PM63-VN Surge Protector "Ground OK" detection circuitry

[BillP]

So I just bought this: https://www.apc.com/shop/ph/en/products/APC-Performance-SurgeArrest-6-Outlet-3-Meter-Cord-230V-Vietn... and I have some questions about its "Ground OK" detection circuitry. Before that, a brief background of my electrical network/system:

I'm from the Philippines and our electrical system has proper grounding. From the service drop, we get two 110V live lines (hot) plus one neutral. All three wires go into the main panel. The main panel's casing is properly grounded to grounding rods with proper length. Neutral is tied to ground in main panel (neutral-to-ground bonding). All 4 wires then get distributed to sub panels where neutral and ground are separated as they should be. From the sub panels, 1 live/1neutral/1 ground go to 110V sockets and 2 lives/1 ground go to 220V socket. This is all following the standard.

For whatever reason, the "Ground OK" LED of the surge protector doesn't light up when I plug it in a 220V socket in our home but lights up when plugged into a 110V socket. I'm 100% sure the sockets are wired correctly so this is not the cause of improper wiring. I've tried several 220V and 110V sockets in our home and I experience the same behavior.

I also noticed that only your 110V Back-Ups Pro (ex. BR1500G) devices have the ground detection in them while the 220V ones (ex. BR1500GI) do not have ground detection. So I'm wondering how the ground detection circuitry works? Is it dependent on the neutral wire for it to work? As you know, a 220V socket do not have a neutral wire. And if this ground OK detection only works for 110V, why is the feature even included in the PM63-VN when that surge protector is advertised as "230V"?

Please help. Thanks.

[BillP]

BUMP! Anyone?

[BillP]

Why is APC support not replying here?!

[Brad_C]

Since nobody else has replied (and I'm not APC), I did a bit of digging.

The data sheet for the devices specifies it as a 220,230,240V surge protector with a NEMA 5-15P plug. The NEMA 5 configuration is an "active-neutral-ground", so it's expecting a neutral. An "active-active-ground" configuration is the NEMA 6.

As it's expecting a neutral, we assume it can correctly ascertain a functioning protective ground by measuring it against neutral. On your 110V socket that is how it's wired. On the 220V socket you have no neutral, you have 2 actives 180 degrees out of phase and a protective earth. Unless the board has circuitry specifically designed for that then it's not going to work.

That seems to correlate what you are seeing.

The other thing to contemplate, is if you plug a device designed for > 200V into a 110V circuit, the surge diversion is going to let through spikes that would pass the >200V devices (generally clamping > 275V). So it's not going to do what it was designed to do. If you plug it into your 220V socket, it's not going to have appropriate protection for the "neutral" line and it probably won't do what you want it to do either.

[BillP]

That's what I assumed at first. But then, I tried it in a different house (let's call it house2) with the same active-active-ground on the outlets and the ground OK circuitry worked. The NEMA 5-15P plug is a standard in our country. We don't use NEMA 6. The only difference from house1 that I noticed is this:

The only weird thing is that the L1-G and L2-G voltages in house2 are both 130VAc but the L1-L2 voltage is 230VAc as expected. I'm not sure what accounts for the additional 15VAc drop (130VAc - 115VAc) L1/L2 to G? Can that potentially affected the APC device's ground detection circuitry? I also checked the breaker in house 2 (condominium) and I only see 2 hot lines and ground. I don't see neutral anywhere.

As a comparison, in house1 I get the expected 115VAc from any hot line to ground. And I double-checked the grounding circuitry of the main panel/sub panel/etc. and they look about just right. We even have two main panels (for two separate buildings) in hosue1 coming from the same 2L+N lines (from the main meter) and are wired exactly the same. They have separate local grounding systems (ground rods) too. But for some reason the APC device's ground OK light really doesn't light up for any of the 230V sockets there.

From the different earthing systems as shown here, I'm sure that house1 has a TT system (where we provide our local grounding rods) as I was there when it was recently renovated. As for the condo house2, I'm not sure of its design but it's possible that it has the same TT system but without the neutral bus bar in the subpanel inside the house because they don't provide us 110V sockets here. But again, the ground OK circuitry of the suppressor works. So what gives?

[BillP]

Also, I've been talking to other people in other forums and this explanation seems to be logical from one member:

"Regarding the APC's ground detection, I figure it's just a weakness of the detection circuitry, meaning it's not 100% accurate. Like I mentioned earlier, it might work by detecting the presence of a voltage between line and ground. However, in doing this you necessarily have to form a connection between the line and the ground (otherwise you cannot test it!). The problem with this is that with enough current flowing to the ground you'll trip a GFCI (as little as 5 mA for a wall receptacle, ~30 mA for a breaker panel-mounted one). So if it works this way then the current being allowed to flow should be very little. If the ground detection works by voltage sensing then it might explain why it works at your place in Taguig and not in Pampanga. On the assumption that the ground in the former has higher resistance, it will generate a higher voltage which can be sensed more easily. Conversely, a good ground (low resistance) would cause a lower voltage that might be below the threshold of the detection circuit. This is consistent with your multimeter readings of 130 V in house2 but only 115 V in house1. Of course this setup is paradoxical in that a better grounding system is penalized, but this is just a theory.

In any case it's not really an issue because I figure the best way to determine if you have a working ground is to 1) do visual inspection that a grounding system is in place 2) test with a multimeter the line to ground voltage. If there is no functional ground then no voltage reading should be present since current can't flow, it's akin to touching one of the multimeter's probe to the hot line and the other one to nothing. Since you have just done these tests and determined that the ground is working, then it's safe to just ignore the false ground light status in your place in house1."

