The question that welders face on many jobs is : What size welding cable do I need for XX amperes when I am YY feet away from the power source?
This article — and the corrected sizes shown in Table 7 — will help you to select the right size cables for your welding or cutting job.
History
Welding cable size recommendations have been around for a long time. The oldest table about welding cable sizes that I found was in a book published in 1921.^{1}
It showed sizes recommended by General Electric Company. At that time American Wire Gage (AWG) was used to describe cable size. Sometimes AWG is referred to as “B & S gage,” which stands for Brown & Sharpe gage.
The AWG system is still used today. See Table 1 for a typical cable table.
The system is well defined and used by engineers all the time.
Cable Numbers
Small diameter cables have high gage numbers, and the numbers get smaller as the cables get larger.
After AWG #1 is reached, the next larger size is 0, then 00, then 000, up to 0000. The 0 to 0000 cables are sometimes referred to as 1/0, 2/0, 3/0 and 4/0, spoken as “one naught” or “one aut”, “two naught” or “two aut” and so on.
After that, larger sizes are described in “circularmil” areas.
Welding cable sizes usually run from AWG #2 through 4/0. It is interesting to note, that in the AWG system, a change of three sizes will double or halve the resistance. For example, a #6 cable has a cross section of about 0.0206 in^{2} and a #3 cable has about 0.0413 in^{2}, while a 1/0 cable has 0.0829 in^{2}.
CircularMils
Since wires are round, the cross section usually is measured in “circularmils”.
In this measurement system, one circularmil is the cross section area of a wire with a 0.001 in. diameter. This is a useful measurement.
Another useful measure is known as the “circularmilfoot”. One circularmilfoot of copper wire is 0.001 in. in diameter and 1 ft. foot long, and has a resistance of 10.37 ohms. This, too, is a very handy thing to know.
“Vintage” Numbers
The 1921 General Electric Co. recommendation1 gave data for three cable sizes in “circularmils”. (See Table 2).
The voltage drop in any copper cable can be calculated as follows:
V = (10.37) × (length in feet) × (current) / (circularmils)
Therefore, from table 2, at each maximum current, for 100 ft. of cable, the voltage drops are as follows:
V = (10.37) × (100 ft) × (200 amps) / (90,000) = 2.3 volts
V = (10.37) × (100 ft) × (500 amps) / (150,000) = 3.46 volts
V = (10.37) × (100 ft) × (1000 amps) / (260,000) = 3.99 volts
This calculation can be used today.
For example, from table 1, for a 4/0 cable, 200 feet long, at 400 amperes:
V = (10.37) × (200) × (400) / (211,600) = 3.92 volts.
More History
In 1928, “The Welding Encyclopedia” 2 stated the following: “When welding is done at great distances from the generator, the copper should be increased so that a drop in voltage across both leads (going and returning) should not be over 10 volts…”.
Just as now, it was possible, then, to adjust the generator — the power source — to make up for the voltage drop. In that article, the data was presented for various “B & S” wire gages.
The data is adapted here in table 3, for a 200 ampere example. With current cable size recommendations, the voltage drop is limited to 4 volts, not 10 volts. This will be explained shortly.
This adapted data (in Table 3) also appeared in the 1932 and 1941 editions of “The Welding Encyclopedia.”
Early Standards
In 1936, the National Electrical Manufacturers Association (NEMA) published a “Recommended Practice”. It was the first welding cable standard that I was able to locate, and stated the following:
“Cable Sizes for Standard Welders
For welding cable up to 90 feet in length, that is 45 feet welding cable and 45 feet return cable, the standard cable sizes shall be as follows:”
(Table 4 is an adaptation of the NEMA data, and shows a calculated voltage drop that was not part of the original data.) Note that all the voltage drops are less than 4 volts.
In 1938 the American Welding Society (AWS) published its first “Welding Handbook”.
It showed, for the first time, recommended welding cables based on a 4 volt drop. The same table appeared in the second volume in 1942.
These AWS cable recommendations were adopted in the first edition of what is now American National Standard Institute standard “ANSI Z49.1XXXX Safety in Welding, Cutting and Allied Processes”. That first edition was known as “American War Standard Z49.11944, Safety in Electric and Gas Welding and Cutting Operations.”
The AWS basic recommendations, from 1938 thru 1963, are shown in Table 5. Various versions of Table 5 have appeared in publications ever since.
Why a 4volt drop, and not 10 volts?
With previous recommendations for cable sizes, a job that requires a 40volt arc needs a 50volt output to account for the 10 volt — or 20 percent — loss of delivered power.
The recommendations for cable sizes in Table 5 lead to less loss of delivered power, so when a job requires a 40volt arc, a 44volt output is needed from the power source. That output provides adequate power in consideration of the expected 4 volt — or 9 percent — loss of delivered power.
Similarly, earlier cable recommendations would need a 30volt output for a job that required a 20volt arc. The 30volt output accommodates the expected 10volt — or 33 percent — loss of the delivered power.
With the cable recommendations from Table 5, a 24volt output is needed to deliver the 20volt arc to the weld. The 24volt output accounts for the expected 4volt — or 16.7 percent — loss in delivered power.
Therefore it should be obvious that a lower voltage drop is more efficient.
When the recommendations of Table 5 are followed, things will be efficient and safe. Cables will not overheat.
In fact, for the same 4 volt drop, longer cables will run cooler than shorter ones. Also, most welding is not continuous, so, even if cables started to heat up there would be cooling off intervals.
Another way to calculate the voltage drop
Instead of using the standard copper “10.37” number and “circularmils,” the information in Table 6 can be used to calculate voltage drop in cables.
The table gives the resistanceperfoot, of various copper cable sizes. The values are derived by dividing 10.37 by the values of circularmils from Table 1, so they present standard resistance figures based on the size of the cable used.
Using these standard resistance figures eliminates the need to use division to find voltage drops in cables.
For example, for 4/0 cable at 100 amps and 700 ft. long:
V = (ohms/foot) × (length) × (current)
V = (0.0000490) × (700) × (100) = 3.4 volts
Tables similar to this one appeared in the 4th and 5th Editions of the AWS “Welding Handbook”.
Voltage drop vs. duty cycle
There is no “duty cycle” factor to consider when calculating voltage drops.
Time is not involved when you calculate a voltage drop. However, the heating effect of current flowing in the resistance of a cable may become a factor if the cable is too small.
An undersized cable will get hot. In addition, undersized cables will have larger voltage drops than are recommended.
When you are welding with the gas metal arc (GMAW) process, or similar processes, the “voltampere slope” of the power source is of concern.
The voltage drop of the welding cables is part of the “slope”. The cable voltage drop should be kept to a minimum, so that it is the power source characteristic that controls the voltampere slope.
Some duty cycle recommendations have appeared in publications for welding cables. These publications even include the “Welding Handbook”.
However, unless you are welding with a process that is not concerned with the power source voltampere slope, avoid following tables with cable sizes based on duty cycles. For GMAW processes, and the like, bigger cables with lower voltage drops are recommended. That way, the effect on the system voltampere slope will be minimized.
The answer to the question
Table 7 is based on recalculating the required circularmil size of copper cables and a 4 volt drop.
The formula is as follows:
CM (circularmil) = 10.37 (amperes) (total cable length, feet) / (4 volts)
Using the calculated circularmil value and data from Table 1, a cable size can be determined.
Always pick the size closest and larger than the result of your calculation; not just the cable that is closest in size.
It turned out that recalculations reproduced the cable sized recommended in the First Edition of the AWS , Welding Handbook, published in 1938.
Except that, two of the 1938 cable sizes for 75 ft. (see Table 5, for 350 and 450 amperes) seemed to be typographical errors. The 1938 errors have been published and republished that way ever since. An updated table with corrected, highlighted values, is shown in Table 7.
The information in this article and the corrected sizes shown in Table 7 will help you select the right size cables for your welding or cutting job.
Table 2
Circularmils  Amperes 
90,000  Below 200 
150,000  200 to 500 
260,000  500 to 1,000 
Table 3
“B & S” cable size  Circularmils  Length feet  Calculated* voltage drop 

