Request for data

[gar]

090429-2309 EST

Bill:

Neutral to EGC voltage has to be measured or calculated somehow even though it may not be displayed. Otherwise there is no way to calculate the EGC impedance back to the panel or the neutral impedance back to the main panel.

.

 

[wptski]

090429-2309 EST

Bill:

Neutral to EGC voltage has to be measured or calculated somehow even though it may not be displayed. Otherwise there is no way to calculate the EGC impedance back to the panel or the neutral impedance back to the main panel.

.

gar:

N-G is displayed. I said that L-G wasn't displayed.

 

[mivey]

Is my analysis of this instrument correct to this point?
.

Yes.

I'm not sure why I got one ohm reading for ground and a different one for the neutral at my panel as I only have a grounded conductor. My panel has a bonding screw, but the ground wire stops at the meter base behind my panel.

The meter might have some assumptions built in that may not always be true.

 

[mivey]

Ideal 61-165 readings

Ideal 61-165 readings

I was thinking the Ideal must put a load L-N and measure the rise relative to the ground. Once that is done, it could put a load L-G and measure the rise relative to the neutral.

If that is true, why did it give me two different readings for the neutral and ground at my panel, which are the same?

Perhaps some tests similar to your OP with some L-N loads with a ground reference, and L-G loads with a neutral reference would be in order.

 

[mivey]

I was thinking the Ideal must put a load L-N and measure the rise relative to the ground. Once that is done, it could put a load L-G and measure the rise relative to the neutral.

If that is true, why did it give me two different readings for the neutral and ground at my panel, which are the same?

Perhaps some tests similar to your OP with some L-N loads with a ground reference, and L-G loads with a neutral reference would be in order.

On that note (I've noticed a delay in how fast I can take the readings, while the voltage and load change, but it is not that bad):

No Load: L-N=122.5V, L-G=122.5V, N-G=0.00V, Load=0.00A
L-N Load: L-N=119.7V, L-G=120.3V, N-G=0.67V, Load=10.79A

No Load: L-N=121.8V, L-G=121.8V, N-G=0.00V, Load=0.00A
L-G Load: L-N=120.1V, L-G=119.4V, N-G=0.67V, Load=10.79A

Add: This was at the microwave outlet from my earlier posts

 

[gar]

090430-2021 EST
090502-1625 EST I have been starting this for several days.
090503-2216 EST

mivey:

Assuming that ideal measures the voltage from neutral to EGC while current is injected into neutral, then this is a measurement of the neutral impedance from the outlet to the main panel, and the ECG is simply used as a test lead to get a voltage reference at the main panel neutral.

A measurement of the outlet hot terminal to neutral for no current load and a known load allows determination of the voltage drop at the outlet for the load. From this one can calculate the total impedance of the transformer, service wires, meter, main panel breakers, and branch circuit wiring and outlet. This includes all connection resistances as well.

Next if the neutral impedance is subtracted from the impedance from the load voltage drop, then a value can be calculated for the hot impedance. This includes more than just the branch circuit hot impedance. From your data this looks like the Ideal approach.

Your post #5:
Microwave --- assume #12 copper wire, then distance from this outlet to main panel is about 0.07*1000/1.588 = 44 ft. Is this close?

The two kitchen counter outlets appear to be slightly longer branch circuit runs, or possibly smaller wire.

I do not believe there is any way for Ideal to measure the neutral impedance from the main panel to the transformer center tap.

Your measurement at the main panel of Zhot appears to be the the sum of all impedances of the transformer, the hot and neutral leads, and any other stuff in this series path. Ideal's measurement of about 0.1 ohms approximates the sum of the two values I calculated from your earlier data, 0.038+0.068 = 0.106 ohms.

Your last post with 0.67 V N-G and 10.79 A is 0.062 ohms. This is a good correlation with your Ideal measurement of 0.07 ohms.

I find that my 87 meter takes a little time to stabilize after a step change. Not so with the 27.

When you made measurements at the main panel with the Ideal and got a slightly different reading for neutral and EGC, it is was one digit, you may consider this to be instrumentation error. Both paths must be very near zero ohms and probe contact resistance or path differences may be the reason.


Bill:

Now to your data.

Until you get to the garage the vast majority of EGC impedances are lower, and substantially lower for many, than the associated neutral. Seems unlikely.

Also there were a number of hot impedances lower than the associated neutral.

50 ft of #12 copper is about 1.588/20 = 0.08 ohms. And #14 for 50 ft would be 2.525*0.00.13 ohms.

I measured two QO20 breakers including contact resistance and the values were 0.011 and 0.008 ohms at 12 A.

I have more comments but this has been dragging too long.

.

 

[mivey]

Next if the neutral impedance is subtracted from the impedance from the load voltage drop, then a value can be calculated for the hot impedance. This includes more than just the branch circuit hot impedance. From your data this looks like the Ideal approach.

