to: Matt NQ6N
Maybe a better test to check PA functionality which includes bias and
that bothpush-pull devices are serviceable (rather than what amounts
to a "design verification test" measuring noise at 90 dB down):
Do a two-tone test. Checks PA linearity, probably a better 'figure of merit'
for a quick "field check" vs noise measurement.
Each tone, about 25 Watts "carrier". Freeware "Spectrum Lab" can gen the
two audio tones into the rig.
Both tones *combined* will be 100W PEP output. Avg reading wattmeter
will show a lot less.
A Rigol should easily show 3rd and 5th order tones down >30 dBc.
73, de AA5CT Jim
.
.
On Monday, February 24, 2020, 6:40:12 PM GMT-6, David Eckhardt
<davearea51a@gmail.com> wrote:
The directional coupler you have is a good start coupled to the spectrum
analyzer. However, for this test, the spectrum analyzer must have a low
noise floor. I have the Rigol DSA 815 and several other HP spectrum
analyzers. The Rigol is highly portable which none of the older HP RF
instruments were. The Rigol is a relatively early unit before they offered
an option with a low-noise LO. Mine hits the noise pedestal at -50 dBc -
pretty bad. All the others, HP, are better. For your radio, you can check
the QST reviews on transmitted noise and LO noise. I'll mention a really
bad one for comparison. The Yaesu FTDX3000 is the poster child for a dirty
transmitter as are all of the recent Yaesu transceivers. Compare that to
the Elecraft K3 in the QST reviews and you will see the vast differences.
Your radio is a Flex and I'm sure it is clean. Flex doesn't put out
anything even approaching the dirt from the Yaesu transceivers. Nor does
Icom or Kenwood or Elecraft. Your noise at +/- 1 to 2 kHz from the carrier
should be at least -100 dBc. This is difficult to measure due to the
strong close-by carrier. Again, I seriously doubt the Flex is the problem.
If your spectrum analyzer is not clean enough to detect the noise, try a
second good receiver - none of these battery powered units from China. Be
sure to attenuate the fundamental sufficiently so no damage is done to the
second receiver (or the spectrum analyzer). You will require at least 100
dB of dynamic range below the carrier to make any meaningful measurements.
It will require a quiet LO in the measurement instrument. The K3 is
outstanding in this respect. The HP 3585 spectrum analyzer ('DC" to 40
MHz) is exceptional in this respect.
The best I could suggest is to set the level into the spectrum analyzer
such that you're close to the permitted max power input to the instrument
(to be safe, back off 6 or 10 dB from that max level). Then measure the
noise 1, 2, 3, 4, and 5 kHz from that carrier. HOWEVER, this assumes the
LO in the spectrum analyzer is cleaner than the potential noise level from
the transmitter (at or more than -100 dBc at 1 kHz removed from the
carrier). Also *be aware* (!!!!) that many modern transceivers deliver a
short pulse near max power on initially keying output that lasts for
several milliseconds. This is enough to damage receivers and spectrum
analyzers. Only connect your measurement instrument AFTER you have keyed
the transmitter (in CW or RTTY mode). NEVER key the transmitter, even
through the directional coupler and attenuators, when connected to the
instruments.
Dave - WØLEV
On Mon, Feb 24, 2020 at 8:32 PM Matt NQ6N <matt@nq6n.com> wrote:
> Dave,
>
> One quick question:
>
> *Any modern solid state transmitter will produce broadband noise. The
> broadband noise is generally worse at less than full output power. I'm
> guessings that's where your BB noise is originating.*
>
> Can you recommend a test procedure I could use to determine how much of
> this broadband noise is originating in my transmitter (so that I can
> determine whether it is within spec)?
>
> 73,
> Matt NQ6N
>
> On Mon, Feb 24, 2020 at 2:00 PM David Eckhardt <davearea51a@gmail.com>
> wrote:
>
>> Just a few thoughts, here:
>>
>> Any modern solid state transmitter will produce broadband noise. The
>> broadband noise is generally worse at less than full output power. I'm
>> guessings that's where your BB noise is originating. It could also be made
>> worse by the 'thing' that is producing the multiplication and mixing of
>> your 7.058 kHz signal.
>>
>> Through the directional coupler, you comment the second harmonic of your
>> 7058 frequency is extremely low. That's good and evidence that the filters
>> are working as designed. Sample the whole enchilada from the air by
>> hanging a shorted o'scope probe ('ground' lead attached to the probe
>> center) on the input of the spectrum analyzer. This forms a
>> quick-and-dirty current probe with the advantage that there will be no
>> 60-Hz and harmonics present, unlike a clip lead attached to the input of
>> the spectrum analyzer. Key your transmitter and see if the resulting
>> second harmonic and the mixing product are any stronger than when connected
>> through the directional coupler. The coupler shows you the output of the
>> transmitter - clean. The quick-and-dirty current probe will display what
>> is 'in the air' due to anything that is locally producing the
>> multiplication and mixing.
>>
>> I'll also bet that the source of the mixing is also producing the 2X your
>> 7058 kHz frequency. It, by itself, could be emitting the 'unpleasant'
>> signature you found due to one of your wall warts.
>>
>> Connect a small wire (very short as a function of the wavelength at 14
>> MHz) to a battery powered radio and walk around the property with the radio
>> tuned to 14.786 MHz. Your transmitter must be keyed for this to work.
