[hpsdr] Pennywhistle q. about negative feedback

[Paul Fletcher]

OK, some good progress.

Firstly, my initial estimate (guess) of feedback resistor values was way 
out. The power gain is set approximately as 1 + R2 (or R12) with a one turn 
feedback loop. Ideally I would have used 33 ohms but didn't have any 
available so ended up with 47 instead (22 was too low).

For 250mW in I want 5W out, a power gain of 20 which would suggest a 
feedback resistor of 20 ohms. In fact this worked out near enough exactly, 
except for 12 and 10M where the gain rolled off a little. The great news was 
significantly improved IMD performance.

I swapped out my 22 ohm resistors (nearest I had to 20) to the next highest 
value I had which was 47 ohms. This is a little over the top as 250mW in was 
producing about 8 watts out on 10M which is actually a positive result, as 
it was twice the 4 watts I was getting with the 22 ohm resistors.

As for IMD - with the bias set as per recommendations, d3 was at least 40dB 
below the two tones which is -46dB referenced to PEP. On the best bands, 
(80, 60, 40) the IMD was 10dB lower than this: -56dB referenced to PEP. d5 
was about 10dB below d3 in all cases.

This is great news for me, but I'm now struggling with the AN762 - it's 
working to spec but I want it to be better!

Lyle, your suggestion of an overall feedback loop is interesting - I need to 
think about this and work out if the amplifiers are inverting or not and if 
they aren't how I deal with this without creating a massive positive 
feedback loop.

Onwards and upwards,
Paul M1PAF

----- Original Message ----- 
From: "Chris Smith" <chris at vspl.co.uk>
To: <hpsdr at lists.openhpsdr.org>
Sent: Monday, November 19, 2012 8:53 AM
Subject: Re: [hpsdr] Pennywhistle q. about negative feedback


> ***** High Performance Software Defined Radio Discussion List *****
>
> Paul
>
> Please keep me informed of any progress you make on this. I also only need 
> about 5W out from P'whistle to drive a 140W PA from CCI and would rather 
> use f/b than some sort of pad.
>
> Cheers & 73
>
> Chris G4NUX
>
> On 19 Nov 2012, at 07:01, David McQuate wrote:
>
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>> Hi Paul,
>>  You're the first I've heard of to apply negative feedback.  It sounds 
>> like a good idea.  It would be great to have a number of data points that 
>> show how IMD performance improves and the corresponding gain & power 
>> output.
>>
>> On 11/18/2012 2:34 PM, Paul Fletcher wrote:
>>> ***** High Performance Software Defined Radio Discussion List *****
>>>
>>> Hi all,
>>>
>>> I'm playing around with the negative feedback option on Pennywhistle to 
>>> try and improve the IMD numbers (I only need about 5 watts drive out 
>>> anyway) and wondered what others have done for the feedback winding on 
>>> T2? I'm currently using a single turn and R2 / R12 at 10k which gives a 
>>> noticeable improvement but still develops more power than I need. Any 
>>> other suggestions on configuration? I'll play around some more but it's 
>>> always nice to know what everyone else is up to.
>>>
>>> 73 Paul
>>> M1PAF
>>> _______________________________________________
>>> HPSDR Discussion List
>>> To post msg: hpsdr at openhpsdr.org
>>> Subscription help: 
>>> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>>> HPSDR web page: http://openhpsdr.org
>>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>>>
>>>
>>
>> -- 
>> Clear Stream Technologies
>>
>> _______________________________________________
>> HPSDR Discussion List
>> To post msg: hpsdr at openhpsdr.org
>> Subscription help: 
>> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>> HPSDR web page: http://openhpsdr.org
>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: 
> http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
> 


 1353346648.0