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

Re: Coil Formers.

TEST NOTES ON MEASUREMENTS
**************************

(1)  Signal Generator : Measured output impedance = 7 Ohms resistive
     over range of frequencies used in tests.
    The generator was built early this year for the purpose of coil
measurement. It has a fine tune control, and the output stage uses
a National Semiconductor LH4001 current buffer (slew rate 125V/uS),
BW = DC to 25MHz (+-10V into 50 Ohms).
Output : 950mVRMS unloaded (output was adjusted where appropriate
during the measurements).
A frequency counter with 1Hz resolution and calibrated is built-in.
It should be noted that the internal impedance of the generator is
going to lower the final Q readings somewhat. The error has been
disregarded in the data tables although it may be significant for
the smallest coils (as much as 10% !!!).

(2)  Oscilloscope : LEADER 3060D 100MHz Digital Storage scope.
Estimated reading error : +-5% max. due to resolution of graticule
and noise on pickup.
Y-Amp settings used were 10, 20, 50mV/Division.
FSD = 8 Divisions (generator and Y amp adjusted to give peak-to-peak FSD
at resonance with generator set to highest possible output for minimum
Y amplification).
-3dB measurements taken at 5.6 Divisions (70%).

(3)  Inductance Measurement : Sencore LC35 LC Analyser

(4)  Resistance Measurement : FLUKE 45  DMM
     Coil Resistances (measured) : closewound (h/d = 5) = 49.5 Ohms
                                   spacewound (h/d = 5) = 176.3 Ohms

(5)  Accuracy for frequency measurements : +-10Hz up to 500kHz
                                           +-30Hz up to 1MHz
                                           +-50Hz to 1.5MHz
Tolerance due mainly to noise of coarse tune control on generator.

ALL tolerances noted above are ABSOLUTE WORST CASE. In most cases, the
readings are claimed to be significantly better.

(6) MEASUREMENTS : There are 2 basic categories.
(a) Measurement of 2 basic coils whose h/d ratio was changed from
    5 to 0.56 during measurements (h/d error : +-1%).
    Winding Diameter = 6.6"
    Coil 1 was closewound with 0.56mm diameter enamelled copper wire
    Coil 2 was spacewound with 0.3mm diameter wire with a pitch to give
    approximately the same number of turns as Coil 1.
    Coil 1 was unvarnished to facilitate recovery of the wire, and Coil 2
    was lightly spray varnished to hold the windings in place.
Two coils like this allowed a comparison between the winding types over a
range of frequencies and also allowed measurements of the disks (below)
at two h/d ratios as the disks were reduced in size.

  Coil 1 (h/d=5) L = 57.45mH   (error +-0.05mH)
  Coil 2    "    L = 56.6mH           "
  N (est) both coils approx. 1360 turns at h/d=5

For the measurements in this category, two terminals were used : a sphere
of 9" diameter, and a toroid (outer diameter = 19.5" and inner dia.= 4.5")
The sphere was mounted at heights of 0.7" and 7" above the windings
and the toroid at heights of 0.7", 7" and 17.3".
Each coil was also measured for no-terminal frequency and Q with top lead
about 2" long.
For measurements at h/d = 5 to 2 inclusive, the bottom turn of the winding
was mounted 19.5 +-0.5" above the floor. For h/d = 1, bottom turn started
26" from floor and for h/d = 0.56, 28.8" above the floor.

(b) With Coil 1 at h/d=4 and Coil 2 at h/d = 3, measurement was made at
    four different heights above the windings of a number of 20gauge
    Aluminium disks, diameters : 200mm, 300mm, 400mm, 500mm, 600mm, 686mm.
    Diameter measurement error = +-2mm max.
For each disk mounted at the lowest height above the windings, the coils
were also measured with :
    (i) the bottom turn of the coil mounted 19.5 +-0.5" above the floor
        where the rest of the readings were taken
   (ii) the bottom turn mounted 13 +-0.5" above the floor (Fe concrete)

Disk heights were : 0.7", 7", 11.3" and 16.9" above the top turn.

(7) To facilitate accurate measurements, I chose this time of year in
    a large empty classroom cleared of desks/chairs and students. The
    building was almost deserted and computer/electrical useage in the
    building was at an absolute minimum.

(8) Finally, some suggestions on what to do with the data. You can
    calculate the coil self-C using Medhurst's formula and hence the
    AC resistance of the coils (including the generator). Note the
    trends vs. wire-size, coil-size etc. You can compare various in-
    ductance formulae predictions with the measurements. You can also
    note the techniques used, and in conjunction with these, you can
    check the effectiveness of the grounds you are using by using
    Mark Barton's coupling technique (injecting signal into the groun-
    ded secondary using a VERY loosely coupled primary). You can also
    check the Q of your coils when mounted at differents heights off
    the ground - it can make a whale of a difference. I did this for
    two h/d ratios and the results speak for themselves. Finally, using
    the calculated self-C for the appropriate coils you can calculate
    capacitances of the disks and terminals when mounted at different
    heights. The toroid has a vertical sliver removed to eliminate any
    possible influence on coil inductance although this has been shown
    in the past to be negligible.