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

Re: Variation of secondary Q



Original poster: "Paul Nicholson by way of Terry Fritz <twftesla-at-qwest-dot-net>" <paul-at-abelian.demon.co.uk>

Terry wrote:

> I have the HP 33120a arbitrary function generator...

Perfect.

> I have the Tek 3012 that does 10,00 point 9 bit captures and can
> put them in a data file to floppy disk.
> Sorry, that is 10,000 points not "10,00"

Fantastic.  0.02% accuracy on frequency, and 0.2% accuracy on Q.
That's assuming that the scope timebase doesn't drift with time. 
Solution if necessary: feed a calibration marker signal into
another scope channel, of known frequency.

> I have the low Z amp and Pearson 4100 1:1 140Hz-35MHz current
> probe.

Fine.  The method is insensitive to variations of drive amplitude
or pickup sensitivity or phase.

> I have a sheltered undisturbed outdoor spot I can put the coil
> and run coax indoors to the toys.

Great.

> Colorado State University's little weather station is only a few
> hundred yards away.  I wonder if its data is on the Internet
> somewhere?...

Drop them a line and tell them you're correlating the performance
of a delicate piece of electronics with temp and humidity - ask if
they can supply data, say hourly records?

> I don't think the scope will do automatically timed captures to
> disk

OK, just set the 33120 to a *very* long period inter-pulse.

>  The ring down current is tiny.  With 10V P-P in, I only get
> about a 200uA base current signal:

Malcolm wrote:
>  Any reason why you can't use the E-field to measure Q? 

Terry wrote:
> I could directly connect my 2.5pF 10Meg ohm probe (Tek5100)
> directly to the top of the coil

which might raise doubts about the stability of the probe Z, which
could potentially affect Q.  Better to use the E field pickup -
remember, the amplitude and phase is not important and can vary
from shot to shot.  No need to average several shots, the accuracy
can be achieved even with a single fairly noisy shot.

> The scope could then save a data file like this:

http://hot-streamer-dot-com/temp/TEK00000.CSV

The data is perfect - will cope with a noisier signal than this if
need be.  The size is a problem: you'll only get half dozen or so
of these onto a floppy, so if you record 1 shot/hour, you'd have to
change floppy 3 or 4 times a day!  Does the scope offer a more
compact binary format?  Maybe it can compress the files?

The other problem is to figure out how to time stamp each file.

Pity you don't have that RS232 option, would solve all these probs
and the PC could also alert you when something goes wrong with the
setup.

> If you can figure out the computer "A, theta, w, and r which give
> the closest (least squares) match to the waveform" part, I can
> supply the raw scope data files.

No problem, I can supply the code to convert those CSV files into
F and Q numbers. 

> So if I put the big coil's SonoTube secondary out there ...

Ah, your trusty 30" coil that has already proved so useful - yes!
Must count as one of the most useful TCs ever built!

> ...take a bunch of data files at various times and note the
> conditions the best I can... would that be ok?

Ideally it needs to be fully automated so that you can forget about
it for a couple of weeks at a time. At worst you should have to
visit the setup once every few days to swap floppy disks.

> How much of a ground plane is needed?  I am thinking of putting
> it on top of a shelf about 5 by 3 feet with very near by walls
> and roof.  I could do the split aluminum foil thing.

In order to establish which part of the coil/ground/environment is
affecting Q, it will likely be necessary to perform runs with
various configurations:  With/without groundplane to determine
sensitivity to ground losses;  With/without faraday cage to 
determine sensitivity to environment;  With/without coil in a 
sealed bag to determine involvment of coil former;  A week or two
in each configuration perhaps?

> I wonder if the Q variation is not so much the coil changing but
> the surroundings changing with weather?  The coil is far more
> stable for temperature and humidity than the ground, walls,
> ceiling, etc. 

Yes, hence the need to apply a process of elimination, runs in
different configs to isolate each possible factor.  Repeat with
the PVC coil former.

As you can see, potentially a long series of tests, hence best to
automate so that the results are standardised as much as possible.

> So I think we got the of the stuff here 

Definately.  You have the kit and the expertise, and this would
generate a valuable dataset of unequivocal measurements.

The only potential stumbling block is if the weather there doesn't
cooperate by being reasonably variable in both temp and humidity.

BTW, this would make an nice accurate Q detector for those current 
profile measurements.
--
Paul Nicholson
--