Receptacle Tester Reliability

[Gunnar Alquist]

My inspection today was a vacant second floor condo. I realize that is not relevant, but decided to include that for your reading pleasure.

I plugged my (cheap) GB 3-light receptacle tester (aka "night light") into a receptacle outlet in the living room and the reading indicated hot/ground reversed. A reasonably dangerous condition.

I moved to the next receptacle with the same reading. Then four more. I also found a couple with hot/neutral reversed. But, as I pulled the tester out of the last hot/ground reversed, the reading on the "night light" changed and became correct. Huh?

I double-checked the receptacle outlets with my (stupidly expensive) Amprobe (similar to the Ideal Suretest) and the receptacles that tested as hot/ground reversed were correctly wired, but the hot/neutral reversed were still wrong. The "night light" continued to give me correct readings on the remainder of the receptacles in the home, which I verified.

I have no idea why the (cheap) receptacle tester gave me these anomalous readings (other than the obvious: that it's a cheap tester). Guess I'll have to chuck this one and get another $7 tester.

Maybe I should buy two?

Similar Threads:

• Blinking 'correct', 'open ground' or 'open neutral' on receptacle tester
• Why all 3 lights of circuit tester are on
• Carbon monoxide tester
• Finally a AFCI Tester
• polarity tester

[Jerry Peck]

Gunnar, I can't find the photo now, but I put 3 of the small meter shows below together, arranged to resemble the neutral and hot side by side, with meter for ground below the other two, wired with a 3-prong cord and plug.

The left (neutral) meter read the N to G voltage, and the right meter read the H to N voltage, with lower meter (ground) measuring the H to G voltage.

Quick and easy to plug in, no waiting time for digital to display what was what.

Plug it in and (hopefully) read:
0v N to G on left; 120v H to N on right; and 120v H to G on bottom which was good

If it read:
120v N to G on left; 120v H to N on right; and 0v H to G on bottom, then it was reverse polarity

Granted, not all encompassing as a SureTest or equivalent, but much better that those 'night light' cubes, but as fast as those 'night light' cubes.

However, the advantage my setup had was going ??? when there was a bad neutral as the voltages could be any weird reading, all depending on the loads through the poor/loose neutral connection off 'ground'.

[Gunnar Alquist]

Jerry,

The other issue with your method is that the inspector has to actually pay attention and think when looking at meters. The advantage of the "night light" is that it can be done while half-asleep.

[Jerry Peck]

The advantage of the "night light" is that it can be done while half-asleep.

Gunnar, maybe that night light tester will wake them up?

[Gunnar Alquist]

Jerry,

Yours wouldn't do neutral/ground switched, would it? I don't see how it could. But, I don't believe the night lights will either, so no real difference.

Except that I end up donating a few of my night lights each year to unsuspecting homeowners. It would be a bummer to go to the trouble to make up a meter tester and leave that behind.

[Jerry Peck]

Gunnar, nope. It would not indicate neutral ground revered.

I haven't used a SureTest in so many years, did it pick up neutral/ground reversed?

H to ground is 120v; H to N is 120v; G to N is 0v ... resistance of N to panel is same as G to panel (unless open G, in which case would not read 120v H to G anyway).

[Gunnar Alquist]

Hi Jerrry,

I do not have a Suretest. I have the Amprobe version which does not. I cannot imagine how it could since they both go back to the same place, they just serve different purposes.

Looking at the Suretest information, I do not see N/G reversed as something that it can find

[Gunnar Alquist]

An update on the weird result that I described at the top of this topic.


I received a call from the listing agent (it was a pre-listing inspection). A potential buyer (I was told he is a retired contractor) checked the "reversed-polarity" receptacle outlets that I found and reported no power at all at several in the home. So, I agreed to go back out to see what I could find.


Indeed, four in the living room and two in the small bedroom did not work. I checked circuit breakers and none were tripped, and then checked for voltage at the load terminals, all of which were fine.


