[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [TCML] predicting phase shift
In the ideal world the inductance will remain reasonably steady. In the TC
world the AC resistance and DC resistance are in parallel, there is
definitely a change in inductance.
It will vary slightly, but every time it hits zero there will be a small
phase shift so we can not hit pure zero crossing with the prediktor
circuit. Close perhaps, but not exactly.
In solid state circuits, the changes in temp, varying current, and parallel
resistances, will change the zero crossing point. It does depend on the
thermal coeff. of resistance and this must be considered especially when you
are working with several thousand feet of wire. The geometrical mean must
be determined and applied. Nodal analysis helps out a lot.
Thermal coeff. of expansion, thermal coeff. of resistance must be considered
in solid state circuits especially at the 60-100 kHz frequencies. When you
hit a tolerance stack it can get out of line in a hurry, and thus changing
the zero crossing point. We are trying to keep the transient range to a
very low value. There will be inductance changes especially as power in
applied in the sec there will be thermal effects. As the caps also heat up,
the value will change slightly. Again, this will cause a shift in the zero
point crossing.
There is not a lot of inductance change in the primary circuit, but there is
in the secondary circuit. If it changes in the sec it changes the ring down
frequency, and this will definitely vary the
frequency. It you are detuned a bit on a solid state coil, you will blow
the IGBTs because the currents become incredibly high.
You have to measure the current in the gate driver xmfrs. The transistors
can not handle large resonance mismatches. You are driving the gate drive
transformer with a square wave pulse, and the tank circuit is not a square
wave. If you don't monitor the current value at the time the gate switches,
and the voltage is not also at zero, then severe thermal effects are
produced in
the IGBTs.
In classic circuits, this doesn't really apply, only with solid state
switching. In classic coils being out of tune only affects the pri cap and
power transformers.
Dr. Resonance
On Wed, Jun 24, 2009 at 7:23 PM, piranha <piranha@xxxxxxxxxx> wrote:
> Odd... The inductance of a secondary coil measured with a 1kHz hand very
> low current held meter matches the coil's inductance at many amps and 100's
> of kV during operation... :)
>
> For an air core inductor without saturation effects, the inductance value
> is pretty darn stable ;) Independent of current and voltage unless you melt
> it or arc it over. In high impedance circuits, Medhurst effects are seen
> but the inductance still stays the same. There are also some transmission
> line effects but those are also independent of current. There is also
> series resistance but, that is still independent of current. Current
> distribution in the wire along with relativistic effects are also
> independent of current.
>
> Inductance can change with frequency. If you measure a coil at say 100GHz
> you will get some odd readings. But in our case at say 100kHz, you just
> don't have to worry about it.
>
> If you see the inductance of your secondary inductor seems to be varying a
> lot, you left the base ground wire off :D
>
> Terry
>
>
> bartb wrote:
>
>> Wow, hate to even get involved in this thread, but...
>>
>> Sure, inductance will vary with the rate at which the current is changing
>> in the inductor (it has to). Phenomenon like current bunching in a coil,
>> long solenoid coils at high frequency, etc. will have a different inductance
>> in one portion of the coil as compared to another. The lumped average over
>> the entire coil would be the effective inductance which the coil will use
>> for it's LC resonance. Hard to measure things like that however and they are
>> best viewed with software. Software predicts this also. We've all heard of
>> high frequency inductances and capacitances.. This is simply "that".
>>
>> However, I have no clue what that has to do with Hammers circuit. The
>> bottom line is that the coil will have an inductance at some frequency.
>> Although it will vary from say a low frequency inductance, it will not
>> change enough at the resonant frequency to make any difference. The IGBT's
>> have a "known" delay, and the circuit is a simple way to provide a
>> consistent zero crossing prediction.
>>
>> Bart B. Anderson
>> Modesto, CA USA
>>
>> Lau, Gary wrote:
>>
>>> Inductance changes as the current through it changes? I don't think I've
>>> ever heard that before. As most inductors are used in AC circuits and the
>>> currents are constantly changing, that would be basically all inductor all
>>> the time. Please clarify?
>>>
>>> Gary Lau
>>> MA, USA
>>>
>>>
>>>> Any time the current is changing dynamically the inductance also changes
>>>> --- it pure physics and basic E.E., not smoke and mirrors.
>>>>
>>>> D.C. Cox
>>>>
>>>>
>>> _______________________________________________
>> Tesla mailing list
>> Tesla@xxxxxxxxxxxxxx
>> http://www.pupman.com/mailman/listinfo/tesla
>>
>>
>>
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
>
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla