GFI Breakers, Pentair pumps and such

[Open Neutral]

So I spec'ed a Pentair Intelliflo I2 pump. This is a variable speed 240V pump.
It keeps tripping the GFI. We tried a different GFI breaker; no joy.

So I finally called Pentair. The helpful support person asked if it was a 5ma or 6ma GFI.

Seems that line of pumps needs the latter. He mentioned the Siemens PA220GF.
He acknowledged that none of the documentation mentions that un-minor detail.
Searching for that number gets me lots of hits at swimming pool supply houses.

But this site is all SqD QO panelboards, so I have a query into SqD asking if they make
a 6ma versions. Nothing yet.

I see mention on the Siemen's site that the Siemens breaker *might* fight a SqD panel. I don't like "mights' on safety critical issues.

And here I thought GFI == 5mA.....

Sigh...

 

[synchro]

The required GFCI trip threshold is 4mA minimum and 6mA maximum. And so 5mA would be the nominal value with a ±1mA allowed tolerance. I doubt there are different breakers for 5mA and 6mA.
It could be that the Siemens breaker is more immune to noise if Pentair has found them to work better with their equipment. But that it just speculation on my part.

Do you have a clamp meter with enough sensitivity and accuracy to measure the actual leakage current?

Oh, and the lower threshold is allowed to be 3.5mA when the ambient temperature is outside of a -5°C to 40°C range. Not that this matters for the problem at hand.

 

[Open Neutral]

Do you have a clamp meter with enough sensitivity and accuracy to measure the actual leakage current?

SQD say: Square D QO220GFI so we'll try that.
I can come with instrumentation but won't be on site for several months. I'll have the GC install the new breaker and we'll observe it.

 

[curt swartz]

SQD say: Square D QO220GFI so we'll try that.
I can come with instrumentation but won't be on site for several months. I'll have the GC install the new breaker and we'll observe it.

The QO220GFI is a standard QO 2-pole 20-amp GFCI breaker. What breaker are you using now?

The issue you are experiencing is pretty common and the solution is to use the Siemens breaker. You could install a standard QO/QOB220 breaker and feed a small panel with a Siemens breaker.

 

[robertd]

The required GFCI trip threshold is 4mA minimum and 6mA maximum. And so 5mA would be the nominal value with a ±1mA allowed tolerance. I doubt there are different breakers for 5mA and 6mA.
It could be that the Siemens breaker is more immune to noise if Pentair has found them to work better with their equipment. But that it just speculation on my part.

Do you have a clamp meter with enough sensitivity and accuracy to measure the actual leakage current?

Oh, and the lower threshold is allowed to be 3.5mA when the ambient temperature is outside of a -5°C to 40°C range. Not that this matters for the problem at hand.

If it's noise that's the problem I wonder if ferrite beads / cores would do any good. You can get good RF noise filters, but they tend to have capacitors from L-N, L-G and N-G, as well as inductors. The capacitor from L-G may be a step in the wrong direction for a GFCI.
Companies really need to start making electronic items that can work on a GFCI...not likely if it's made in China.

 

[texie]

This issue has been going on for years with Pentair. The idea that a listed product won't work with any standard GFCI device where GFCI is required by the NEC is just absurd to me. I can't believe they have been getting away with this. And this problem in general is getting worse for the whole industry. For example, under current code many kitchen appliances now need to be GFCI depending on their location in a kitchen yet almost all install manuals typically state that a GFCI is not recommended and when you have an issue they just tell you that they don't "recommend a GFCI" with their product knowing full well that it may be code required. I don't understand why the NRTLs allow them to put this verbiage in the manuals.

 

[Open Neutral]

The QO220GFI is a standard QO 2-pole 20-amp GFCI breaker. What breaker are you using now?

Not sure, but thanks. Will get a unGFI swapped in.

The issue you are experiencing is pretty common and the solution is to use the Siemens breaker. You could install a standard QO/QOB220 breaker and feed a small panel with a Siemens breaker.

So how small a panel will hold a sole PA220GF? Is there such a beast as a "one-armed bandit" disconnect that holds a sole PA breaker?
{More research...}

Thanks to all for your inputs. As usual, I've learned from you.

 

[synchro]

If it's noise that's the problem I wonder if ferrite beads / cores would do any good. You can get good RF noise filters, but they tend to have capacitors from L-N, L-G and N-G, as well as inductors. The capacitor from L-G may be a step in the wrong direction for a GFCI.