>> Maybe, just maybe, this might reveal the source of the multiplication and
>> mixing.
>>
>> I had a problem a bit like this long....long ago (a half-century
>> ago!!)). We had a "budget" AM BC station down the hill and about 0.75
>> miles line-of-site distant from my location. When I ran CW at 100 watts on
>> 80, 40, and 20 meters (no amp at the time) to my wire antennas, all kinds
>> of mixing products and noise reared their ugly heads. Problem turned out
>> to be neglected and improperly dressed and tightened guy wires at the AM
>> transmitter location (FCC got involved).
>>
>> Dave - WØLEV
>>
>> On Mon, Feb 24, 2020 at 7:26 PM Matt NQ6N <matt@nq6n.com> wrote:
>>
>>> Hello --
>>>
>>> I'm not sure how much background to offer about my shack, so I'll skip to
>>> the specific question I have and then below that I will describe in a bit
>>> more detail the steps I've been using to figure out the source of RFI.
>>>
>>> When I transmit at 7058 KHz at low power (testing today transmitting 6W)
>>> there is a very strong mixing product of AM 670 that appears at 14.786
>>> MHz. It seems to be 7058 * 2 + 670 = 14786.
>>>
>>> There are MANY such mixing products all over the bands when I am
>>> transmitting at 7058, including a loud mixing product of Radio Martí at
>>> 296
>>> KHz.
>>>
>>> Question 1: Is this normal, something that always happens and likely has
>>> nothing to do with my interstation interference?
>>>
>>> Question 2: If it is not normal, then would the source of radiation of
>>> the
>>> mixing products be a likely culprit for the interstation interference?
>>>
>>> The mixing product at 14.786 is audible on a portable short wave receiver
>>> as I walk around the property.
>>>
>>> Next I'll briefly describe the station configuration:
>>>
>>> The rig is a Flex 6600. At first I thought my interstation issues had to
>>> do
>>> with the typical scenario such as the second harmonic on 40m overwhelming
>>> the receiver on 20m. However, in my setup the interference seems to be
>>> much more severe.
>>>
>>> Transmitting 6 watts on 40m, most of 20m sees a noise floor increase of
>>> over 10 dB. This is transmitting on the Flex 6600 which has additional
>>> BPF
>>> filtering on contest bands AND a VA6AM high power filter.
>>>
>>> When I connect a spectrum analyzer to the output of the VA6AM filter via
>>> a
>>> directional coupler, the second harmonic is incredibly weak to the point
>>> where visualizing it is difficult on my Siglent 3021x spectrum analyzer.
>>> It is hard for me to believe that this tiny signal is strong enough to
>>> cause a 10 dB noise floor increase on 20m.
>>>
>>> The signal path includes the Flex 6600, a 403A 8x2 switch, and wire
>>> antennas. Interconnections betweenn the rig, VA6AM BPFs, and switch box
>>> all use RG-400 jumpers, and all antenna feed lines are LMR-400.
>>>
>>> The antennas are suspended in the trees above the house. The minimum
>>> spacing between the 40m Vee and 20m dipole is approximately 4 feet.
>>>
>>> Does it make sense that transmitting *6W* with all this filtering there
>>> would be enough signal getting into the 20m antenna to cause a 10 dB
>>> noise
>>> floor increase?
>>>
>>> I have tested using a different transmitter AND a different receiver and
>>> the 20m noise floor increase while transmitting QRP appears to be present
>>> in both cases.
>>>
>>> In my last RFI search I discovered a few wall warts whose power cords
>>> seemed to be radiating some kind of mixing products, and so I added
>>> ferrite
>>> or disconnected each of the ones that did this. Unplugging them
>>> completely
>>> does not reduce the harmonic.
>>>
>>> Question 3: What should I be RDFing to find the source?
>>>
>>> My hypothesis is that the mixing products described above are NOT normal
>>> and thus I have been under the impression that RDFing them would help
>>> locate whatever is doing the mixing. Is this correct?
>>>
>>> One other data point:
>>>
>>> I found one wall wart and mains circuit that appeared to be adding "junk"
>>> onto the second harmonic of the transmitted signal. The second harmonic
>>> does not sound like a pure sine wave. Instead it is wide and has a very
>>> unpleasant sound. I found that removing that one wall wart improved the
>>> purity of the second harmonic.
>>>
>>> Question 4: Would it make sense to try to RDF the source of the "junk"
>>> sound on the second harmonic?
>>>
>>> I have tried turning off each of the breakers in the breaker panel
>>> (except
>>> the one powering the TX), and none have any impact on the tone of the
>>> second harmonic OR the BCB station mixing. I will repeat this test today
>>> and also try transmitting on the KX2 to verify that the mixing products
>>> are
>>> still there.
>>>
>>> I'd also welcome ideas for things to test/try that might shed some light
>>> on
>>> what is going on.
>>>
>>> 73 and thanks much for any advice. I have temporarily removed my amp
>>> from
>>> the system and have installed the 403A switch and RG400 in an attempt to
>>> make it easier to deduce useful things from various tests.
>>> _______________________________________________
>>> RFI mailing list
>>> RFI@contesting.com
>>> http://lists.contesting.com/mailman/listinfo/rfi
>>>
>>
>>
>> --
>>
>> *Dave - WØLEV*
>> *Just Let Darwin Work*
>> *Just Think*
>>
>
--
*Dave - WØLEV*
*Just Let Darwin Work*
*Just Think*
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|