I then chose a receptacle that I felt was a good candidate for first in the circuit to pull out from its box (it was non-functional) and found scorched receptacle, scorched wiring, and oxidized copper conductor. After looking a bit closer, I realized that these were #12 conductors that were "stabbed" into the device and the connection had clearly failed. I find it interesting that my "night light" gave that anomalous reading in the same circuit (not the same receptacle though) as the reversed hot/ground. But, I cannot imagine how the first condition would be directly related to the second.


According to one of the online real estate websites, this condo was built in 1989 and my research tells me that UL standard 498 to only allow #14 wires to be stabbed was modified in 1996. Prior to that, stabbed #12 was acceptable, but I don't know how commonplace.


I realize that pulling receptacles out of the wall is beyond the scope of a home inspection, but I have now seen two sets of significant failures that are apparently caused by stabbing #12 conductors to devices. My own personal experience in 2020 and now this one. As a result, I am beginning to wonder if it wouldn't be a bad idea to inspect an occasional receptacle outlet in pre-1996 homes to spot-check how the receptacle is wired.


I realize that the chances of finding a bad connection in this manner are pretty low, but this might allow identification of potentially problematic stabbed #12 connections that the client can then be informed about.

I believe it would be too time-consuming to check all or even a "representative sample" of devices, but what about one or two in the entire house, at random? Possibly only if the panel label indicates 20A circuits to receptacles?


Obviously, my suggestion is well beyond the scope of a home inspection, and I am not sure how I would go about doing that; but, I was wondering... does anyone out there in InspectorLand do anything like this?

[Jerry Peck]

Gunnar, this is from personal experience too.

When we bought the house we had in Ormond Beach, which built in 1979, my inspector (who was a friend of mine) found what we presumed were loose connections in several connections in receptacles (erratic indications on his tester). Additionally, the receptacles were so old that just making contact with his SureTest and the receptacles was a problem.

I decided that I should replace all of the receptacles based on what we found.

After closing, I went to replace the first receptacle in the living room, removed the cover, removed the screws holding the receptacle to the box, pulled the receptacle outward so I could reach the wires ... however as I pulled the receptacle out, I was holding the receptacle and the wires were sticking out of the box, but still in the box.

All wires pulled out of their backstabbing holes. The wires may have been touching the contacts in the receptacle ... may ... but there was nothing making firm contact in the receptacle.

Almost all of the receptacles were like that.

Yes, #12 copper in backstabbing ONLY receptacles. There were no screw terminals to use.

That went with the FPE panel as an indication of the quality of the electrical contractor who wired the house. At least it was copper and not aluminum wiring. All boxes were good bakelite boxes, plastic, but good plastic.

[Gunnar Alquist]

... however as I pulled the receptacle out, I was holding the receptacle and the wires were sticking out of the box, but still in the box.

Jerry,

Which brings up another issue/problem. If an inspector were to pull a receptacle during an inspection and the connections pulled-out as they did in your case, what next? Having gone beyond SoP, it seems that the inspector would now have the responsibility to "fix" this. The receptacle cannot be left off and the wires exposed. Pushing the wires back into the holes might even cause other issues.

And then the current owner's inevitable response... "It was fine before you got here."

Additionally, the receptacles were so old that just making contact with his SureTest and the receptacles was a problem.

Yes, I routinely find "tired" receptacles on inspections. Nothing to do in that case but spread the prongs on the plug and keep going, right? (NO NO NO! Replace the receptacle!!!)

I have read various sources with different information on receptacle lifespan. I have seen as low as 15 years and as much as 50 years. I realize it depends on whether it is frequently used to plug in the vacuum or if it is hidden behind the credenza and never used. 50 seems a bit too long though.

Does the SureTest put enough of a load on the receptacle to check the connections? I guess I need to review my Amprobe instructions to see if mine can.