I agree. But with a 2-pole GFCI, if the L1-G and L2-G capacitances are equal then their GF leakages will cancel out, at least to some extent. The amount of cancellation would also be limited by the matching between the L1-N and L2-N voltage waveforms.

 

[retirede]

I agree. But with a 2-pole GFCI, if the L1-G and L2-G capacitances are equal then their GF leakages will cancel out, at least to some extent. The amount of cancellation would also be limited by the matching between the L1-N and L2-N voltage waveforms.

It may be better today, but back when I dealt with filter capacitors, their standard tolerances were 20%. So it’s possible that the filter might not be well balanced with regard to leakage current.

 

[Jraef]

The issue with Pentair saying you need to use Siemens GFCI breakers has been going on for years. It has nothing to do with whether it s “4mA or 6mA”, whomever told you that was making it up. It has to do with the internal microprocessor detection algorithm (math code) of that brand of GFCI being more tolerant of the electrical noise made by the VFD on the pump. All VFDs create noise, there is no way around it. But some GFCI mfrs have tweaked their algorithm to avoid nuisance tripping, some have not, and it is not the same from one VFD to another either, so some GFCIs will tolerate some VFDs that others will not. Pentair discovered that the Siemens GFCI is somehow compatible with their VS pump and have been advertising that for a long time now.

I have heard the idea floated about using ferrite beads several times and have even suggested it as an experiment, but nobody has reported back if it worked or not. I have a separate VFD I have wanted to try it on, but it doesn’t trip the Square D GFCI that I have so I don’t have the actual conditions necessary to test the ferrite beads.

 

[hbiss]

I thought I heard here how Pentair has their own breaker to take care of this problem.

-Hal

 

[LarryFine]

So how small a panel will hold a sole PA220GF? Is there such a beast as a "one-armed bandit" disconnect that holds a sole PA breaker?

Home Depot shows several Siemens 4-space indoor load centers.

 

[LarryFine]

I thought I heard here how Pentair has their own breaker to take care of this problem.PA220GF

The PA220GF shows as a Pentair-labeled breaker.

 

[Open Neutral]

Pentair discovered that the Siemens GFCI is somehow compatible with their VS pump and have been advertising that for a long time now.

I'd be less annoyed if they DID advertise that, or mention it in the Fine Manual. Their own support person admitted they do not.

I have heard the idea floated about using ferrite beads several times and have even suggested it as an experiment, but nobody has reported back if it worked or not. I have a separate VFD I have wanted to try it on, but it doesn’t trip the Square D GFCI that I have so I don’t have the actual conditions necessary to test the ferrite beads.

Wish I had time to experiment with same.

 

[mtnelect]

This issue has been going on for years with Pentair. The idea that a listed product won't work with any standard GFCI device where GFCI is required by the NEC is just absurd to me. I can't believe they have been getting away with this. And this problem in general is getting worse for the whole industry. For example, under current code many kitchen appliances now need to be GFCI depending on their location in a kitchen yet almost all install manuals typically state that a GFCI is not recommended and when you have an issue they just tell you that they don't "recommend a GFCI" with their product knowing full well that it may be code required. I don't understand why the NRTLs allow them to put this verbiage in the manuals.

This is the same thing that is going on with variable speed condensing units ... The NFPA issued a TIA to suspend GFCI on HVAC units until 2026.

 

[brantmacga]

I thought I heard here how Pentair has their own breaker to take care of this problem.

-Hal

It is a Siemens breaker with a Pentair label. They stock them here at the pool builder supply house.


Sent from my iPhone using Tapatalk

 

[brantmacga]

But this site is all SqD QO panelboards, so I have a query into SqD asking if they make
a 6ma versions. Nothing yet. .

They do not. And that is not the issue causing it to trip.

I see mention on the Siemen's site that the Siemens breaker *might* fight a SqD panel. I don't like "mights' on safety critical issues...

It will not.

Set a Siemens sub-panel at the pump, install Siemens 20/2 GFI breaker.


Sent from my iPhone using Tapatalk

 

[Open Neutral]

So I asked Siemens Support which "small" box would hold one PA220GF.

8 days later, I was told:

Per our research team,
We do not offer enclosures for other manufacturers products. We can only advise on Siemens products only
We CANNOT advise as PA220GF is NOT a Siemens product.

Of course, it is on their catalog page.

Another phone call and finally shuffled to someone, who when confronted with the URL, said something like "Well, that's different.."
Oh? How so?

But he did finally come up with their W0204ML1060 which is what I need.

Case closed.. {I hope.}