G8MNY > TECH 07.10.18 09:45l 256 Lines 11637 Bytes #2 (0) @ WW
BID : 40812_GB7CIP
Subj: Reducing Electronic RF QRM
Sent: 181007/0733Z @:GB7CIP.#32.GBR.EURO #:40812 [Caterham Surrey GBR]
To : TECH@WW
By G8MNY (Updated Aug 07)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)
There are 3 principles to removing QRM & 1 for putting up with it.
1/ Stop the RF being generated at the source. RFI Enforcement?
2/ Stop Source RF from radiating.
3/ Stop Rx system seeing the RF QRM.
4/ Living with it.
1/ In electronic systems the use of radio frequencies is common, but the need
for wideband signals is less common.
Lets look at a logic gate that runs at 1MHz but uses a fast 40MHz totem pole
output stage. This produces a high current power rail pulse @ > 120MHz on
the switching edges.
Gate │ │ High bandwidth
Current ───┴──────┴─── High QRM
______ │ `\ ______
/ 1MHz \_____ ──┤Gate├── │ │____
Low Bandwidth │__./ 40MHz edges
low level QRM │ Medium Bandwidth
0V ──┴─── Medium level QRM
The output fast edge shape is often wanted, but power rail noise is not.
Filtering is normally provided on the power rails to provide an on PCB path
for these high pulse currents..
+5V ──┬─────┬─── 5V ──┬─┴───┬──┴──
│ │_ But due to ) │_
=== │ `\ PCB track 1uH ) │ `\
10n │ ──┤Gate├── lengths this ) ─┤Gate├──
│ │__./ often looks 10n │ │__./
│ │ like this... === │ \│/
0V──┴─────┴─── 0v ──┴─────┴──────┴──
One cure is to put a C directly across any noisy ICs, these can identified
by scopeing the power rails on a PCB looking for the worst RF noise sources.
Then putting a Cs (surface mount?) across the power pins without loads of
track en-route.... Cs of 10n-10u in the right place can make a huge
difference to power rail noise radiation.
\│/ \│/ Much less
5V ──┬─┴───┬──┴── RF Noise
) ┌──┤_ on PCB It is not usually possible to
1uH ) ===│ `\ add extra L to the PCB power
) │ ┤Gate├── rails to help the decoupling.
10n │ │ │__./
=== └──┤ \│/
Data bus radiation is more difficult to cure as there are more lines & the
RF power level is normally quite low from any particular line.
But filtering of a R or LC is sometimes effective (e.g. on a PC printer
port), but both type must consider the timing delay & loss & any undamped
resonance on the data line.
__ QRM __ Much less __ Very little
│ `\ \│/ │ `\ \│/ │ `\ \│/
──┤Gate├──┴─ ─┤Gate├─100─┬┴─ ─┤Gate├─((()─┬┴─> Z load
│__./ ___ 1MHz │__./ | stray │__./ L │
│ │ │___│ │ === C │ 1uH === 1n
───┴────── ──┴─────────┴── ─┴──────────┴─
Noise Sources have less With R C the RF With L & C the
dB than -6dB/Octave Harmonics drop Harmonics drop
│ │ | for odd harmonics @ 12dB/O after @ 18dB/O after
│ │ │ │ | -3dB point the damped
│ │ │ │ │ │ | where Xc=R resonant Freq.
0 1 2 3 4 5 6 7 8 9 11 MHz
The actual radiation from PCB track is proportional to the frequency in
addition to any resonant track. So a 1MHz square wave may be very weak at
11MHz as the PCB track length is very small compared to a 1/4 wave, but
radiate quite loudly at 50 to 144MHz where the track becomes a good aerial.
These principles also apply to SMPSUs as well as TV/Monitor Line stages,
just that it is very difficult to apply in those high power cases.
Quite often to save money makers drop the use of RFI components in mains
SMPSU. e.g. they have bought in GREY illegal imports of unfiltered SMPSU, &
put them in their PCs etc. hoping no one will notice! If you have a look
look inside one of these offending PSUs you will see the straps that bridge
out the missing C-L-C mains filter.
GOOD FILTER NO FILTER!
L >───┬────┬──┐ ┌────┬─> to Bridge L >──┬────┬──----───┬──> to Bridge
2n7=== │ (((() │ Rectifier ??? │ wire │ Rectifier
│ === ===== === Smoothing │ ??? links ??? Smoothing
E >───┤ 4n7│ (((() 4n7│ & HF E >──┤ │ │ & HF
2n7=== │ │ │ │ Switching ??? │ │ Switching
N >───┴────┴──┘ └────┴─> device N >──┴────┴──----───┴──> device
balance Mains QRM 80dB worse!
As the ferrite & mains rated caps are relatively expensive, quite a saving
is made. But as the main smoothing capacitor ages & dries out, its' ability
to suppress HF on the HT rail is greatly reduced & QRM levels reaching the
mains can only get worse! Some SMPSU carry on working fine with no mains
DC smoothing C !!
If your track down a new example of this practice, your local radio
administration/RFI enforcement body should be informed, & if they are on
the ball (some could not care these days) a prosecution could result.
2/ Applies the last filtering principle to the whole noise source..