#2  66,370  320  10.0 
#1  83,690  400  9.9 
1/0  105,500  500  9.8 
2/0  133,100  640  10.0 
3/0  167,800  800  9.9 
4/0  211,600  1000  9.8 
*For 200 amps, V = (10.37) × (200) × (length) / (circularmils) 
Table 4
Current  Cable Size  Voltage drop* 

100 amps  #2  1.40 
200 amps  #2  2.80 
300 amps  1/0  2.65 
400 amps  2/0  2.80 
600 amps  3/0  3.33 
* From Table 3, V = (10.37) × (current) × (length) / (circularmils) 
Safe cable use — Things To Do
Here are a few things you can do to protect your cables and yourself.
Cables will last longer and work better if you follow these recommendations:

Turn off the power before connecting or changing cables.

Be sure to use the correct cable size.

Follow the manufacturer's recommendations.

Relocate cables to prevent tripping and entanglement.

Protect cables from damage by other equipment.

Keep connections clean and tight.

Use the right size connectors for splices.

Examine cables regularly for damage.

Repair or remove damaged cables.

And, use the correct length needed for a job. Remove and store extra long cables.
Table 6
AWG Size Resistance (Ohms/ft)  
6  0.000394 
5  0.000313 
4  0.000249 
3  0.000197 
2  0.000156 
1  0.000124 
1/0  0.0000983 
2/0  0.0000779 
3/0  0.0000618 
4/0  0.0000490 

”Electric Welding”, by Ethan Viall, McGrawHill, 1921 page 18

”The Welding Encyclopedia”. Editors Mackenzie and Cord, The Welding Engineer Publishing Co., 6th Edition, 1928, page 208.