[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: grounding question



Original poster: "Gerry  Reynolds" <gerryreynolds@xxxxxxxxxxxxx>

Hi Gary,

I'm not as concerned about the lower frequency stuff (100KHz and below lets say) as I think the RF choke could easily be designed to block these frequencies. It is the high frequency stuff created by the "impulse" current of the strike that Im worrying about. If most of the return current for this impulse is in displacement current then maybe this is not a problem. If any of this impulse return current conducts up the RF ground wire connection to the TC base, the same could happen thru the green wire connection where the impulse current either travels thru the parasitic capacitance of the choke or arcs across the choke.

With parasitic coupling thru the NST from its core (RF ground) to its primary, a common mode noise spike results on the LINE and NEUTRAL. An EMI filter will help this greatly because of the common mode choke in the filter will turn this noise spike into a differential voltage between the LINE and NEUTRAL (on the line side of the choke) that can easily be filtered. However, with a RF choke between the RF ground and the safety ground, there is nothing to filter the noise to. One only gets an attenuation due to the impedance division between the choke impedance and the green wire impedance at this higher frequency of interest. Green wire impedance at high frequency is not well controlled and the parasitic capacitance of the RF choke could significantly reduce the impedance of the choke and result in a significantly large portion of the voltage spike being on the green wire. This is really an application that deserves some measurements.

Gerry R.



Original poster: "Lau, Gary" <Gary.Lau@xxxxxx>

Yep - any choke will have some parasitic end-to-end capacitance that
would couple RF from the RF ground to the green-wire ground.  But to
keep that in perspective, remember too that the NST core is connected to
RF ground, and that is even more capacitively coupled to the NST primary
and to the mains than you might get across a choke.  A very good reason
to use an EMI filter ahead of the NST, though I'm not sure most filters
are very effective below 1MHz.  A quick survey of Corcom filters shows
even their best 2-stage filters only attenuating 2db at 70KHz.  I should
probably rethink the filter that I'm using...

Regards, Gary Lau
MA, USA




> Original poster: "Gerry  Reynolds" <gerryreynolds@xxxxxxxxxxxxx>
>
> Hi Gary,
>
> Much of what you say is true.  When a streamer strikes earth ground,
> there is a return current to the base of the coil.  This current
> results in a voltage spike on the RF ground of a certain magnitude
> with respect to earth ground.  The magnitude of the spike may not be
> well controlled.  If an RF choke connects the base of the coil
> (counterpoise) to the green wire ground, this spike voltage will also
> appear across the RF choke.   Some have observed that the choke
> breaks down and arcs result across the choke.  The choke,
> unfortunately, is not ideal and a parasitic capacitance appears
> across the choke.  This parasitic could results in a significant
> portion of the "voltage spike" appearing across the green wire ground
> (that is also connected to earth ground) and may cause problems with
> other equipment also connected to the green wire ground.  Again this
> depends on the green wire topology.  If the green wire branch is not
> common with any other equipment, this will certainly help.  My
> thinking is the best way to not have a voltage spike across the green
> wire is to maximize the impedance between the green wire and RF
> ground (aka open circuit).  Of course, this requires that no one
> comes in contact with the RF ground during operation.
>
> Gerry R.
>
>
> >Original poster: "Lau, Gary" <Gary.Lau@xxxxxx>
> >
> >Hi Gerry:
> >
> >I'm not clear on why you suggest that when using a choke, any
streamers
> >must connect only with the counterpoise (which IS the RF ground), and
> >not with some other target not directly connected to the
counterpoise,
> >say a metal shelf.
> >
> >Both the counterpoise and the shelf may be viewed as separate
capacitors
> >to earth.  Should a streamer hit the shelf, the current flows:
> >streamer -> shelf -> Cshelf -> earth -> Ccounterpoise -> counterpoise
->
> >secondarybase.
> >
> >Since Ccounterpoise is finite and is an imperfect RF ground, there
will
> >be some significant RF voltage across the counterpoise and earth, but
I
> >don't see this as being any different than when using most
conventional
> >RF grounds.
> >
> >The shelf will also be at some significant RF potential above earth,
but
> >again, this is no different with or without a choke or with a
> >conventional RF ground.
> >
> >I believe that adding a choke between a counterpoise (or a
conventional
> >RF ground) and green-wire ground serves only to add safety against
the
> >RF ground becoming "hot" in the 60 Hz sense due to some fault
condition.
> >
> >
> >I agree that if a choke is used, attention must be paid to the
voltage
> >that it can withstand.  But assuming that it's properly designed
(TBD!),
> >I don't see why RF would enter the green-wire, regardless of what the
> >sparks hit.
> >
> >John - As with many topics relating to TC construction, there is not
yet
> >a yes/no consensus on this.  So far it's all just talk.  While it
seems
> >like a good idea, I've not heard of anyone actually using such a
choke.
> >
> >Regards, Gary Lau
> >MA, USA
> >
> >
> > > Original poster: "Gerry  Reynolds" <gerryreynolds@xxxxxxxxxxxxx>
> > >
> > > Hi John,
> > >
> > > Its sorta up to you.  If you do use a counterpoise and choose to
only
> > > ground it to RF ground,  You will need to keep people away from it
> > > while the TC system is plugged in.  If you do put a RF choke
between
> > > the counterpoise (thats still RF grounded) and mains safety
ground,
> > > you will need to make sure that a streamer doesn't hit a wall,
> > > ceiling, or something else that is earth grounded and not
connected
> > > directly to the counterpoise else return current could come up the
> > > green wire and arc across the RF choke.
> > >
> > > Gerry R.
> > >
> > > >Original poster: John <guipenguin@xxxxxxxxx>
> > > >
> > > >Well moving back to when I asked my second question, I god mixed
> > > >opinions on what should be done.
> > > >
> > > >When you use a counterpoise which is connected to the base of the
> > > >coil, should you install a RF choke to connect the counterpoise
to
> > > >mains safety ground? Yes or no.
> > > >
> > > >Thanks.
> > > >
> > > >On 10/16/06, Tesla list
<<mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx>
> >wrote:
> > > >Original poster: "Barton B. Anderson"
> > > ><<mailto:bartb@xxxxxxxxxxxxxxxx>bartb@xxxxxxxxxxxxxxxx>
> > > >
> > > >Hi Gerry,
> > > >
> > > > >If you or anyone else only connects the safety ground to the
variac
> > > > >(or control panel) and dont connect it to the NST or  PIG
chassis,
> > > > >or counterpoise (and many may not want to do this to protect
their
> > > > >house wiring), then these items must be considered dangerous
and
> >off
> > > > >limits while the system is plugged in.
> > > >
> > > >
> > > >Exactly! Touch nothing that isn't "green grounded" unless power
is
> >off.
> > > >Bart
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > >
> > >
> >
> >
>
>
>