Message-ID: <5B60A72B.90103@posteo.de>
Date: Tue, 31 Jul 2018 20:15:07 +0200
From: DK7FC <selberdenken@posteo.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <CA+GfORvAA10STf3zWSQ-dRFwk5BBS25Ht0x8cDZ3cYTa09dYbw@mail.gmail.com> <5B54CCCB.6060903@posteo.de> <5B59E687.5030805@posteo.de> <5B5C3ABF.3030809@posteo.de> <5B5C4A63.3040504@posteo.de> <5B5C5446.30105@posteo.de> <14080281532786049@iva8-6be7d23653d9.qloud-c.yandex.net> <5B5CD03F.40602@posteo.de> <14811811532865577@myt4-415a3339794b.qloud-c.yandex.net> <5B5DC8BC.9030601@posteo.de> <5B5E2D7D.6050908@posteo.de> <5B5F678B.7000506@posteo.de> <6dc15e3e-a325-4be2-55e3-44413e814f8a@web.de> <5B6050DA.60905@posteo.de>
In-Reply-To: <5B6050DA.60905@posteo.de>
Subject: Re: VLF: Earth antenna transmissions on a guide rail | calculations
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Hi all,

Now i had a closer look to the scope traces i photographed last weekend.

For the 900m long wire the phase is 42 deg at 8270 Hz. That would mean 
that the ground loop has an L of 3 mH if the resistive component is R = 
175 Ohm. So i would need 123 nF to series resonate the loop.
The speaker cable should arrive in 2 days so i can do a next test next 
weekend, this time with much lower losses. 1 A antenna current is the 
goal! Ah and i like to try 2.47 kHz and 1.67 kHz and 1.57 kHz and 0.97 
kHz!!!

I even modified our expensive Rohde & Schwarz scope. It's internal power 
supply has an output voltage of (guess!) 12.0 V. So i spent a series 
diode and now i can run it on batteries, just a normal '12 V' 7 Ah lead 
gel accu. It consumes 2.85 A during operation, quite much. With the 
scope i can accurately measure the phase and capture the screen. So this 
will be another improvement.

Well, 3 mH at 8270 Hz. If this is true, then the loop size can be 
estimated. Assuming a simple model of a normal single turn loop with a 
rectangular cross section area, where one side is known to be 900m, the 
depth of the current into the soil can be calculated. I found the 
formula from ON7YDs website 
http://www.strobbe.eu/on7yd/136ant/#TransmitLoop and build an Excel 
table where i can tune the parameters until they appear reasonable. The 
results show something like a loop size of 108000 m^2 and the current 
flows back in a depth of 120 m. I'mnot sure whether this is the right 
order of magnitude at all :-) The excel sheet also calculates the 
radiation resistance (210 uOhm) and efficiency of the antenna (-59 dB) 
but this appears to be quite optimistic...

