While reasearching my service upgrade I bought Rex Cauldwell’s Wiring a House, 2nd Ed., Taunton Press. I have picked up a few pointers and think Rex is a bit overboard in other places.
Rex talks about and has pictures of high end plug in analyzers. I have the cheapy that has the three LEDs that indicates somethings like open ground. But this SureTest model Rex pictures claims to be able to measure ground impedance in ohms.
Is there a cheap way to make this measurement without the tool?
I have never seen this tool for sale, who are the suppliers? What does one of these jobs cost?
Does SureTest have any competitors?
Big Macs – 99 cents
Replies
I don't remember the details, but David Shapiro had an article in FHB discussing ways to do some tests cheaply. Basically it involves finding a known good ground. He also feels that the cheap plug in ones are not guaranteed to work properly.
You should be able to test impedance with a digital multi-tester. They've come waaaaay down in price. My poor analog one was driving us nuts on the Nu-heat mat...
>>who are the suppliers?
http://www.Inspectortools.com (my favorite)
http://www.ProfessionalEquipment.com
I don't know about yours, but my church isn't a hotel for the holy, it's a hospital for sinners
Sojourners: Christians for Justice and Peace
Mac--
The Sure Test runs about $250 and up, depending on the model.
There is no easy or safe way to test grounding conductor impedance without a special tester like the Sure-Test. There are one or two other testers (by other manufactoreers) that'll do some of the same tests, but are about the same price.
Using a multimeter is not the same. The key is to measure the impedance under a load, and a multimeter doesn't do that. It measures resistance under a 9 volt, milliamp load. The Sure Test drops a 12 amp load (at 120V AC) across the conductors under test for 8 cycles, about one-eighth of a second, and measures voltage drop, then calculates impedance. It does the Vdrop test for hot-neutral, and hot-ground, and then tells you what the impedance is of each conductor. It's extremely helpful to know which conductor the bad splice or connection is in.
Before the sure test, I used a space heater (1500-1800 watts) and a multimeter to measure voltage drop, but I was always concerned that on a circuit with bad joints, the test itself could cause a serious problem (like burning open a connection, or starting a fire).
Those three-light polarity testers are useful but tell only part of the story. They can give false readings--like if there's a bootlegged ground (neutral jumpered to the ground terminal at the outlet), the polarity tester will show "ground OK".
The Sure-Test compares the impedance of the neutral and the ground condictors and if there's not much of a difference, it indicates "false ground". Of course, I've tested receptacle outlets close to the panel and gotten a "false ground" reading, because there wasn't enough of an impedance difference. Any test instrument or measurement has to be interpreted correctly before it can be relied upon. That comes with understanding how the test is done and the causes of erroneous results.
One feature of the Sure Test that I like is the GFCI test. The unit actually measures the trip current and time. The GFCI self-test button passes something like 30mA, several times the current that ought to trip a GFI. So a GFI self-test is more a check of the relay function than the leakage current. Better than nothing. I've picked up GFIs that self-test OK, but pass more current than allowed, or for longer than allowed, based on Sure Test results.
BTW, I've got three of the Sure-Tests, and they all compare closely to each other. Why three? First, they heat up in use, and if I'm testing every receptacle in a house, it slows me down to let a single unit cool down. Second, I bought the first one 15 years ago (before Ideal Industries bought the rights to make 'em), and the newer ones have much better capabilities (like AFCI breaker performance testing).
I do a lot of old work. Before I add on to a circuit in a remodeling situation (even if the load calcs show no problem), I sure feel better if I've done Vdrop and grounding impedance tests to get a good idea of the condition of the circuit.
And no, I don't have any connection to Ideal, the mfgr. It's just a good product.
Cliff
The Sure Test drops a 12 amp load (at 120V AC) across the conductors under test for 8 cycles, about one-eighth of a second, and measures voltage drop, then calculates impedance. It does the Vdrop test for hot-neutral, and hot-ground, and then tells you what the impedance is of each conductor. It's extremely helpful to know which conductor the bad splice or connection is in.
What do you think the chances are?
1) they have a precision resistor of about 10 ohms - RL.
2) they measure the open load voltage - Vo
3) they add the load resistor and measure the voltage across it and the calculate the current current through it VL and IL= VL / RL.
4) Now they assume Vo = (IL * RL) + (IL * Rg) where Rg is the the ground resistance.
This assumes the resistance in the black wire is null.Big Macs - 99 cents
Sounds 'bout right. But the instrument reports a value for the hot conductor impedance, so it it appears that they don't assume the impedance of the hot is zero.
I've seen cases where the hot and neutral impedance values are different, so I don't think they take the hot-neutral result and divide by two, attributing half to each conductor. And I don't think it's a random number generator...
The patent would probably reveal their method or algorithm. I've had the thought of getting the patent, and really delve into the details. But to paraphrase Bones McCoy on the original Star Trek, "Dammit Dave, I'm a sparky, not an electrical engineer!"
All I know is, it's a lot easier, quicker, and safer to use the tester than to haul a space heater (or hair [blow] dryer), multimeter, and calculator around. The tester has helped me identify and locate more han a few cases where there was a weak link in branch circuit wiring. Especially the grounding conductor, you know, the one that's like a military pilot's parachute. You don't need it very often, but when you do, it damn well better work.
I'd think that anyone doing remodeleing wiring on a regular basis (an electrician, or even a GC, to check the health of a circuit or the work of his subcontractor) would find the expense well worth it.