[Brad_C]

On 27/10/2020 2:38 PM, Kevin Mychal said:

That's what I assumed at first. But then, I tried it in a different house (let's call it house2) with the same active-active-ground on the outlets and the ground OK circuitry worked. The NEMA 5-15P plug is a standard in our country.

You didn't mention that up front. I just spent half an hour reading the Philippines Electrical Codes and trying to decipher the supply arrangements. Challenging. So as you say, some supplies are split phase 115/230V and some are single phase 230V. From what you've intimated in a split phase house you could have active-neutral-earth 115V and active-active-earth 230V and they are wired to the same specification NEMA 5-15 sockets? How to you stop someone plugging a 115V appliance into a 230V socket and causing disappointmentt?

On 27/10/2020 2:38 PM, Kevin Mychal said:

The only weird thing is that the L1-G and L2-G voltages in house2 are both 130VAc but the L1-L2 voltage is 230VAc as expected. I'm not sure what accounts for the additional 15VAc drop (130VAc - 115VAc) L1/L2 to G?

One thing that could cause that is them not being exactly 180 degrees out of phase, but how that'd happen on a split phase transformer is beyond me. The other thing is some weird harmonic on the Earth/Neutral line or significant neutral imbalance. Not unheard of at all in multi-phase systems with a significant phase to neutral loads, but it'd be unusual to see the same imbalance to both phases.

On 27/10/2020 2:38 PM, Kevin Mychal said:

But again, the ground OK circuitry of the suppressor works. So what gives?

Without stripping it down and looking at the circuit, who knows? If it's designed for "active-neutral-ground" then plugging it into an "active-active-ground" is in completely unknown territory.

[Brad_C]

On 27/10/2020 2:53 PM, Kevin Mychal said:

In any case it's not really an issue because I figure the best way to determine if you have a working ground is to 1) do visual inspection that a grounding system is in place 2) test with a multimeter the line to ground voltage. If there is no functional ground then no voltage reading should be present since current can't flow, it's akin to touching one of the multimeter's probe to the hot line and the other one to nothing. Since you have just done these tests and determined that the ground is working, then it's safe to just ignore the false ground light status in your place in house1."

While what you say is potentially true, using a meter (which has an input impedance of some 10Mohm to test for the presence of a reliable ground isn't a reliable test.It doesn't take much capacitive coupling on a disconnected "ground" wire to see lots of nice volts on the meter when testing with a super high impedance load. I'd be more trusting of the old 3 light neon or led socket tester. At least they draw a couple of milliamps.

The real point is you can't rely on a surge supressor that is designed for an "active-neutral-earth" supply to do the right thing on an "active-active-earth" supply. It might make you feel warm and fuzzy but not really do much to protect your gear. Without stripping it down to analyse the circuit it's all a bit of a guess.

[BillP]

On 10/27/2020 8:52 PM, Brad said:

You didn't mention that up front. I just spent half an hour reading the Philippines Electrical Codes and trying to decipher the supply arrangements. Challenging. So as you say, some supplies are split phase 115/230V and some are single phase 230V. From what you've intimated in a split phase house you could have active-neutral-earth 115V and active-active-earth 230V and they are wired to the same specification NEMA 5-15 sockets? How to you stop someone plugging a 115V appliance into a 230V socket and causing disappointmentt?

My bad. When I posted my OP, I didn't get the change to test it on house2 yet. I forgot to update the OP since then. Correct, both 115V and 230V sockets use the same NEMA 5-15 sockets. And as frustrating as it sounds to be, you can't stop someone from plugging a 115V appliance on a 230V socket. It's just one of those quirks that we have here. If we buy 110V devices from the US, there's always the chance of it being destroyed. We just label the sockets 110V and 220V, that's it.

One thing that could cause that is them not being exactly 180 degrees out of phase, but how that'd happen on a split phase transformer is beyond me. The other thing is some weird harmonic on the Earth/Neutral line or significant neutral imbalance. Not unheard of at all in multi-phase systems with a significant phase to neutral loads, but it'd be unusual to see the same imbalance to both phases.

I see.

Without stripping it down and looking at the circuit, who knows? If it's designed for "active-neutral-ground" then plugging it into an "active-active-ground" is in completely unknown territory

While what you say is potentially true, using a meter (which has an input impedance of some 10Mohm to test for the presence of a reliable ground isn't a reliable test.It doesn't take much capacitive coupling on a disconnected "ground" wire to see lots of nice volts on the meter when testing with a super high impedance load. I'd be more trusting of the old 3 light neon or led socket tester. At least they draw a couple of milliamps.

The real point is you can't rely on a surge supressor that is designed for an "active-neutral-earth" supply to do the right thing on an "active-active-earth" supply. It might make you feel warm and fuzzy but not really do much to protect your gear. Without stripping it down to analyse the circuit it's all a bit of a guess.

If it's being sold officially here by APC, I would think that it should be designed for our main electrical system here which is the split phase 115V/230V system, no? I understand that there are single phase 230V systems here too but those are in the minority (provinces). In the capital of the country, it's almost always 115V/230V split phase. So it's really counterproductive for APC to sell these in our country if it won't even work correctly. But then again, I'm still baffled by the fact that it the ground OK light works on one house but not on another house when they have similar electrical systems.

[BillP]

BUMP! Can anyone help out here?

[BillP]

Can anyone please help?! How do I get support from APC here?

[BillP]

You can always reach out to our team in the Philippines:

https://www.apc.com/ph/en/support/contact/ask_apc.cfm

-Gavan 

[BillP · Accepted solution]

I already did and their technical support is not competent enough, at least the one who I've talked to. He gave me the wrong details for our standard electrical system here so that's why I resorted here.