Message-ID: <53E4C41D.4040504@gmx.net>
Date: Fri, 08 Aug 2014 14:35:41 +0200
From: Tobias DG3LV <DG3LV@gmx.net>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.6.0
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <F2308B40167B42B486EAD4657D9D78F6@AGB> <53E3983D.9050606@gmx.net> <D300C456F3F2421E913AD4C9D41EE57C@AGB>
In-Reply-To: <D300C456F3F2421E913AD4C9D41EE57C@AGB>
Subject: Re: LF: Receivers for LF and MF    136 KHz  @ 477KHz   ???
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

Hi Graham !

The foremost problem of the Barlow-Wadley receiver is the generation of 
the LO by a comb harmonic oszillator (in German : 
"Lattenzaungenerator"). Receiving an unknown signal you could never be 
shure, from which of the 1MHz wide Bands the signal actually originated. 
The installed IF-bandpassfilters were much too wide and the resulted 
purity of the LO was poor. At best it was just 1 MHz up or 1MHz down 
from the selected MHz band, but large signals got through even farer 
away. And the mixer did his job very well, it mixed everything with 
everything. In the 1940s, when Mr. Wadley developed his Patent, this was 
certainly not an issue. But even in the 1970s, when I listened first to 
such a radio, the bands were much too crowded already.
http://en.wikipedia.org/wiki/Wadley_Loop
http://televideo.ws/wadley.html

Using a direct conversion receiver/transmitter (at any samplerate) you 
must use a nyquist-filter (=anti-aliasing filter). Without such you will 
receive the noise of all the "harmonics" and thus your noise-figure is 
degrading a lot.

73 de dg3lv Tobias

Am 07.08.2014 22:35, schrieb Graham:
> Yes Tobias,
>
> All  is not as it  seems , but as you say , adding  filtering , reduces
> the problems  , and  with  direct conversion  , the  noise  floor  is
> the lowest possible , stability  is  at maximum , having only  1
> oscillator , In have  good  decode  results  on 477 using  the  £5
> dongle  , behind  the  TX atu and inv L ae
>
> For the  £150  sdr , the  description  lists  80 msps , which  , I
> assume moves the  image  problems  out  side the  HF spectrum ? at  12
> bits  , that starts to  provide a  reasonable  , post processed  dynamic
> range ?
>
> 12- bit 80 MSPS A/D conversion
>
> I would  not  link the  barlow-wadley , too  closely  to  lack  dynamic
> range , may  be a lack of  engineering  integrity , the  RA17  with
> pentode rf  stage  , then  later  cascode  , with  beam deflection
> mixer , was reasonable , though  there  was a  pre selector  for  use
> at  close  tx/rx sites ,  the  ra1771/1772  was  the  first  to  better
> the  ra17 , but again that also tends to  question the  models before
> ..   the  ra1772/1  is fitted  with  rf-pre selector , for those
> 'unexpected' situations ,  the ra6790gm , with  no  pre-amp  is the
> closest  I have  seen  to  a  bullet proof  front  end , that  runs ,
> where  the  ra1778 needs  the  pre-selector
>
> But in terms  of  noise  etc , one of these  'reasonable  sdr's'  may be
> better .. I don't think I would  recommend any one  go  down the  racal
> path  these  days !
>
> 73-G,
>
> --------------------------------------------------
> From: "Tobias DG3LV" <DG3LV@gmx.net>
> Sent: Thursday, August 07, 2014 4:16 PM
> To: <rsgb_lf_group@blacksheep.org>
> Subject: Re: LF: Receivers for LF and MF    136 KHz  @ 477KHz   ???
>
>> Hi Graham !
>>
>> This type of device has a dynamic range of 48dB at maximum (8 Bit), so
>> they need help from a narrow-band frontend plus input-attenuators to
>> reduce the load to the A/D-converter. The conversion speed is variable
>> from 1 to 3 Megasamples/second, leading to a Nyqist-frequency of 0.5
>> to 1.5 MHz.
>>
>> This filter should have a deep attenuation above the Nyqist-frequency
>> to avoid unwanted reception at harmonics/aliases of the sampling
>> frequency (= "undersampling").
>>
>> The advertized usage of 100kHz to 30MHz (in direct-sampling method) is
>> based on this normally unwanted harmonics/aliases, i.e. the receiver
>> uses a method that has become "famous" with the ancient
>> "Barlow-Wadley" all-band receivers. (at least equivalent to). Their
>> (large-signal-)problems of the past are reborn at the direct-sampling
>> method of these DVB-T sticks. Without narrow-band selective frontends
>> this is just a "proof of concept" and not a usable receiver.
>>
>> For the use at 136 kHz and 475 kHz a steep lowpass-filter (7 to 9 pole
>> Tscheby with toroids) at (e.g.) 500kHz would be mandatory. An actual
>> bandpass may not be necessary. Equipped with such filters the lack of
>> resolution (8 Bit) will become more acceptable. Using a pre-amplifier
>> without using filters will do no good.
>>
>> When home-brewed, such filters will not cost much, but it takes time,
>> measurement-tools and effort to build and tune them.
>>
>> 73 de dg3lv Tobias
>>
>> Am 07.08.2014 14:13, schrieb Graham:
>>> Receivers for LF and MF136 KHz@ 477 KHz
>>>
>>> A question,
>>>
>>> Startingat the£5dongleriggedfordirectsample , as
>>> acheapeffectivestartingpoint
>>>
>>> A pre amp and  pre - selector [ band pass filter ] would  help  for
>>> lf/mf
>>>
>>> Whatwouldbein a scaleofincreasingperformance[ notcost !]
>>>
>>> be seen asreasonable in terms ofconfigurationand hardware ?
>>>
>>> Any particular equipments  stand out   as  good cost/performance
>>> choices ?
>>>
>>> Tnx
>>>
>>> Graham
>>>
>>> G0NBD
>>>
>>
>
>