Work safe,
Cliff
"Sounds 'bout right. But the instrument reports a value for the hot conductor impedance, so it it appears that they don't assume the impedance of the hot is zero.I've seen cases where the hot and neutral impedance values are different, so I don't think they take the hot-neutral result and divide by two, attributing half to each conductor. And I don't think it's a random number generator...The patent would probably reveal their method or algorithm. I've had the thought of getting the patent, and really delve into the details. But to paraphrase Bones McCoy on the original Star Trek, "Dammit Dave, I'm a sparky, not an electrical engineer!""Well I r an electrical ingeneer, but I have not worked on something like this since school. And mainly what I remember from school was Edison's lectures about the dangers of Westinghouse's crazy idea of alternating current systems.Anyway you are trying to get 3 different values and that requires 3 INEDPENDENT TEST. Could not figure out where the 3rd one came from.Then it dawned on me.1. With load applied from hot to neutral measure the change in voltage between the neutral and ground. From that you can compute the neutral resistance. 2. With load applied from hot to neutral measure the change in not to neutral voltage. That gives you the combined resistance of hot and neutral. And subtracting off the neutral resistance found in #1 gives you the hot resistance.3. With load applied hot to ground measure the change in voltage between ground and neutral. That will give you ground resistance.4. As a confirmation with the load applied hot to ground measure the voltage change and will give you the combined hot and ground resistance. That should equal the sum of #2 and #3.
Agree with Bill.
I ar en EE two <G>
With a $3 HF meter, a couple of light bulbs, any calculator (or even a slide rule) and the E*I charts from any web site with tungsten light bulb characteristics you can measure anything on a 60 Hz single phase line except displacement factor (gotta add in a capacitor for that). The $250 for a fancy do-little meter is ok if you can't learn the basics.
Is there a test done right at the panel so there's a reference value for the neutral/ground interface?
"I cut this piece four times and it's still too short."
You could measure it at the panel. It will be lowest there. But you would like to measure it where you intend to plug the surge suppressor for your $5,000 plasma TV.Big Macs - 99 cents
Bill,
Ah so! Nice explanation. Thanks.
Cliff
I have not made the trip to the local electronics store, yet. What do you think my chances are of finding something like quantity 120 1W 1 Ohm resistors. With a nominal 120V input I can draw a 1 amp load and measure my ground impedance.Big Macs - 99 cents
chances are of finding something like quantity 120 1W 1 Ohm resistors.
Well, first, 120(what?) 1W 1 ohm does not compute - you mean 120 V, 1 amp, 120 ohms, > 120 W???
Think flexibly - 3 setting room heater works well, 240 V HW replacement heaters at 120 V work well, light bulbs work well (but need corredtion factor), old elec clothes dryer elements, etc. etc..
I agree with junk hound. Use a heating element (or heater) or a light bulb.But for best results you want to also use a clamp on ampmeter because the resistance of the load with vary as it heats up, specially the light bulb.But are you trying to measure the EGC for the ground electrode (earth ground)?
I don't have a clamp-on ammeter, I have a Fluke DMM I could put in series with a load. I'd need a fixture. Could be thing evolve but my original goal was to measure the resistance on the EGC. I am worried about surge supression. I never gave much thought to poor conductance on the black or the white.
I don't have any moral objection to running current on the green/bare for a test.
I was talking a whole bunch of those little carbon deals they call resistors. Like Cap says the meter heats up, that's a resistor. IIRC, resistors are sold in certain values ( Ohms ) and a rated power capability. The formula above calls for a known resistance which does not vary much with heat.
So, and remember I am not getting any younger, power = current squared times resistance. If I shoot for a 1 amp load I need 120 ohms. 120Watts = 1 * 120. This assumes a nominal 120 Volt input.
My opinion is that a light bulb's resistance will vary greatly with age and temperature of the filament. Short cycling it on and off will accelerate it's end and change it's physical properties quickly. Dryer element may be too low in ohms and pop the breaker on a 15 A circuit.
So a cheap precision resistor that's a 120 ohms in resistance and a 120 watts in power disapation was my quest.
Big Macs - 99 cents
Edited 12/13/2004 10:10 am ET by FarmerDave
Quikcly looking at my print Mouser catalog http://www.mouser.com and it looks like 1 watt resistors are about 77 for 10% units, but that brand has a minium value of 3.3 ohm.Here is another Xicon metal oxide. They have 1 to 5 watt units starting at 13 cents 5% units ans starting at .47 ohms. Those will work, but you still have a lot of connections and you will need to space them out and let air get in the middle or they will overheat.A better idea is to go with power resistors.They have 25 watt wire lead units for $1.09And metal case 25 watt 1% for 2.20 (and 50 to 300 watt units).4 25 watters or 2 50 watters ($2.75) would be the way to do.
1W resistors are hard to find anymore, at least at Radio Shack type places. A big city electronics distributor might have them, or you could order from somewhere like Allied (http://www.alliedelec.com).But keep in mind that 1 ohm is pretty low resistance, and putting 120 of those in series will result in a fairly high connection resistance. Better to buy higher resistance parts and put them in some sort of a series/parallel arrangement.Or you might be able to buy some big honkers from Allied. I've seen 50 and 100W devices cataloged in the (distant) past (and some that were probably 500W installed in some equipment). Try here: http://www.alliedelec.com/catalog/pf.asp?FN=1191.pdf If I were doing it I'd probably use a space heater (or light bulbs) and an ammeter, vs try to get precision out of the resistors.The device described likely uses only 8 cycles so it doesn't need to have high wattage devices (and a fan to cool them). For 8 cycles a 5-10 watt device can probably handle the load. However, 8 cycles isn't a good test from the standpoint of "smoking out" bad connections elsewhere that will only show up when they warm up a little.