73, Stefan

>>
>> Am 30.07.2018 um 21:31 schrieb DK7FC:
>>> Hi VLF and ULF friends,
>>>
>>> In the attachment you can find 3 spectrograms from transmission of 
>>> the last weekend. For the 8270 Hz and 2970 Hz transmission i took 
>>> the data of the 900m ground loop, 5170 Hz is from the 450m ground loop.
>>> The 8270 Hz signal was very strong even in 3.8 mHz ('DFCW-600'). 
>>> Some real QRSS-60 would have been possible!
>>> The spectrograms also show that there was no trace on the frequency 
>>> of interest before and after the transmission (no traces from the 
>>> right channel ( 1 PPS) ).
>>>
>>> A few minutes ago i built the switchable network for resonating the 
>>> antenna properly. And i actually ordered 1000 m of that 0.75 mm^2 
>>> loudspeaker cable. Thus, in the next experiment i think the Q of the 
>>> antenna will be much higher and it will be possible to tune it 
>>> accurately. Then, the tuning capacitance will tell us what the 
>>> ground loop inductance is! And then, we can calculate the depth 
>>> where the current is flowing backwards in the ground, asuming a 
>>> simple model in first steps.
>>> Next, we can see whether the inductance or depth stays constant at 
>>> lower frequencies (hopefully it will increase!) by resonating the 
>>> antenna at 2.97 kHz for example.
>>>
>>> BTW i measured the DC resistance of the wire i used last weekend, it 
>>> is 116 Ohm, so it agrees well with the calculated 120 Ohm...
>>> The new wire should have 21.4 Ohm only.
>>>
>>> The big advantage of this antenna is that no high voltage and large 
>>> coils are needed. Usually the voltage is the limiting factor for a 
>>> VLF or ULF TX antenna. But not here. Here it is the power of the PA. 
>>> The antenna could easily handle 2 kW, which means 5.16 A, no problem 
>>> for the wire. And it would be a 20 dB stronger signal than last 
>>> weekend...
>>>
>>> 73, Stefan
>>>
>>>
>>>
>>> Am 29.07.2018 23:11, schrieb DK7FC:
>>>> As a next result from today's experiment, here is a spectrum peak 
>>>> of the 2970.005 Hz transmission. I reached 20 dB SNR in 180 uHz! 
>>>> This far more than expected, in July anyway.
>>>> It is the first ULF transmission (by amateurs) from an earth 
>>>> electrode TX antenna, detected in the far field.
>>>> Spectrograms will be produced soon...
>>>>
>>>> 73, Stefan
>>>>
>>>>
>>>> Am 29.07.2018 16:01, schrieb DK7FC:
>>>>> Hello Roman, VLF,
>>>>>
>>>>> I transmitted pure carriers each time. It was just a first test.
>>>>>
>>>>> Today i've done a second experiment, this time using a 900m long 
>>>>> wire! So the wire should have about 125 Ohm DC resistance. I 
>>>>> measured 220 mA DC at 38.4 V, so the overall loss is 175 Ohm. Very 
>>>>> interesting; the ground loss resistance stays at 50 Ohm although 
>>>>> the distance between the electrodes has doubled!
>>>>> If i would use the 0.75 mm loudspeaker cable then i can reduce the 
>>>>> losses by 100 Ohm which is more than 50 %, so i will gain 3 dB at 
>>>>> the same output power! Then with a switch mode PA having nearly 
>>>>> 100% efficiency i may gain 2 dB more (mmy lonear mode PA was not 
>>>>> well matched today and quite warm). So maybe i can reach 1 A 
>>>>> antenna current with just 75 watts??!!!
>>>>> Today i tuned to 550 mA antenna current again, at 8270 Hz. With 
>>>>> the scope, i measured the phas, it was more inductive than 
>>>>> yesterday. I actually found a 1 uF MKP-10 cap in my car and 
>>>>> switched it in series to the antenna. This improved the phase 
>>>>> slightly. Then i also found a low pass filter for 137 kHz, the 
>>>>> pypical pi configuration, i.e. it acts as 4*22 nF in parallel here 
>>>>> on this frequency. Switching this in series leads to a low 
>>>>> current, so the C is to small. So for the next experiment i'm 
>>>>> preraring a switchable C network (47n, 100n, 220n, 470n, 1u, 2.2u, 
>>>>> 4.7u). It will be even more necessary when i lower the wire 
>>>>> resistance (higher Q).
>>>>> Well, today the QRN was much lower, especially for the 2970.005 Hz 
>>>>> transmission period. Yesterday this was totally buried in the 
>>>>> noise on my RX on the tree. But today! I transmitted another 90 
>>>>> minutes with much lower QRN background. I already have a clear 
>>>>> spectrum peak but i like to try to improve it a bit more before 
>>>>> presenting it here.
>>>>> Todays carrier transmissions:
>>>>> 8270.000 Hz : 08:12...09:33 UTC
>>>>> 2970.005 Hz : 09:42...11:15 UTC
>>>>>
>>>>> Sorry for the confusing email ;-)
>>>>>
>>>>> 73, Stefan
>>
>>
>>
>