Voltage Drop Calculations with Amprobe INSP-3 Tester

Status
Not open for further replies.

glene77is

Senior Member
Location
Memphis, TN
Using an Amprobe INSP-3 tester.
Circuit Voltage is 121.5V RMS Unloaded.
Reading Volt Drop @ 15A = 4.2 %
Indicating 1.1% on hot leg and 3.1% on neutral leg.
Indicating Loaded Voltage = 116.4 V.
Indicating Ground Impedance = .27 Ohms.

I calculate Loaded Voltage = 116.4 V.
Anybody explain how and why the VD is different on hot vs. neutral legs ?
Anybody explain how and why the Ground Impedance is .27 Ohms ?

Sounds like one for Gar, or another experienced engineer. :)
 

jghrist

Senior Member
Anybody explain how and why the VD is different on hot vs. neutral legs ?
Anybody explain how and why the Ground Impedance is .27 Ohms ?

If there is more resistance in the neutral, there will be more voltage drop. 3.1% drop in neutral = 0.031?121.5 = 3.77V. R = V/I = 3.77/15 = 0.25 ohm.
 

glene77is

Senior Member
Location
Memphis, TN
If there is more resistance in the neutral, there will be more voltage drop. 3.1% drop in neutral = 0.031?121.5 = 3.77V. R = V/I = 3.77/15 = 0.25 ohm.

JG,
True, and thanks.

If you apply a 1KW load, and measure the VD,
then how do you differentiate the VD Hot vs. Neutral ?

This instrument does this kind of measuring,
and I would like to know the method.

I do not yet see that it is useful information.
I am curious "how" Amprobe does it, and "why"
so that I can understand if it is of any use to me.
 

bob

Senior Member
Location
Alabama
What size wire was used? Was it romex?
To what are you referring when you say "Ground Impedance is .27 Ohms"
If the hot and neutral were the same size and length there should not be a difference.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101206-2015 EST

My guess is there is a problem on the neutral circuit. If the wire size is the same on both hot and neutral then as noted by bob the voltage drop should be about the same. I would favor a little higher on the hot than the neutral because the breaker adds some impedance.

glene77is:

I suspect the Fluke does something like this:

1. Measure the unloaded voltage. Assume it remains sufficiently stable for the purpose at hand.
2. Apply a 15 load.
3. Use the EGC as a test lead back to the main panel neutral bus.
4. Then measure the voltage difference between EGC and neutral at the test point with the 15 A load. This provides information for the neutral path voltage change. From this the percentage is calculated.
5. Measure the hot to EGC voltage under the 15 A load. Subtract this from the unloaded voltage, and calculate the percentage.

The result for the hot side is a little distorted, because there is drop from the transformer to the main panel on both the hot and neutral from the transformer, and there is also the transformer impedance.

Keep in mine with the measurements I described above the reference point is the main panel neutral bus as connected to the EGC bus in the main panel.

If the reference point was the ground rod at the transformer and negligible current was flowing thru it, then you would have impedance measurements back to and including the transformer.

I just do not expect to see a higher voltage drop on the branch circuit neutral vs its hot wire.

Install an outlet at your main panel directly connected to the neutral bus and EGC bus, and the hot thru its own breaker. Now you have three tests points from the main panel to use anywhere you want. Run a long extension cord from this main panel outlet to the location you want to test. Use a small 1500 W heater to supply your load. Now you can directly measure the voltage drops and also see how much this load effects the voltage at the main panel.

For #12 copper wire the loop resistance per 1000 ft is about 1.6 ohms. Thus 1 ft is about 0.0016 ohms. A 50 ft run from the main panel will have a loop resistance of about 0.16 ohms. A 12 A load would produce a change of voltage of about 1.92 V, or about 0.91 V on the neutral and maybe a little more on the hot wire.

At my work bench I get a change of 5.2 V from 134.2 to 119.0 from a 1500 W heater. There are various wire sizes, and 5 circuit breakers in the circuit, and I believe 2 plugs. The neutral path voltage drop is 1.5 V, and that makes the hot path about 3.7 V. I just used the EGC at the bench for my probe back to the main panel.

.
 

glene77is

Senior Member
Location
Memphis, TN
What size wire was used? Was it romex?
To what are you referring when you say "Ground Impedance is .27 Ohms"
If the hot and neutral were the same size and length there should not be a difference.

Bob,
That is what I thought, neutral and egc should show same impedance, close.
But the Amprobe INSP-3 tester gives the information as stated in the original post.
It also does a test for Amp Short Circuit for neutral impedance and ground impedance.
 

glene77is

Senior Member
Location
Memphis, TN
By measuring the voltages with respect to ground...

Dbuckley,
Let me hold onto that thought for a moment.
So, I start with 121 V and load it, and have 116 V.
That is 116 V presented to the top of my load:
(1) referenced to Neutral I have 116V
(2) referenced to ground (egc I think) I have something different ?.
Neither Neutral nor EGC are "real" grounds.
Well, seems something is missing.

