Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
In a situation like a buck converter where the inductor current is basically constant and there is little ripple on the inductor, does skin effect apply? The voltage across the inductor changes a great deal but isn't skin effect only for changes in current?
I understand what you are asking. But as I understand skin effect starts affecting current flow at frequencies at about 500kc and higher. I thought it was higher, but I have seen inductors for the low end of the AM broadcast band wound with litz wire.
Isn't the upper frequency of switching power supplies about 80khz?
Skin effect starts happening for frequencies greater then Zero.
YES the effect at 1e-10000Hz is minimal BUT it is still there
The current in a buck converter cannot be constant, it is switching
Yes there will be a majority of a DC-component but there still will be an AC component on that, be it small and be it high freq.
I shouldn't worry abt it, since you want DC you want minimal ripple IF you had alot of ripple then you would be more concerned with getting rid of it
since you need to size the cable for the DC component and since the DC-component is the significant part that is all that matters. its not like you have a 1A DC component and then a 5A pk-pk at 100kHz then you might want to start concidering making sure you are not wasting copper (since that is all the skin effect is really to do with - why use 10m^2 when the current only flows in 2m^2)
and by the way there is no upper switching freq for switching PSU.
I have a 5W DCC from Crain at worki that is going at 600kHz.
The limiting factor is how much current you want to switch. The higher the freq the smoother the current is, but the switching losses go up proportional to switching freq.
Equally I have a 150kW converter switching 600A and that has an average switching freq of 3.3kHz
The question had just occurred to me because I was looking at what it would take to build a transformer supply capable of working off a wide variety of high voltage battery configurations. For getting power like 100 amps @ 12v I'm having trouble seeing how to do this with reasonably sized toroids. The conductors have to be huge and thus switching at 500kHz and using thin wire does not seem practical. A great many parallel conductors would be needed and it would be so sloppy on the core that leakage may become large. At least that's what it looked like.
Anyways, the real troublesome factor was trying to design it to work with a large range of possible input voltages and a preregulator was suggested which would keep me from having to design a transformer that would have to be able to work off so many different voltages. Then I thought, well then we've still got a similar problem with moving that kind of power density through a magnetic component... but apparently a buck with enough inductance that its current is fairly smooth wouldn't have skin effect problems thus high freq can improve the power density substantially.
Figuring out which capacitor to use is SO much simpler than figuring out inductors/transformers...
I didn't realize that there were high power convertors out there. I knew that the higher the frequency to rectify the less the ripple, and as a result the less the capacitance required.
I once mainted a high power RF transmitter that used a plate voltage of 15KV and the average plate current was 2A. The power supply used 3 phase at 60 Hz, so the filtering only required about 32 uF.
Skin effect is always there, just like capacitance. They are more pronounced at some point and then they have to be dealt with.
In a situation like a buck converter where the inductor current is basically constant and there is little ripple on the inductor, does skin effect apply? The voltage across the inductor changes a great deal but isn't skin effect only for changes in current?
I know that the skin effect occurs when we apply AC current on an inductor , so i don't know about current or voltage... i think the matter is with the frequency not the changes in current or voltage...may be i'm wrong i don't know!!
anyway here's a PDF about skin effect, it might be useful for you:
In a situation like a buck converter where the inductor current is basically constant and there is little ripple on the inductor, does skin effect apply? The voltage across the inductor changes a great deal but isn't skin effect only for changes in current?
I know that the skin effect occurs when we apply AC current on an inductor , so i don't know about current or voltage... i think the matter is with the frequency not the changes in current or voltage...may be i'm wrong i don't know!!
anyway here's a PDF about skin effect, it might be useful for you:
changes in current or voltage is what defines something as AC!!!!
it is hte current that causes skin effect, voltage has nothing to do with it, except in being the driving force to create the current flow
Gotta question for the pros on skin effect. I work with ESD and am wondering how the skin effect works on conductive polymers or similar materials. Dumba$$ mispelled my login name. Responses may be sent to robert.miller3364@sbcglobal.net
Urgent response please. I've got a patent up for grabs and I need this info!
In a situation like a buck converter where the inductor current is basically constant and there is little ripple on the inductor, does skin effect apply? The voltage across the inductor changes a great deal but isn't skin effect only for changes in current?
Yes you are quite right. The dc component just needs wire large enough to handle the dc current. Skin effect is only relevant for the ac current ripple.
Now suppose you have 500 Amps dc, and five amps Ac.
The wire will be fairly huge, to carry the dc component. The ac ripple will be reduced to the outer surface, but because the wire is so large, it will easily have enough surface area to handle the ripple without any problem.
Where you can get into strife is with transformers, and ac inductors, and particularly resonant inductors that carry a high circulating current. That may requires copper tubing, foil, or litz wire, depending upon the power level. It is also a particularly huge problem with transformers used in switching power supplies, because it is all ac.
But for a buck regulator, or any type of dc filter choke, with any sane percentage of ripple current, you can probably forget about skin effect.
The same applies to core loss. You will probably need to run ferrite with ac inductors and transformers, because of the good high frequency performance. For dc chokes, powdered iron is far more suitable, because it can carry much higher dc flux before saturation, but is much more lossy at higher frequencies.
Design your wires and cores for the dc, if the high frequency ripple current is fairly low, the chances are, core loss and skin effect can just be ignored..
ESD events occur instantaneously therefore are effectively an active high- current event. What will be the effect on a conductive polymer or similar material with a desired resistance of 100KΩ/square?
I would expect the voltage gradient to be fairly high at the point of actual "strike", but would reduce rapidly as the charge dissipated over the wider area of your "conductive" film.
Skin effect may be an issue, a lot depends upon the discharge rate, and that can vary with both the source impedance and capacitance it is discharging into.
An EMC testing lab can perform an ESD discharge test for you, with the standard human body model and standard "test finger".
All other factors being equal, skin depth increases as the square root of bulk resistivity. A very highly resistive material will have a large skin depth.
But your number of ohms/square is a sheet resistivity which already took the material thickness into account, skin depth is a factor of bulk resistivity.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
C from Crain at worki that is going at 600kHz.
