5mA Siemens does not.
 
Since I had already ordered the part (FT240-43 toroid core), I figured let’s try it out of curiosity …

I had it installed right before the pump circuit enters a SquareD Q0 20A GFI breaker as an experiment. This is mostly for the benefit of this community since I had already solved this with a Siemens GFI breaker.

If you don’t hear back from me for a while, it means it hasn’t tripped yet with this ferrite ring. If it trips, I will post an update as soon as that happens. If it goes a long time without tripping, I will eventually update that as well.


IMG_7686.jpeg
 
Updating to let everyone know that this toroid core solved the SquareD Q0 GFI tripping. It used to trip every night, but hasn't tripped anymore after adding this filter.

What we learned from this experiment:
  • Needs to target a specific frequency range (FT240-43 toroid) with multiple wraps around the coil, a simple inline ferrite bead for general purpose filtering like the ones you find on the ends of USB cords is not sufficient.
  • The tripping has nothing to do with the mA threshold (some have speculated maybe Siemens allows a bit more than 5mA) because a 30mA EPD breaker also trips.
There are many threads on the internet on this issue. I believe this is the first reported solution of something other than a Siemens brand GFI.
 
ddbear said:
Updating to let everyone know that this toroid core solved the SquareD Q0 GFI tripping. It used to trip every night, but hasn't tripped anymore after adding this filter.

What we learned from this experiment:
  • Needs to target a specific frequency range (FT240-43 toroid) with multiple wraps around the coil, a simple inline ferrite bead for general purpose filtering like the ones you find on the ends of USB cords is not sufficient.
  • The tripping has nothing to do with the mA threshold (some have speculated maybe Siemens allows a bit more than 5mA) because a 30mA EPD breaker also trips.
There are many threads on the internet on this issue. I believe this is the first reported solution of something other than a Siemens brand GFI.
Click to expand...
It actually does...some of the GFCI devices trip at a level below 4mA when they see high frequency leakage current. The members of the technical committee for UL 943 are working a proposal that, if passed and published, will require GFCIs to have some immunity to high frequency current. I assume the same thing happens with the 30 mA trip devices.
 
Status
Not open for further replies.
Top