So, to take a different tack,
If I measure Voltage from the bottom of the load,
and reference it to Neutral, I have X Volts,
and reference it to EGC, I have Y Volts.
I cannot reference my Voltage reading to ground,
because I do not have acess to a ground reference
when I am measuring Voltages at fair distance out on the circuit.
And the Amprobe Insp-3 tester does not have access to any 'real' ground either.

I think I am missing something conceptual that the tester is doing.
Something that I should be able to do with my loads and volt-meters and amp-meters.
 

glene77is

Senior Member
Location
Memphis, TN
101206-2015 EST

My guess is there is a problem on the neutral circuit. If the wire size is the same on both hot and neutral then as noted by bob the voltage drop should be about the same. I would favor a little higher on the hot than the neutral because the breaker adds some impedance.

glene77is:

(more . . . . . .)

Thank you, GAR !
It will be tomorrow morning before this settles in good.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101207-2130 EST

glene77is:

In my post number 5 I believe I provided you with the answer of how the Amprobe is functioning. But keep in mind it is a guess, although logically it is what you can expect from a measurement at the test point.

It is basically using the EGC conductor as a long lest lead back to the neutral bus at the main panel. This will not provide good information on the impedance of the hot lead from the main panel, because more than this segment of wire in involved. Should be a good measurement for the neutral from the main panel to the test point.

.
 

glene77is

Senior Member
Location
Memphis, TN
101207-2130 EST

glene77is:

In my post number 5 . . .

.

GAR,
Thanks for sharing your experience.

Thoughts:

(1) On old circuits, the EGC was smaller than the conductors,
thus resistance of Neutral and EGC will measure differently.

(2) Could the Neutral be used to reference the EGC (ground) resistance,
or is this even necessary to mimick what the Insp-3 is doing ?
I would call this a two-pass double-blind test.

So, today moves forward. :)
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101208-0841 EST

glene77is:

(1) On old circuits, the EGC was smaller than the conductors,
thus resistance of Neutral and EGC will measure differently.
Yes. I believe that the INSP-3 will just shift the load path from neutral to the EGC to to test the EGC resistance and use the neutral for reference. A problem might arise here if there is some other load on the circuit when the test is made. This will depend upon the design of the INSP-3.

I am assuming you are doing these experiments in your home. Therefore, you have more flexibility in what you do.

I think you need to do some tests with an ordinary DVM as I described above to verify that the INSP-3 measurements are correct.

The INSP-3 will use only a short burst load, quite possibly only several cycles. If we assumed a maximum test rate of one measurement every 10 seconds, then there are 600 cycles in this period. If a 10 W resistor was used as the load, then the maximum number of cycles per test at 20 A would be (20*120 = 2400 W, 2400/600 = 4 ave watts per 10 seconds per cycle of 60 Hz) 2.5 cycles. In an instrument of this size maybe one full cycle is used. Or they could also do a short pulse test at the peak of the AC waveform.

So I am suggesting you do an independent test on your neutral path resistance, and compare this with the Amprobe measurement. I am interested in how this compares with the Amprobe measurement.

Assuming your neutral resistance is high as it appears to be, then I would look along the neutral path of this circuit with a DVM and a reference from the main panel to see if you can spot where there is a high resistance connection.

At a duplex at my main panel the voltage difference between neutral and EGC is about 0.001 V with my TED system turned off. The TED system uses carrier current signalling and introduces a lot of unwanted noise for this test, at least 20 MV. Adding my 12 A load to the other side of the duplex the difference between neutral and EGC is about 0.1 V. Approximately 0.008 ohms. #12 wire alone might account for 0.0016*3 = 0.005 ohms.

If the voltage difference between neutral and EGC at your main panel looks good, then use an extension cord to bring the EGC reference at the main panel to your test location. Go to each outlet on the circuit and measure the voltage difference between neutral and the EGC reference. Have your test load, 1500 W heater, in the last outlet of the circuit. If the load is not in the last outlet, then the voltage measurements will indicate this.

What are your results? This data should point to the high resistance source. Do you have back-stab receptacles?

.
 

glene77is

Senior Member
Location
Memphis, TN
What are your results? This data should point to the high resistance source. Do you have back-stab receptacles?

Gar,
Will send results when available, maybe a week, on this schedule.

Don't assume a problem, such as high resistance source.
This is a developmental question, based on the published data in the manual.
This is (1) to explore an understanding of methodology, and
(2) establish criteria/requirements prior to purchase.

No back-stabs,
replaced everything in the house with push/clamp types, and upgraded to GFCI as required.
 
Last edited:

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101209-1214 EST

glene77is:

To try to evaluate the Amprobe INSP-3 you might try the following approach.

