toggle quoted messageShow quoted text
I really appreciate the clear fashion in which you (and Scott) explain the SiLabs chip's operation! You are correct; I did not realize that the chip tunes for proper inductance after it detects a change of frequency; I thought it "senses" the inductance continuously, and displays any signal strength changes during the "refresh" cycle of the RSSI indicator.
I'd forgotten about the maxim of best "Q" with a centered coil. After so many years of seeing stock loopsticks with factory coils positioned so far from center, I've not given it much thought. I like Scott's idea of starting coil winding right at the center, and seeing if there's any difference between a 330 uH and 550 uH coil.
The purpose of the black arrow marker in my video was simply to show the positioning of the coil for which I had previously found to give the best reception. With that precise point in mind it was easier to see and hear the signal fall-off when the coil was shifted. However, I agree that by not retuning the receiver after each shift it becomes a moot point.
Despite the flaws in the method, it still remains that I'm seeing a greater than 20 dB improvement over the stock loopstick, and that's excellent. If performance increases further with a coil at the rod's exact center, that will be even better.
I'm knee-deep in finalizing Wellbrook arrays and QDFA phased array antennas plus other preparations for a coastal DXpedition later this month, so I won't be able to return to ULR radio experimentation for a while. However, I'm sure Gary and others will be able to follow up on your and Scott's suggestions soon. I look forward to reading about these experiments of others.
Puyallup, WA USA
Re: Guy Atkin's Independent Testing of PL-380 7.5" Loopstick
Tue Feb 2, 2010 12:56 am (PST)
Looking at the video of the coil being moved indicates that the all that
is happening is that you are moving the coil away from a peak that has
already been established from a previous tuning operation that has set
the varactors capacitance to tune the loopstick at the black mark for
the coils inductance at that position on the rod. It appears to me that
the coil was originally tuned at the black mark then moved off peak
before the video started.
You cannot expect the Si4734 to track while moving the coil because it
does not know the coil has moved and keeps the capacitance unchanged,
but you have changed the inductance by moving the coil and of course the
signal will rises and fall when you move the loopstick.
Had the radio been retuned each time the coil was moved the Si4734 would
have returned its varactor capacitance for the new inductance brought
about by the movement and really the signal would not have changed by
much more than about 3dbu across the rod.
The only time the Si4734 changes its capacitance to match the loopsticks
inductance is during a tuning operation and this was not done in the
video. The capacitance does not change automatically when you slide the
coil. The whole idea of the Si4734 is it is self aligning. And that is
why you must step off and step back on frequency every time you move the
It would be very interesting to see the video redone showing the method
described above because every time the coil was moved and the frequency
retuned the audio and the signal levels (within about 3dbu) would have
come back up.
We must understand how this chip works and use this chip as it is
designed and not copy old methods from completely different technology
without applying them to this new technology correctly.
I think there has been a misunderstanding as to the correct operation of
how the Si4734 works and I now that criticism can be taken the wrong way
especially when someone has put a lot of work into their results and
presented them sincerely.
I really believe that the video showing a peak signal on the rod is
incorrectly presented and had the radio been retuned between movements
it would have only shown how cleaver the Si4734 really is at realigning
itself and not a signal dropping off from an established peak.