[hpsdr] Unstable HPSDR?

[Christopher T. Day]

Has anyone further thoughts on this e-mail? If I'm reading it correctly,
it sounds easy for us to wind up with a Horton receiver that is off
frequency by a Hertz or two anywhere in the HF bands. Given the flack
that the SDR-1000 has received about its stability and that I doubt
Richard Hambly will let us do worse that 10**-10 at 10MHertz on
Gibraltar, I find this prospect disturbing. Is chasing spurs the wrong
problem?


	Chris - AE6VK


-----Original Message-----
From: Philip Covington [mailto:p.covington at gmail.com] 
Sent: Saturday, June 17, 2006 5:51 AM
To: High Performance Software Defined Radio Discussion List
Subject: [hpsdr] Fwd: Horton LO - further thoughts

***** High Performance Software Defined Radio Discussion List *****

<Posted on behalf of Phil VK6APH>

I've been thinking further about the VHF VCO that we may need for
Horton and have come to the conclusion that a packaged solution may
not be the best answer.  The reason for this is as follows.

Lets assume that the packed VCO covers 100-200MHz with a control
voltage range of 0-10v.  This means that for each volt change of the
control voltage the VCO moves

(200-100)/10 = 10MHz.

OK,  lets say we have 1mV of noise on the control voltage due to earth
loops, power supply noise etc. This voltage will FM the VCO by

10,000,000/1000 = 10kHz

Even if we can get the noise down to say 10uV then we still FM at
100Hz - not good for CW. In practice we will do better due to the
divide ratio, but on 10m this is only 4 so we still have 25Hz of  FM.

The way round this is to build our own low noise VCO using a nice high
Q inductor and 'pre-tune' it by switching capacitors across the
inductor for each band. Lets take an example, the widest VCO swing is
on 80m where the VCO goes from 112 to 128MHz, a 16MHz swing.

If we build our VCO to require a 10v control voltage swing to produce
this 16MHz swing then the same 1mV of noise will produce

16,000,000/1000*10  = 1.6kHz and our 10uV down to  16Hz.  On 80m we
divide by 32  so this drops to 0.5Hz.

On 10m we have a VCO swing of  8MHZ and with the same 10v control
voltage swing we get

8,000,000/1000*10 = 800Hz for 1mV noise and 8Hz at 10uV, we divide by
4 on 10m so this drops to 2Hz by the time we reach  I and Q.

We can improve these figures by using a higher control voltage swing
but either way this will need a low noise power supply and careful
attention to screening.

73's Phil...VK6APH
_______________________________________________
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/