Received-SPF: none (google.com: owner-rsgb_lf_group@blacksheep.org does not designate permitted sender hosts) client-ip=195.171.43.25;
Date: Wed, 18 Jun 2014 17:58:42 +0000
Message-ID: <20140618175842.Horde.mF3mR0NUi0-gYNZbN367hg1@posti.anvianet.fi>
From: phl@netikka.fi
To: rsgb_lf_group@blacksheep.org
References: <890FEB25EF8D41A2A9843DC7D7972D3F@PCFIXE>
 <8D1588BE7B132A5-2024-D718@webmail-m237.sysops.aol.com>
 <0D65E2D5B33E454CBBBA70E173A51425@AGB> <53A0BBAC.9030803@gmail.com>
 <20140617223859.Horde.suUdzUDUx2ua3dsFDcvN8g1@posti.anvianet.fi>
 <DAB318C9906D458C918B736C436651EA@AGB>
 <20140618112356.Horde.w3hY7UPcNXQNRpET7dQ4tQ8@posti.anvianet.fi>
 <E4EF545AC128446BAFFCCB5E6D3CC233@AGB>
In-Reply-To: <E4EF545AC128446BAFFCCB5E6D3CC233@AGB>
User-Agent: Internet Messaging Program (IMP) H5 (6.1.7)
MIME-Version: 1.0
Content-Disposition: inline
Subject: Re: LF: MF CW now!
Content-Type: text/plain; charset=UTF-8; format=flowed; DelSp=Yes
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Hello Graham, LF

I'm using the +I and -I inputs that were connected to the R820 tuner  
chip before.

The 1:4 matching transformer is not my own idea, I just made my  
transformer a bit beefier than the design I borrowed from. It had a  
FT37-43 core that provided a minimum frequency of 1500 kHz at 50 ohms  
system impedance with the given number of turns.

RF transformer rule of thumb: XL at min frequency = 4 x system impedance.


Various "sources" on the web claim that the AD input impedance is  
highly undefined, or at least not specified. However the article I  
borrowed the 1:4 XFRM idea from claimed it had some merits without  
going deeper on the subject.


I can not find any "mirrors" below 14.4 MHz, but above that hell  
breaks loose... so I added a simple PI LPF at around 15 MHz with the  
main intention to make sure no strong VHF-signals make it to the  
AD-chip.

You can see at least 2.4 MHz chunks of spectrum although there are  
settings above that. The higher settings above 2.4 MHz don't produce  
good results with my setup, signals get choppy or I get temporary  
freezes. Maybe a USB-system limitation? If I go below 1 MHz strange  
things also start to happen and mirrors appear. Running the standard  
2.4 MHz seems to be the best option for me.


You can probably make a dual-purpose device from a stick that has the  
R820 tuner that connects to the AD using only +I and -I since the +Q  
and -Q inputs are then free to use as the direct sampling input to the  
AD. I wanted to try this but my eyes and hands don't cope too well  
with the tiny dimensions since +Q and -Q are not brought out to any  
pads on my stick. Therefore I just soldered away the coupling  
capacitors going to +I and -I and then used their pads for interfacing  
my transformer.


The article that spurred my modding interest was in the last swedish  
magazine QTC, but very similar mods are all around the web.


Hope I did not add confusion.

BR

Paul-Henrik


Quoting Graham <g8fzk@g8fzk.fsnet.co.uk>:

> Ok Paul,
>
> Its  round  the  Q  and  Q + not Q  , configuration , the  latest   
> drivers allow  that to  be  selected  , the  single  feed seems  to   
> work  fine , but  still  needs  some  filtering to prevent images ..
>
> I think,  possibly  a  pass  band limited  pre-amp  would  help  for  
>  MF/LF , the  gain drop's  past a couple of  MHz  and  the  200 ohms  
>  z  needs some matching , may  be the  mini-whip is providing this   
> aspect  ?
>
> There are image responses  linked to the  sample  rate , a  
> pre-selector as opposed  to  a  'tuner'  would  be  'ok' for  HF , I  
> think .
>
> I have another 'now'  low band  dongle  . used to  be  high  band ,  
> but after  a 7 MHz session , with it connected  to a 10 mtr  Ae ,  
> the  vhf front  failed ..  so  that's  a  candidate for  mf/lf   
> conversion , these have  two life's  !
>
> One of  Jim's   two active  device band limited  pre-amps  , with   
> 200 ohms o/p z would be  ideal  , or  use the  toroid as  the o/p   
> coupling  device ? that  may  be  the  most  elegant  configuration ?
>
> 73-G,
>
> --------------------------------------------------
> From: <phl@netikka.fi>
> Sent: Wednesday, June 18, 2014 12:23 PM
> To: <rsgb_lf_group@blacksheep.org>
> Subject: Re: LF: MF CW now!
>
>> Hello Graham, LF
>>
>> To be honest I have not tried other feeding methods for the AD than  
>> a trifilar 1:4 transformer. I used a beefier toroid (FT50-43) and  
>> more turns to get a XL of 200 ohms at 400 kHz for the 50 ohm  
>> winding. The concept of having the AD and antenna line galvanically  
>> separated and  the antenna input DC-shorted appealed to me since  
>> there is no other protection against static.
>>
>>
>> I will give capacitor-coupling a try at some point.
>>
>>
>> A simple LPF is mandatory at least if one plans to use wire  
>> antennas, otherwise there is a large risk of interference from VHF  
>> FM  broadcasting.
>>
>>
>> BR
>>
>> Paul-Henrik, OH1LSQ
>>
>>
>> Quoting Graham <g8fzk@g8fzk.fsnet.co.uk>:
>>
>>> Paul,
>>>
>>> Q , do  you find the  transformer  feed  better  than  , simply  
>>> connecting via a  capacitor  to  the  AD ?
>>>
>>> 73-G,
>>>
>>> --------------------------------------------------
>>> From: <phl@netikka.fi>
>>> Sent: Tuesday, June 17, 2014 11:38 PM
>>> To: <rsgb_lf_group@blacksheep.org>
>>> Subject: Re: LF: MF CW now!
>>>
>>>> Hello Stefan, LF
>>>>
>>>> Your CQ's are 539 here on a Mini-Whip directly into the  
>>>> demodulator"  -chip on a slightly hardware-hacked RTL-SDR TV   
>>>> USB-dongle :-).
>>>>
>>>> The "hack" is just bypassing the VHF/UHF tuner-chip and running  
>>>> straight into the +I and -I inputs of the RTL2832U via a 1:4 XFRMR.
>>>>
>>>>
>>>>
>>>> No TX here at the moment but receiving you so well on a hacked 10  
>>>>  euro USB-dongle and the old MiniWhip is just a tad more than I   
>>>> expected...
>>>>
>>>>
>>>> BR
>>>>
>>>> Paul-Henrik, OH1LSQ
>>>> Vaasa, KP03SD
>>>>
>>>>
>>>>
>>>> Quoting DK7FC <73dk7fc@googlemail.com>:
>>>>
>>>>> 472.2 kHz...
>>>>>
>>>>> Anyone for a QSO?
>>>>>
>>>>> 73, Stefan
>>>>
>>>>
>>>>
>>
>>
>>