[hpsdr] New RF Digitizer

[Nyall Davies]

Hello Ahti,
130 MHz compared with 100 gives another 1.13 dB of processing gain.
or compared with SDR-IQ at 67 MHz   2.87 dB
I should say that the 78.2 dB SN figure used for the LTC2207 is the
"headline" figure and is the best achievable. Depending on the mode used and
input frequency may not be as good as that.
The comparison between SDR-IQ and HPSDR shows how dependant we are on the
performance of the A/D to squeeze the odd dB.
I see Phil N8VB has given some other advantages of the HPSDR

73 Nyall G8IBR


----- Original Message ----- 
From: "Ahti Aintila" <oh2rz.sdr at gmail.com>
To: "Nyall Davies" <nyalldavies at ukonline.co.uk>
Cc: "Lyle Johnson" <kk7p at wavecable.com>; "High Performance Software Defined
Radio Discussion List" <hpsdr at hpsdr.org>
Sent: Tuesday, November 14, 2006 10:22 AM
Subject: Re: [hpsdr] New RF Digitizer


> Nyall,
>
> Thank you for your explanation. Obviously HPSDR plans to use the
> 16-bit LTC2208 with specified 130 MHz sampling. If there are no faster
> replacements for the AD6620, so possibly some FPGA should be used
> instead. How much that would help?
>
> 73, Ahti OH2RZ
>
> On 14/11/06, Nyall Davies <nyalldavies at ukonline.co.uk> wrote:
>>
>> COMPARISON OF SDR-IQ with HPSDR
>> The signal noise ratio of the fast AD converters does not allow the full
>> use
>> of all the bits. The extra bits in the AD should mean that the
>> quantization
>> loss in the AD can be ignored - it should provide its own dither noise.
>>
>> The picture of the SDR-IQ shows that it uses the AD6620 which comprises
>> the
>> digital mixer and set of CIC filters. The processing gain the AD6620 is
>> such
>> that 23 bits are required at its output
>> I cannot see what AD it uses but we have to assume that it is the AD6645.
>> This part has a S/N ratio around 74dB
>> It states that the sampling is done at 67MHz (the max spec of the AD6620)
>> If we are using the LTC2207 sampling at 100MHz - it has a quoted S/N of
>> 78.2
>> dB so we are better due to the AD SN by 4.2dB
>> If we sample at 100 MHz we gain 10 log(100/67) dB of processing gain
>> =1.739
>> dB
>>
>> So we can expect a performance 4.2 +1.7 = 5.9dB better than the SDR-IQ
>>
>> It has been suggested to me that we Linear tech can supply a device to
>> sample at higher rates say 150MHz so we may, depending on the S/N of this
>> device, be able to gain another 1.7dB
>>
>> Nyall G8IBR
>>
>> >> _______________________________________________
>> >> HPSDR Discussion List
>> >> To post msg: hpsdr at hpsdr.org
>> >> Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
>> >> HPSDR web page: http://hpsdr.org
>> >> Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/
>> >>
>>
>>
>>
>
>



 1163518881.0