A recent comparison, which I highlighted at the time was only a quick test done in poor weather and needed more work, resulted in my concluding that the generic Chinese mini-whip system has some broadband noise. This noise was absent when using the PA0RDT mini-whip at the same time.
 |
| My generic Chinese mini-whip system. |
I realised during an early morning hours session of reception today that the Chinese whip does not in fact suffer from broadband noise as a system. The noise, in fact, is transmitted by a mouse cable connected to my 1 year-old laptop.
In addition to mouse RFI, it became apparent that touching the laptop, notably the mousepad, also caused noise which was in fact considerably stronger than that transmitted by the physical mouse and its cable.
The level of the noise is not high, but is certainly of a magnitude where it would mask weaker NDB (or other) signals, and is thus a significant issue.
On closer examination, it's clear that the PA0RDT whip is also affected by the RFI, as the following waterfall demonstrates:
 |
| PA0RDT whip with wired mouse connected. |
Here's what the Chinese whip's waterfall looks like when listening at typical NDB frequencies with the wired mouse connected:
 | |
| Chinese mini-whip with wired mouse connected at the laptop. Note the constantly-repeating RFI. |
And this is what the waterfall looks like when the mouse is not connected to the laptop; I've tried to display the waterfall as close as possible to the previous waterfall:
 |
| Chinese mini-whip with the mouse disconnected. Note the lack of repeating RFI now. |
So that's cleared up: both whips are affected, to much the same degree, by RFI from the computer and its wired mouse. How about the signals? PA0RDT himself told me recently that he considers the signal-to-noise ratio reading to be a more meaningful measure of a whip's performance than, say, an 'S' meter reading, noting that a SNR reading is more difficult to obtain, given QSB and so on.
I looked at some low frequency commercial signals and one time signal, and the results are as follows:
189kHz,RÚV1, Iceland (peak signal strength and peak SNR)
Chinese: -78.5dBm, +12.1dB SNR
PA0RDT: -81.5dBm, +9.3dB SNR
198kHz, BBC Radio 4, UK
Chinese: -45.6dBm, +47.1dB SNR
PA0RDT: -39.7dBm, +47.3dB SNR
225kHz, Polskie Radio
Chinese: -72.4dBm, +17.2dB SNR
PA0RDT: -67dBm, +14.9dB SNR
252kHz, Radio Algérienne Ch. 3
Chinese: -60.5dBm, +30.6dB SNR
PA0RDT: -56.3dBm, +29.1dB SNR
60kHz, time signal, UK
Chinese: -52dBm, +51.7dB SNR
PA0RDT: -44.8dBm, +51.9dB SNR
Other than in the case of RÚV1 (Iceland), the PA0RDT yields significantly stronger signal strength than the Chinese unit, by a mean of 5.7dBm. In terms of SNR, the Chinese unit comes in at a mean of 1.24dB better than the PA0RDT (maximum of +2.8dB, minimum of -0.2dB SNR). The SNR difference is not significant and is likely to disappear if a longer test were conducted and this might also prove true with the signal strength. The two units can reasonably said to be equivalent, in my view.
I've also listened to the signals by ear and compared their quality. Whilst signal strength is comparable, there is very definitely more noise to the Chinese whip when listening, for example, to voice (plain AM mode) transmitted by Polskie Radio.
Here's a short clip from the Chinese unit:
And compare that with the output from the PA0RDT unit:
The next step will be to look at some real-world use of the signals from both units (without a wired mouse connected to the laptop) and see whether very weak NDBs are detected more clearly by one of the units. That will, of course, take some time to complete.
