Detached garage grounding

[pete_c]

A Ufer ground, obtained by connecting the grounding conductor to the rebar in your foundation, should provide a good earth ground.
 
In Florida the elevated ranch sits at sea level and mostly on sand.  Grounding does go to the rebar there. 
 
In the Midwest it is to the water ingress and another stake in ground.  Soil is totally different.
 
 
Measuring earth resistivity.
 
The effectiveness of grounding rods largely depends on whether the soil surrounding the rods can conduct large electrical currents. To design a buried grounding system correctly, you must measure earth resistivity with a ground resistance testing instrument. This instrument should also have switches to change the resistance range. You can use various test methods to measure earth resistivity, but the three most common are:

• 
  Four-point method, the most accurate.
  
• 
  Variation in-depth method (three-point method).
  
• 
  Two-point method.
  

After determining the soil resistivity, you are in a better position to determine what kind of buried grounding scheme will be most effective. Depending on the soil resistivity and grounding scheme requirements, the particular system can vary from a simple buried ground conductor to an extensive ground rod bed. The latter could include a grid system or a ground ring. To decrease the grounding system impedance, you can use ground enhancement material or chemical-type electrodes.

 

[kurtmccaslin]

Thanks for the response.  I feel better about my Ufer ground now.   The Fluke article explains the testing very well.   I can also see why a residential electrician would be hesitant to buy a $3K tester for the occasional customer who knows enough to ask about ground impedance.   Easier to just drive a second ground rod and be done.

 

[Del91574]

DEL


Very easy if you know the AWG and material (AL or CU) tables for standard resistance values are known and distance can be ballparked if the footage isn't labelled on the cable. You're not metering out the electrode itself, but the path. Meter between the ground clamp and the opposite lead.
 
-------------------------
Yes I understand that the impedance can be calculated if you know the wire type and length but I still don't see how you can use a meter to measure the impedance when the two points that you are testing are a great distance from each other. If I am installing a surge device in the Elk box that is 30 feet from the outside earth ground and I want to measure resistance from that ground point inside the box to the earth ground outside 30 feet away then my test lead on the meter would have to be 30 feet long in order to reach the rod from teh Elk box. No? Wouldn't that affect the meter reading? Wouldn't any conductor 30 feet long cause impedance that would be measured on the gauge?
 
Mike.

You subtract the added cable resistance. You can get the resistance of say, 100' of 14 AWG THHN as a known constant and extend the test leads accordingly, then subtract the RT from R1.  The length and AWG of your added test leads are known constants.

 

[mikefamig]

You subtract the added cable resistance. You can get the resistance of say, 100' of 14 AWG THHN as a known constant and extend the test leads accordingly, then subtract the RT from R1.  The length and AWG of your added test leads are known constants.

 
That is possible but is it practical? Have you ever actually run a test lead that long? According to a calculator that I found online the resistance of a 100' long awg14 cable is .0.25 ohm which tells me that pretty much any AC outlet in my house would serve as a ground for a surge suppressor. Am I correct? Am I missing something?
 
http://www.cirris.com/learning-center/calculators/133-wire-resistance-calculator-table
 
Mike.

 

[Del91574]

That is possible but is it practical? Have you ever actually run a test lead that long? According to a calculator that I found online the resistance of a 100' long awg14 cable is .0.25 ohm which tells me that pretty much any AC outlet in my house would serve as a ground for a surge suppressor. Am I correct? Am I missing something?
 
http://www.cirris.com/learning-center/calculators/133-wire-resistance-calculator-table
 
Mike.

I'd have to go dig out the books but the charts generally run temperature, circular mils and resistance which does vary slightly, but it is close enough.
 
I have done similar with TDR's and ballparking distances with known cable putups that I didn't have a known VP provided via the scope or manu.
 
You have to do what you have to do in order to test and/or connect to what you need to test. While it'd be great to use factory made test leads and ends, it's not a huge deal using bright copper, connected using known methods and then metering out the RT through them prior to connecting to what is being tested. Just adding the resistance to the RT obtained in the test itself.

 

[LarrylLix]

The wire resistance is not much of a factor. The major factor in the impedance is from the ground rod to the earth's crust.

The CEC states a minimum wire size of #6 awg. for mechanical reasons, not the current capacity.

The NEC is usually the same.

 

[RAL]

That is possible but is it practical? Have you ever actually run a test lead that long? According to a calculator that I found online the resistance of a 100' long awg14 cable is .0.25 ohm which tells me that pretty much any AC outlet in my house would serve as a ground for a surge suppressor. Am I correct? Am I missing something?
 
http://www.cirris.com/learning-center/calculators/133-wire-resistance-calculator-table
 
Mike.

 
For a device like a power strip surge protector, you usually don't have much choice but to use the AC outlet ground.  It's not an ideal ground, but usually it's your only choice and better than nothing.
 
You really want to protect your house from surges wherever wires enter the house from outside (as you know from your Ditek experience).  And you should ground those entry point surge protectors with a wire that has as direct a path to earth ground as possible - preferably with a continuous wire. 
 
An AC outlet doesn't usually give you that.  For one thing, the ground wire path from the outlet probably doesn't go directly to earth ground, but may be daisy chained through a bunch of outlets.  Each outlet usually means another splice in the ground wire and that increases the impedance of the wire.
 
The other thing you want to avoid is sharp bends in the ground wire.  Sharp bends also change the impedance of the wire and can cause a lightning surge to jump from the cable to any other conductive surface nearby. 
 
I don't know about your house, but in my house just about all the romex cables have sharp bends in one place or another as they make their way back to the main panel.
 
Remember, the ground wire in an AC outlet is a safety ground that is meant to protect you from a short in an appliance that you have plugged into it. It's far from perfect for surge protection.

 

[LarrylLix]

The outlet ground also maintains an equipment equipotential ground.

Connecting a device directly to it's own grounding point would endanger that device by introducing another potential between it's case and electronics.

This is why only one bonding point between any neutral system and the earth is permitted in a building. We don't want lightning and other common mode transients to find paths through your appliances to ground.

Common transients include your Uncle Eddie.

 

[Del91574]

The wire resistance is not much of a factor. The major factor in the impedance is from the ground rod to the earth's crust. The CEC states a minimum wire size of #6 awg. for mechanical reasons, not the current capacity. The NEC is usually the same.

Not necessary or required to be a #6. He needs either a #14 or #12 MAXIMUM. There is a difference here between a LV bond and a AC system ground bonding. In the case of surge protection, it is not necessary for a #6 to be used.
 
Mike is grounding a LV branch circuit. Grounding is not required until certain items are met. He's not being supplied from an AC transformer at greater than 150V primary; The transformer is not connected to an ungrounded system; He does not have supplying conductors being installed overhead.
 
All covered in NEC Art 250.112(I) and on the related tables. Using a #6 for this application is unnecessary, not required by code and a significant waste of money.
 
http://freenec.com/T90.html
 
https://www.industry.usa.siemens.com/automation/us/en/industrial-controls/products/Documents/Grounding_Conductors_0507.PDF