[Jerry Peck]

Having gone beyond SoP, it seems that the inspector would now have the responsibility to "fix" this. The receptacle cannot be left off and the wires exposed. Pushing the wires back into the holes might even cause other issues.

And then the current owner's inevitable response... "It was fine before you got here."

While we all know that the "real" answer is that the issue/problem already existed and the inspector "just found it" ... such a scenario as you described is why THE INSPECTOR SHOULD NOT pull receptacles from their boxes.

At this point, there are a few choices for the inspector:

1) Align all the wires (in my case, it was a feed through, so there were wires going in to that receptacle and wires coming out to another receptacle) ... ALL the wires with their respective holes and push them back in so that there are no exposed wires, reinstalled the receptacle enough to start the screws and hold the receptacle in place, then reinstall the cover to protect the still not safe condition as best possible. Then call an electrical contractor to install a new receptacle in that receptacles places BEFORE the inspector finishes the inspection and leaves the premises. All to reduce the inspector's liability to the least amount possible.

2) through however many options there are - it doesn't matter ... the inspector needs to call a licensed electrical contractor, explain what happened, and pay for the emergency trip out to replace the receptacle. Provide the electrical contractor with sufficient information to make sure that they have as compatible looking receptacle as possible to install as a replacement.

3) Now deal with the owner who is complaining that the new receptacle does not match the other receptacles in the room. But at least the safety issue as been addressed and cured.

I would not want to be that inspector. That would be worse than the time I (and the real estate agent - she pressed the button )had an overhead garage door fall off its track and crash onto childrens play toys below. I found the issue and, fortunately, it failed when no children were under the overhead garage door. There was no safety issue which needed an emergency service call to correct.

That was when I first said "Failed under testing", and explained that it was a very good thing that I found it when your (talking to a screaming seller) children were not under had you opened the garage door the next time.

(NO NO NO! Replace the receptacle!!!)

NO NO NO! a licensed electrical contractor needs to replace the receptacle, otherwise you, the inspector, is taking the liability for THAT receptacle and ANYTHING that happens after you, the inspector, touched it and replaced it.

or if it is hidden behind the credenza and never used. 50 seems a bit too long though.

Those are likely similar to NOS parts for cars. New Old Stock. Brand spanking new 40 year old parts.

Does the SureTest put enough of a load on the receptacle to check the connections? I guess I need to review my Amprobe instructions to see if mine can.

No tester puts pressure on the receptacle contacts. The receptacle contacts puts pressure on the prongs plugged into the receptacle. There are receptacle tension testers, but the SureTest does not have that function. I think I recall having one of those tension testers, but only using it a few times as they were, in my opinion, worthless. What pressure should one find on new receptacles? What pressure is deemed to be "safe" and "useable" on other-than-new receptacles?

Without documented standard pressure numbers are acceptable, and calibrated testers to measure, what could does the tester do? If you plug a SureTest or any tester, or any plug, into a receptacle and the tester or plug is 'loose' (yeah, right, define 'loose') include that in your report and include that all receptacles likely need to be replaced as loose plugs in loose receptacles are a potential hazard.

[Gunnar Alquist]

No tester puts pressure on the receptacle contacts.

Hi Jerry,
I was given one of those "tension testers" a while ago. I agree, it is spring-loaded and not reliable beyond being good theater for a demonstration to a client.

I was thinking about the stabbed (or side-wired) connections to the device in the box. What I meant was does the SureTest apply an electrical load that would test the quality of the electrical connection between the household (NM-B) wiring and the receptacle? For example, if I plugged a skilsaw into a receptacle outlet, it is going to draw about 12 amps (probably more to start). A stabbed (or loose) connection will not be able to carry the current and the saw will not operate properly.

Does the SureTest simulate or put some form of of electrical load on a receptacle? I am trying to imagine how something could test for current without generating a whole lot of heat in doing so.

