400HZ Feeder Cable

Status
Not open for further replies.
I am bidding some work where I will be feed some 400A 400Hz distribution panels and some 100A 400Hz sub-panels I am confused as to what type and size of wire I should be using for this installation. This is design build work so I do not have an engineered set of drawings I only have concept drawings. I have some spec sheets on 400Hz cable sent to me from my supplier however I am unsure as to what I need because the spec sheets do not show ampacity.

Does anyone know this information and where I can find it in the code book.
 

templdl

Senior Member
Location
Wisconsin
I've never ran across anything that addresses wire size in a 400hz application. What I do know is transformers are smaller because st 400hz they neeed less core. When you go from 60hz to 50hz there is more core in the transformer design as well as that of a transformer that is 25Hz. I believe that it is the same with motors.
But sizing the conductor for circuits current is current.
With breaker ratings becasue of the frequency you would have to consider the raing of the breaker for 400hz. As I recall 400hz works toward your advantage.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
I am bidding some work where I will be feed some 400A 400Hz distribution panels and some 100A 400Hz sub-panels I am confused as to what type and size of wire I should be using for this installation. This is design build work so I do not have an engineered set of drawings I only have concept drawings. I have some spec sheets on 400Hz cable sent to me from my supplier however I am unsure as to what I need because the spec sheets do not show ampacity.

Does anyone know this information and where I can find it in the code book.

You will not find it in the code book.

There is not a simple answer to this question.

Most circuit breakers manufacturers have charts and graphs that will show the effect of using them at 400 Hz instead of 60 Hz. However, as far as I know, none are UL listed at 400 Hz. Usually the AIC is greatly reduced and you need to increase the nominal rating of the CB.

Fuse manufacturers also have data for at least some fuses used at extended frequencies. most of them seem to end up using about 110% of the 60 Hz rating at 400 Hz.

Most of the contactor manufacturers have data for 400 Hz as well.

Conductor size is usually not that big of a deal at 400 Hz, especially with smaller sizes. My design criteria is for 900 Hz. I do not derate wire sizes for extended frequencies until I get over #6. However, the derating is a function of both the frequency and the cross sectional area of the wire.

I would probably use (3) 1/0 for the 400A feed and a single 1/0 for the 100A feed.
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
I am bidding some work where I will be feed some 400A 400Hz distribution panels and some 100A 400Hz sub-panels I am confused as to what type and size of wire I should be using for this installation.


I can't see why 400 Hz vs 60 would affect cable size. The issue is all I^2R loss and the subsequent heating.

Higher-frequency transformers have less iron and maybe fewer turns, but I can't see why the gauge would be smaller.
 

Hv&Lv

Senior Member
Location
-
Occupation
Engineer/Technician
I can't see why 400 Hz vs 60 would affect cable size. The issue is all I^2R loss and the subsequent heating.

Higher-frequency transformers have less iron and maybe fewer turns, but I can't see why the gauge would be smaller.

I believe that there is a difference in the ac/dc ratios with 400 hz. vs 60 hz. There are correction factors for other than 60 hz that are used to determine the ampacity adjustments.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
I can't see why 400 Hz vs 60 would affect cable size. The issue is all I^2R loss and the subsequent heating.

Higher-frequency transformers have less iron and maybe fewer turns, but I can't see why the gauge would be smaller.

Self inductance. Sometimes called skin effect. The current actually flows only in a band around the surface of the conductor with AC. As the frequency increases, the size of this band decreases. For smaller wires, the part of the wire where it conducts includes all or most of the wire, but as the wire gets bigger or the frequency increases, less and less of the wire is actually used.

At 60 Hz it is usually not significant, but it is why DC impedance of a particular wire is less than the same wire's impedance at 60 Hz.
 

suemarkp

Senior Member
Location
Kent, WA
Occupation
Retired Engineer
For our 400Hz circuits we:
  • Use aluminum conduit
  • Sometimes use "planetary cable", but I see that as sort of unncessary. Just upsize a bit or use normal paralleled conductors. There are some additional exceptions for paralleling wires at 400Hz that you can't use at 60 Hz.
  • Can't remember what brand and type of breakers/panelboards we generally use. But we never push the bus rating (a 200A distribution will generally use a 400A panel, and a 100A will use a 200A panel).
 

ghostbuster

Senior Member
We have had a number of 400HZ. sites where the moulded case breakers (200-400A.) would trip out on thermal overload.The higher order harmonics (eg. 5th-2000HZ. etc.) would heat up these breakers and trip.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
We have had a number of 400HZ. sites where the moulded case breakers (200-400A.) would trip out on thermal overload.The higher order harmonics (eg. 5th-2000HZ. etc.) would heat up these breakers and trip.

A chart I got from Square D suggests their 200A breakers used at 400 Hz are only good for about 184 A non-continuous or 147 A continuous on the thermal side.

On the instantaneous side, it takes about 2.6X the current to trip at 400 Hz than at 60 Hz.

The interrupting capacity at 400 Hz is about 1/10 what it is at 60 Hz.

It is a non-trivial design problem and why I generally use fuses for these kind of projects.
 

