The probe is a standard 10X Hi-Z
probe with a 10 Megohm input resistance in parallel with 10 picofarads.
Figure 3 shows the resulting 60 Hz pickup of the probe with a higher
frequency signal riding on it. The 60 Hz peak to peak amplitude is
about 600 mV. Near the center of the screen, the amplitude of the high
frequency component is about 700 mV. A new plot is shown in Figure 4
where the time scale is expanded from 5 ms/div to 20 us/div.
Figure 3. Probe Pickup at about One Meter Showing EMI Riding on 60 Hz Waveform
(Vertical = 200 mV/div, Horizontal = 5 ms/div)
The peak to peak amplitude in Figure 4 is also about 700 mV and the
frequency is about 48 kHz. This signal was picked up by the test lead
on the table over a meter from the lamp! In order to estimate the
source impedance of this voltage, a 100 kOhm resistor was placed across
the probe input as shown in Figure 5 to reduce the probe input
resistance from 10 MOhms to 100 kOhms.
Figure 4. Probe Pickup at about One Meter Showing ~48 kHz EMI
(Vertical = 200 mV/div, Horizontal = 20 us/div)
Figure 5. 100 kOhm Resistor Across Probe Input
Figure 6 shows the result. The peak to peak amplitude was reduced to
about 400 mV indicating that the source impedance was somewhat larger
than 100 kOhms, I suspect a capacitive reactance of around 150 kOhms.
This would imply a source capacitance (frequency was ~48 kHz) of 21
pF. The 21 pF equivalent capacitance would include the probe
input capacitance and capacitances to nearby objects from the test clip as well. The
experiment was performed on a wooden table so that should not affect
the calculation much.
Figure 6. Probe EMI Pickup at about One Meter Across 100 kOhm Resistor
(Vertical = 200 mV/div, Horizontal = 20 us/div))
The test was also run with a square sheet of aluminum foil about 1
foot, ~30 cm, on a side as shown in Figure 7. The peak to peak
amplitude was increased to about one Volt under this condition, not a
large change.
Figure 7. Overhead View of Foil Attached to Probe Input
When the probe and test lead were held close to the lamp as shown in Figure
1, repeated below for convenience, the recorded amplitude was much
larger as shown in Figure 8. A peak-to-peak reading of ~6.5 Volts
was recorded.
Figure 1. Electronic Fluorescent Light Near Scope Probe
(repeated for convenience)
Figure 8. Probe Pickup at ~5 cm Showing ~48 kHz EMI
(Vertical = 1 V/div, Horizontal = 20 us/div)
This kind of EMI would not likely be a problem for low impedance
digital circuits. However, some analog circuits can have a high
impedance at 48 kHz and could be affected. In fact, this Technical
Tidbit was written as a result of EMI from the same light fixture shown
in Figure 1 affecting some of my experiments over the last several years.
The waveforms of Figures 4, 6, and 8 also are rich in harmonics. Given
the amplitudes observed, enough high order harmonics to cause RF
interference to AM radios and other equipment is a possibility,
especially radios with high impedance antennas like AM car radios. A
small AM radio does in fact pickup broadband noise modulated by 60 Hz
in the vicinity of the lamp.