[david shapiro]

I find this discussion fascinating. What I'd been told is that the standard was changed to restrict backstabbing to 14AWG, oddly, was that if a 12AWG wire had been used and subsequently someone removed that wire and stabbed a 14AWG into the same hole, the internal clip (undoubtedly the wrong term) would be too stretched out to hold the new wire securely. Your finding that the clips simply did not hold onto a 12 reliably makes more sense.

If I remember right, both my VD testers did simulate loads, very briefly.

[Jerry Peck]

If I remember right, both my VD testers did simulate loads, very briefly.

Gunnar, as David said (but I'm not sure if SureTest is one he used), the volt drop test does put a load on the circuit, VERY briefly.

Shortly after the SureTest testers came out, I called them with a few questions. My recollection of a couple of questions and answers are:

- Does the SureTest put an actual load on the circuit, and for long enough to create any heating of the circuit connections?
- - Yes, the SureTest puts a 15 amp load on the circuit when the voltage drop test is activated.
- - No, there is no heating of the connections.

- How long is the 15 amps applied by the SureTest? It doesn't seem like it could do it for very long without possibly tripping a breaker or blowing a fuse?
- - The SureTest applies the 15 amp load for 1/2 cycle?
- 1/2 cycle? You mean 1/2 of a 60 Hertz cycle?
- - Yes.

- Does the SureTest put a 20 amp load on the circuit for the 20 amp voltage drop test?
- - No, the 15 amp test result is mathematically converted to the 20 amp test result.

[Gunnar Alquist]

I find this discussion fascinating. What I'd been told is that the standard was changed to restrict backstabbing to 14AWG, oddly, was that if a 12AWG wire had been used and subsequently someone removed that wire and stabbed a 14AWG into the same hole, the internal clip (undoubtedly the wrong term) would be too stretched out to hold the new wire securely.

David,

That makes sense, but in this case I am certain these were #12 conductors. I just figure that this "internal clip" thing just doesn't have enough surface contact or pressure to securely attach the conductor. I forget the torque required for wiring to a side terminal, but I always (ok, maybe not "always and forever", but I didn't stab conductors) was concerned about the stabbed terminals.

I had a similar discussion a while back with Douglas Hansen. He told me that while teaching a class on electrical code and standards updates, he notified the class that #12 stabs in receptacles and switches had been discontinued and was asked by one of the electricians what size drill should be used to enlarge the hole for #12 conductor.

Gunnar, as David said (but I'm not sure if SureTest is one he used), the volt drop test does put a load on the circuit, VERY briefly.
- - No, the 15 amp test result is mathematically converted to the 20 amp test result.

Jerry,

Thank you. Good to know.

I just re-read my Amprobe manual and it looks like it does the same thing. I did use it on the inspection at the top of this thread, but not the failed receptacle and I wasn't looking for voltage drop, I was looking for reverse Hot/EGC.

I also (finally) realized that there is a second screen on the Amprobe that can be toggled that has more information. I need to play with this thing a bit more.

[Jerry Peck]

I was looking for reverse Hot/EGC.

Gunnar, that should be easy to detect.

Facing receptacle, using receptacle slots as measuring points:

properly wired
N to H =120V
N to G = 0V
G to H = 120 V

reversed N and H (reverse polarity)
N to H = 120V (N is now actually H; H is now actually N)
N to G = 120V
G to H = 0V

weird wired reversed G and H (don't know what to call it 'Can't Fix Stupid'?)
N to H = 0V (N is still N; H is now actually G)
N to G = 120V (N is still N; G is now actually H)
G to H = 120V (G is now actually H; H is now actually G)

reversed N and G (meter doesn't show the reversed N/G, reads same voltage between N/G as not reversed does)
N to H =120V (N is now actually G; H is still H)
N to G = 0V (N is now actually G; G is now actually N; G and N are bonded together back at the panel, both conductors would have, should have, essentially the same resistance/impedance)
G to H = 120 V (G is now actually N; H is still H)

Check to make sure I did that correctly, the above was a bit weird to do with a drawing to look at.