Roll out about 50 to 100 ft of #14 Romex. #14 for higher resistance, but use #12 if that is more readily available. All new Romex of these sizes will have the EGC equal in wire diameter to the current carrying conductors. Rolled out to reduce the inductive effects of being coiled in a box.

This piece of Romex becomes your test item. Probably one big loop so the input and output ends are near each other. Possibly terminate both ends on terminal blocks, and from the terminal blocks to a short cable with a plug, and at the other end to a duplex receptacle. There should be single wires used so that a clamp-on current probe can be used.

With this setup you could do some controlled experiments using the INSP-3 and other test equipment. You can calculate the wire resistance, but use my copper wire values from the Reference Data for Radio Engineers. These values are 20 deg C figures of 1.588 ohms/1000 ft for #12, and 2.525 for #14.

If you have a clamp-on current probe that can connect to your oscilloscope, then you will be able to see the test current waveform that the Amprobe generates.

Using a separate piece of Romex will provided a better means of evaluating the INSAP-3 than trying to use unknown wires in a wall with unknown terminations.

In this test the hot wire impedance should be substantially higher on the INSP-3 than the neutral. This is because I do not believe the INSP-3 has any way to sort out all the hot impedance that is not part of the Romex cable. Meaning the breaker for the branch circuit, the main breaker or fuse, the service drop (both hot and neutral), and the transformer.

This Romex test should be run by jumpering the EGC to the neutral of the Romex at the terminal block at the input end, and not connecting to the EGC in the plug at the input end. This is for the purpose of running a more controlled experiment. The INSP-3 should get an almost identical impedance mesurement for the neutral and EGC wires.

.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101209-1711 EST

glene77is:

I picked up a 250 ft roll of #14 Romex. With black and white in series I measured a total loop DC resistance of 1.3 ohms. Probably not worse than +/-0.1 ohm error.

Calculated resistance for 500 ft is 2.525/2 = 1.2625 ohms.

Then using a 60 Hz AC test with 10.52 A current the two voltage drops were 6.90 and 6.82. Summing these the loop drop was 13.72 V. Total loop resistance was 13.72/10.52 = 1.30 ohms.

The correlation is good for the accuracy of the experiment. Whatever inductive component exists in this coil of wire is insignificant. It should be small because it is basically a non-inductive type of winding. In other words an opposite phase current flows in one wire relative to the other largely canceling one another.

.
 

glene77is

Senior Member
Location
Memphis, TN
101209-1711 EST

glene77is:

I picked up a 250 ft roll of #14 Romex. With black and white in series I measured a total loop DC resistance of 1.3 ohms. Probably not worse than +/-0.1 ohm error.

Calculated resistance for 500 ft is 2.525/2 = 1.2625 ohms.

Then using a 60 Hz AC test with 10.52 A current the two voltage drops were 6.90 and 6.82. Summing these the loop drop was 13.72 V. Total loop resistance was 13.72/10.52 = 1.30 ohms.

The correlation is good for the accuracy of the experiment. Whatever inductive component exists in this coil of wire is insignificant. It should be small because it is basically a non-inductive type of winding.

In other words an opposite phase current flows in one wire relative to the other largely canceling one another.
.

Gar,
Thanks. :) You are most encouraging.
I plan to repeat this experiment, and use your others to build some better understanding. I have a fresh roll of 12-2.
Need to review your prior posts, also.

BTW, one day a Master tried to explain to me that the difference in impedance between D.C. and A.C. was tremendous. I did some calculations, and figured that the difference was "only" in the neighborhood of .02 %. (by memory)
Without a re-calc, it supports your effort.
Needless to say, that did not earn me any points on that job.

I guess I'll have to install an electronic ignition on this VW van of 42 years age.
Have spent the last week fighting with failing points,
repeatedly pulling over to file the pits.
I put together a kit several years ago, need to install it,
The sputtering and coughing is worse
than the usual 'shake, rattle, and roll' of an air-cooled antique Van.
My van is especially air-cooled with these temperatures going down to 20 F.
All the other electricians I know have fairly new 'leased' vans with heaters.
I always invite them to laugh at mine, 'cause it is paid for ! :grin:
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
101210-2127 EST

glene77is:

Replace the points and the capacitor across the points. A good capacitor is very important. Also if you can find points for the VW with a hole in the middle of one contact I believe you will get longer point life. This was an invention of Ford in the early 50s. I don't think it was the late 40s.

Mechanical contacts in a DC circuit show a migration of contact material in one direction producing a mound on one contact and a cavity in the other. The hole in the contact where the mound grows allows some of the metal to blow thru the hole.

The importance of the capacitor is that it reduces the rate of rise of voltage when the points open. It is also part of the compound resonant circuit of the primary and secondaries. Basically there are two tuned circuits with mutual coupling.

.
 
Status
Not open for further replies.
Top