SOG38

Member
Location
USA
400 Hz cable

400 Hz cable

The wire characteristics are different for copper vs. aluminum and stranded vs. solid.
My practice has been to use only stranded wire and I have not had any issues.
I have worked with frequencies up to 10 KHz. It makes a difference if it in magnetic and non-magnetic pipe/ducts. It is different with 3 single conductors or a 3 conductor cable. Years ago I found impedance data for the various types of conductors
(This is a good source http://www.arcadvisor.com/reference.html#CopperData ). You also need to look at circuit breaker and fuse impedance at the higher frequencies. I have used Z Tot = R2+Z2 R and Z from the tables
Z is raised by the ratio of High Freq /to 60 or 400/60 = 6.666
I think what you will find is that the impedance will only increase maybe 2% at 400 Hz.
Hope this helps.
Sog
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
The wire characteristics are different for copper vs. aluminum and stranded vs. solid.
My practice has been to use only stranded wire and I have not had any issues.
I have worked with frequencies up to 10 KHz. It makes a difference if it in magnetic and non-magnetic pipe/ducts. It is different with 3 single conductors or a 3 conductor cable. Years ago I found impedance data for the various types of conductors
(This is a good source http://www.arcadvisor.com/reference.html#CopperData ). You also need to look at circuit breaker and fuse impedance at the higher frequencies. I have used Z Tot = R2+Z2 R and Z from the tables
Z is raised by the ratio of High Freq /to 60 or 400/60 = 6.666
I think what you will find is that the impedance will only increase maybe 2% at 400 Hz.
Hope this helps.
Sog

I think you will find it is not a fixed number but a function of the frequency and the cross sectional area so that as the frequency goes up the impedance increases increases by the ratio of the frequencies but the impedance also increases by the square of the diameter of the circular conductor.
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
SIEMENS400HZ.jpg
The wire characteristics are different for copper vs. aluminum and stranded vs. solid.
My practice has been to use only stranded wire and I have not had any issues.
I have worked with frequencies up to 10 KHz. It makes a difference if it in magnetic and non-magnetic pipe/ducts. It is different with 3 single conductors or a 3 conductor cable. Years ago I found impedance data for the various types of conductors
(This is a good source http://www.arcadvisor.com/reference.html#CopperData ). You also need to look at circuit breaker and fuse impedance at the higher frequencies. I have used Z Tot = R2+Z2 R and Z from the tables
Z is raised by the ratio of High Freq /to 60 or 400/60 = 6.666
I think what you will find is that the impedance will only increase maybe 2% at 400 Hz.
Hope this helps.
Sog

I think you will find it is not a fixed number but a function of the frequency and the cross sectional area so that as the frequency goes up the impedance increases increases by the ratio of the frequencies but the impedance also increases by the ratio of the square of the diameter of the circular conductors as the conductor size increases.
 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I think you will find it is not a fixed number but a function of the frequency and the cross sectional area so that as the frequency goes up the impedance increases increases by the ratio of the frequencies but the impedance also increases by the ratio of the square of the diameter of the circular conductors as the conductor size increases.

And interrupting ratings can go down because the higher frequency allows less time at the zero current point of the wave form for the arc to cool to the point where it will not re-strike as the voltage increases. At some point you end up with a nice high-frequency arc welder. :)
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
Self inductance. Sometimes called skin effect. ... At 60 Hz it is usually not significant, but it is why DC impedance of a particular wire is less than the same wire's impedance at 60 Hz.
You are correct, I was not considering skin effect. I'd never thought of it in the Hz. range, only in the microwave arena. {Waveguide is a magic hollow copper pipe that you pour microwaves in at one and it comes out the other...}
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
You are correct, I was not considering skin effect. I'd never thought of it in the Hz. range, only in the microwave arena. {Waveguide is a magic hollow copper pipe that you pour microwaves in at one and it comes out the other...}

Unless the waves are so big that they do not fit. :p
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
Thanks for all the replies everyone.

Just another quick question.



The cable I found is here http://www.servicewire.com/content/downloads/2012_Catelog/2-13-12 p 27.pdf.

So with this cable would I be okay to use the one with 6-250MCM on the 400A and the 6-1/0 on the 100A and then just pull a separate ground?

you really need to run the numbers on this kind of thing and not make guesses about what might and might not work.

according to my chart, a single 250 kcm 75C conductor at 400 Hz has an ampacity of about 190 amps, so I suspect it is probably adequate.

I would also point out that you should give serious consideration to using non-ferrous raceways/conduit.
 
you really need to run the numbers on this kind of thing and not make guesses about what might and might not work.

I did run numbers, this is what I used: 250kcm copper, with 7-9 current carrying conductors in cable bundle, 2 parallel runs (which I assume would this cable would account for both runs?), xhhw-2 90C insulation, termination temp of 75degrees C, gives me a derated ampacity of 406A, and If I replace the 250 with 1/0 I have 238A which is plenty enough for 100A. If I am wrong please correct me.

I would also point out that you should give serious consideration to using non-ferrous raceways/conduit.

Aluminum conduit is already going to be used :D
 

petersonra

Senior Member
Location
Northern illinois
Occupation
engineer
I did run numbers, this is what I used: 250kcm copper, with 7-9 current carrying conductors in cable bundle, 2 parallel runs (which I assume would this cable would account for both runs?), xhhw-2 90C insulation, termination temp of 75degrees C, gives me a derated ampacity of 406A, and If I replace the 250 with 1/0 I have 238A which is plenty enough for 100A. If I am wrong please correct me.



Aluminum conduit is already going to be used :D

The point I was trying to make is that the numbers in the code book are for 60 Hz. At 60 Hz you get 255A of ampacity for a single 75C 250kcm conductor. At 400 Hz, that 255A number drops to 190A. I would be nervous if you are trying to run both big cables in the same conduit. I would suggest running them in a cable tray - fiberglass is what i would use.
 
Last edited:
Status
Not open for further replies.
Top