Filter ┌────────┐ little
All leads │ NOISE │ \│/ RF
─┬─()))─┤ SOURCE ├─()))─┬┴──
=== └────┬───┘ ===
You can see this approach used on all PCs nowadays in an attempt to meet not
very stringent the RFI laws. Ferrites are used on all leads close to any
noise source box to reduce (but not eliminate!) any currents flowing into
the leads & reduce them acting as aerials.
Full screening is often not used due to dangerous voltages screen needs to
be see etc. But it may also be left off due to cost, & design / aesthetics /
sales considerations. e.g. a metal box around an RF economy lamp. :-)
For TV -> HF QRM, consider the UHF aerial system as a HF aerial, put a HF
braid breaker in the aerial & ferrites on the mains lead etc. as these
filters work both for Tx & Rx QRM.
If QRM source items in 1 or 2 are not yours, then modification can be risky
I have heard of hams with interference cases, who are prepared to buy a
neighbour a brand new TV etc. to solve the problem. But note well, modern
items can be a lot worse than some older designs! e.g. low voltage 12V
lighting now uses 80kHz 150W square wave power inverters with radiating
unscreened leads, rather than a straight forward 50Hz transformer!
3/ Reducing the QRM seen by the Rx can be simple or hard to achieve depending
on the location etc.
a) If it is yours, turn the QRM source off. (TV / PC / RF lamp)
b) Prove the path of the QRM to the Rx, e.g. not 12V pick up/mains pick up.
Put the Rx on battery, or via wound up extension lead to test. If not down
the aerial, then filter mains/other leads.
For UHF/VHF QRM try braid cover leads of 1/4 wave long bonded to Rx earth.
└─────════Á Rx Ă════───DC
c) Increase the distance from Rx aerial to the QRM source. Move TV/ Aerials.
True RF fields die away @ the square law e.g. 6dB per doubling distance.
But near fields with transformer/stray capacitance coupling like TVs
timebase etc. obey a cube law. e.g. 9dB per doubling of distance.
d) If QRM is steady & only from one source, it can be nulled out with a 2nd
aerial & phase/amplitude system.
Main \│/ ┌─────┐
Ant └───────────┤Phase│ ┌────┐
Noise \│/ │ & ├─>┤ Rx │
Ant └───┤Level│ └────┘
4/ Operate when the QRM is not there. Operate on bands/frequencies unaffected.
Or filter it out after/in the Rx.
Impulse Noise gates or blankers were the 1st attempt to do this, they remove
car ignition & spiky mains noise quite well on SSB mode.
┌──────┐ ┌──────┐ ┌──────┐ ┌─────┐
SSB +> │ Mixer├>─┤ WIDE ├─┬────>──┤ GATE ├─>─┤ SSB ├──> No
Spikes └──────┘ │ IF │ │ _ └──┬───┘ │ IF │ Spikes
└──────┘ │ │\ /| === └─────┘
The noise amp AGC is fast enough (300uS) to follow a single SSB voice
envelope, maintaining a fairly steady RF signal. When a wideband spike
occurs the rise time is too fast & it produces an AGC spike that is used
to turn off the IF gate to the narrow SSB filter. The result is very narrow
holes (e.g. 1mS) in the audio rather than loud clicks, that could also have
turned the Rx AGC down as well. Drawbacks are if there are too many stations
in the passband or big signals (e.g. contests) the noise amp miss operates
the gate making signals break up or sound very wide.
Try a modern noise DSP processing in or after the Rx, as much of some types
of QRM can be 99% removed without too much loss of intelligence in the Dx
signal underneath. One external "add on DSP LS system" I tested with a
steady carrier noise was reduced by 65dB automatically at the max noise
reduction setting, or 35dB for more broadband noises.
Signal > \│/ ┌────┐ │ D.S.P. │ 99% of
+ QRM > └──┤ Rx ├──┤NOISE LS│ Signal >
└────┘ │ SYSTEM │
It works by digitally analysing the AF spectrum into 100s of narrow
frequency bands, with an "FM noise squelch" system on each one, deciding if
the signal in the band is significantly above the noise to let through. If
the level on any band has not changed after 2 seconds ignore that band.
The drawback is that a voice with high levels of noise will sound like a
nasty cellphone call at high noise reduction levels, as the voice pitch to
allow through may be the wrong one!
When living with QRM, remember that our hobby is just a good a pastime a any
other. So you don't have to put up with it really. If you can, get help to get
the problem sorted. Try an intermediary, local radio club, radio authority,
other radio users affected? (e.g. police, ambulance, aircraft, broadcast bands
etc). This can all help if you can find the right person to take up the case.
Remember too loose plugs & wires/screws can be a QRM cause & fire hazard too.
However Andy G0FTD reminded me that this hobby is protected by:-
DIRECTIVE 2004/108/EC OF THE EUROPEAN PARLIAMENT & OF THE COUNCIL
of 15 December 2004 on the approximation of the laws of the Member States
relating to electromagnetic compatibility & repealing Directive 89/336/EEC.
Where Amateur radio is _specifically_ mentioned/protected.
See my bul "Stopping HF Tx/Rx SMPSU QRM", "Mains Plugtop Filter Adaptor" &
"Clip on QRM Probe", "Smart Meter Refusal".
Why don't U send an interest bul?
73 de John G8MNY @ GB7CIP
Read previous mail | Read next mail