Manual FCKL 1528 page - 23
12.2.2.2 Logarithmic Scale
This scale permits a wide overview in a dB-linear scaling.
This is done by an analogue lin/log-converter.
In its right part the logarithmic scale can indicate 20 dB higher voltages, in its left part
15 dB smaller voltages than the linear scale.
Fast overview is easier this way, because switching of the attenuator is avoided.
On the other hand there are some limitations and problems, which could cause errors.
A) Errors caused by receiver noise:
Measuring within the frequency range of band B (150 kHz-30 MHz) and low distortion, a
noise level of approx. -14 dB is present on the logarithmic scale, which might be
considered as an interference signal caused by the d. u. t..
It is a good practice to listen to the loudspeaker to decide, if its noise or interference.
Disconnecting the input coaxial cable shows clearly if the signal is produced inside or
outside the receiver.
B) Overloading the receiver:
Even though the receiver is protected against damage especially in the protected input
mode, there is a danger of wrong measurement especially when slow pulses are
measured.
The situation is even worse, when the frequency spectrum shows big difference between
minimum and maximum, which is often the case with high power inverters.
Connected to a l.i.s.n. they may well deliver interference voltages up to some volts to the
EMI-receiver in the frequency range up to about 100 kHz.
At higher frequency the amplitudes are rapidly decreasing by 50 dB to 80 dB.
Such spectrum completely „consumes“ the available dynamic range of the receiver.
There is no safety margin to use the logarithmic scale near is +20 dB point in the low
noise mode.
Under these circumstances a superimposed slow pulse might be measured too low,
because the receiver has no more „breath“ left for the pulse.
On the other hand, the high voltages in the in the low frequency band might cause
intermodulation distortion which could result in measurements at higher frequencies which
are wrong, because they are produced in the receiver itself.
These problems occur far beyond the limits of standard measurements especially with
high power, slow pulse equipment.
Standard equipment such as PCs and microprocessors don’t show these characteristics.
It is good practice to check measurement with the receiver on the safe side, which means
low distortion on linear scale.