Eddy Currents Generated in a Ferrous Conduit
Eddy Currents Generated in a Ferrous Conduit
(OP)
Hey all,
I'm looking for the equation/s to determine the heat energy created in a galvanized rigid steel conduit (10 inches long 3.5in diameter)
The cables running through the conduit are 3 500MCM THHN Copper (all the same phase) at 1500A.
Ultimately I am trying to take that heat energy and transfer it back to the conductor core to determine the time it would take to melt the cable.
I'm looking for the equation/s to determine the heat energy created in a galvanized rigid steel conduit (10 inches long 3.5in diameter)
The cables running through the conduit are 3 500MCM THHN Copper (all the same phase) at 1500A.
Ultimately I am trying to take that heat energy and transfer it back to the conductor core to determine the time it would take to melt the cable.






RE: Eddy Currents Generated in a Ferrous Conduit
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: Eddy Currents Generated in a Ferrous Conduit
-Heat conduction from conduit to material around outside of conduit (air, ground, concrete or other). Depends on thermal resitivity of materials and their interfaces.
-Starting temperature of conduit, surrounding material and wire.
-Heat loss to cooling air flow through conduit.
-Radiation heat loss from exposed conduit.
-Heat transfer along wires to outside the conduit.
-Resitivity of the conduit.
-Method of attachment at each end. If the conduit terminates on a metal panel, the panel will also have significant heating, raising the ambient temperature.
500 amps per conductor exceeds the 430 Amp 90C rating of the THHN per NEC, so there may already be overheating problems. Also, my experience is that the metal can get hot enough to blister paint and smoke insulation within 10 minutes. We had to repaint switchgear steel after high current testing a large molded case breaker with improper routing of our test cables and bus bars.
RE: Eddy Currents Generated in a Ferrous Conduit
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: Eddy Currents Generated in a Ferrous Conduit
RE: Eddy Currents Generated in a Ferrous Conduit
Is it worth the time? Absolutely not. Unfortunately I have been tasked by my VP with trying to back up a field error that occurred and is being disputed.
I am aware of the results as we validated them via experimentation.(And your right, it took about 20-30 minutes to start smoking.) I am trying to determine the calculations involved.
Thank you for the posts.
RE: Eddy Currents Generated in a Ferrous Conduit
"THIS IS WHY!!"
Bill
--------------------
"Why not the best?"
Jimmy Carter
RE: Eddy Currents Generated in a Ferrous Conduit
I'll try to overcome this by supposing that it is an adiabatic process and no energy will be wasted outside .
I'll neglect the hysteresis losses also and I'll follow the Steinmetz formula for eddy current.
The steel pipe of 3.5" has inner diameter of 3.548 and the outside dia 4 inches. So the thickness will be 0.226 inch= 5.74 mm
The average dia =95.86 mm then the length of circumference will be pi()*9.586=30.12 cm
The field intensity will be H=1500/30.12=49.8 A/cm. From B=F(H) curve B=1.65 Wb/m^2[=16500 Gs]
Peddy=(pi()*B*t*f)^2/(6*10^7*d*ro) w/kg
where flux density (B) is in Gauss, frequency (f) in Hz, laminate thickness (t) in cm, density (d) in g/cm3, and electrical
resistivity (ρ) in micro-ohm-cm.
Peddy=(pi()*16500*0.574*60)^2/(6*10^7*7.8*10) =681 w/kg
conduit weight[10 inches long]=3.4 kg
Pfe=Peddy*weight=3.4*681=2315 w
Let's say all the conduit iron losses will penetrate inside and will heat the cables.
dQinput=dQoutput
{I^2*Ro*[1+alpha*(T-To)]+Pfe}*dt= TCAP*volume*dTemp [ if we shall neglect the outside heat dissipation.]
Integrate we'll get:
t*Kb/TCAP/volume=1/Kb*LN((Kb*Tf+Ka)/(Kb*Ti+Ka))
where Ka=I^2*Ro*(1-alpha*To) +Pfe and Kb=I^2*Ro .
Ro=1/58*long[m]/sqmm 3*500 MCM=910 SQR.MM
For copper Ro=1/58*0.254/910=4.81E-06
TCAP=3.42[ J/cm^3/dgr.C]
Extracting Tf from the above equation:
Tf=(exp(t*Kb/TCAP/volume)*(Kb*Ti+Ka)-Ka)/Kb
volume =439 cm^3
For t=232 sec Tf=1089 dgr.C