I agree

Your post #5:
Microwave --- assume #12 copper wire, then distance from this outlet to main panel is about 0.07*1000/1.588 = 44 ft. Is this close?

Very. I measured 42ft, from the inside and made allowance for going over joists, box loop, etc.

The two kitchen counter outlets appear to be slightly longer branch circuit runs, or possibly smaller wire.

Longer runs. The receptacles are at the end of the daisy chain and behind a GFI receptacle.

 

[wptski]

090430-2021 EST
090502-1625 EST I have been starting this for several days.
090503-2216 EST

Bill:

Now to your data.

Until you get to the garage the vast majority of EGC impedances are lower, and substantially lower for many, than the associated neutral. Seems unlikely.

Also there were a number of hot impedances lower than the associated neutral.

50 ft of #12 copper is about 1.588/20 = 0.08 ohms. And #14 for 50 ft would be 2.525*0.00.13 ohms.

I measured two QO20 breakers including contact resistance and the values were 0.011 and 0.008 ohms at 12 A.

I have more comments but this has been dragging too long.

.

gar:

Speaking of my garage! Since those readings were taken, I rewire a bit. Nothing added, just actually replaced the NM between two receptacles. I found a few loose terminations. I retested with the Suretest which showed no drastic change!

 

[gar]

090511-1154 EST

So far there have been only two responders that have run the experiment of my first post. Are there not more people that are interested in finding out what information can be obtained from simple measurements on a residential system?

This thread has deviated somewhat to branch circuits and what information the Ideal instrument provides.

I have also deviated and have been doing some measurements on a couple of my branch circuits.

When circuits have good continuity, all connections are good, then voltage drop measurements I can make at a known load current adequately correlate with calculations based on wire resistance and length. In other words if rough calculations and measurements do not have a good correlation, then look for problems.

I have been running experiments in my kitchen area. Later I will discuss my results.


wptski:

Back to your data.

Consider living room 2.
Hot 0.15 ohms, Neutral 0.20 ohms, and EGC 0.08 ohms.

The neutral impedance is from the main panel neutral bus into your Ideal meter. This includes contact resistance at the neutral bus, wire resistance, contact resistance at receptacle screw, or what ever termination, contact resistance of plug blade to receptacle contact, and a little wire into the Ideal instrument.

The hot impedance includes all of the above points plus the circuit breaker, and all circuitry from the main panel to and including the pole transformer. I have measured a Sq-D QO20 at about 0.01 ohm. My guess is that what Ideal displays as hot impedance would be the total loop impedance minus the measured neutral impedance.

Hot being less than neutral does not make sense. EGC way less than neutral also does not make sense.

Following is some limited data from one of my kitchen circuits:

My kitchen circuit #9 to my pantry sub-panel has a neutral resistance of about 0.053 ohms (this calculates to 33.4 ft of #12), and the hot side is about 0.070 ohms including the breaker. The EGC is about 0.148 ohms. The EGC is about #16 and the current conductors are #12. 1965 Anaconda NM 12-2 w Gnd. The conductor resistance ratio for #16 to #12 is 4.016/1.588 = 2.52 based on copper tables. My measured ratio of resistances is 0.148/0.053 = 2.79 . This is a close correlation on the ratios, but I might have expected a closer value,

Summary of these values
Receptacle #9 to subpanel.
Neutral is about 0.053 ohms.
Hot is about 0.070 ohms and includes one breaker.
Sum of hot and neutral is 0.123 ohms.
EGC is 0.148 ohms.



Same #9 circuit:
Receptacle #9 thru subpanel to main panel.
Neutral is about 0.073 ohms.
Hot is about 0.094 ohms and includes two breakers.
Sum of hot and neutral is 0.167 ohms.

Total loop impedance including transformer is about 0.23 ohms.

None of these measurements include the resistance of the plug blade to the receptacle contacts. The Ideal device includes this resistance.


My experimental procedure in most cases does not require opening the breaker panel. The theory is that a high impedance meter requires only a very small amount of current for full scale relative to what is being tested. A 10 megohm input resistance meter could have an additional 10,000 ohms in series with its input and only produce a 0.1% error. Thus, other circuits going to a main or subpanel with no load on that circuit and of the same phase can be used as a means to connect the meter probe into the desired panel as an extension of the probe without removing the panel cover. This does not provide a means to get at a breaker output terminal.

This means that the reference circuit can have high resistance and have little effect on the circuit being tested.

Also I do not measure the voltage on the load side of a plug. Rather I use an adapter cable that plugs into a duplex receptacle that the load is also plugged into. This way I am measuring the voltage on the contact assembly of the receptacle. This eliminates errors from the plug to receptacle contact resistance.

Plug contact resistance is done as a separate test.

I will have more comments later.

.