[hpsdr] Pure Signal RF Sampler

[Richard Grubb]

Thanks Scott.

The resistive coupler I built is peculiar to the contents of my Junk 
Box. I used a series parallel arrangement of 9 half watt metal film 
resistors in three parallel - series groups and a 51 ohm terminator to 
provide -51 dB coupling and the necessary voltage and power rating for 1 
kW. It is flat to a dB or so from 160 - 6m. Actually the Pure Signal 2 
implementation is very non critical on sample levels. The software is 
completely happy with the linear enabled at 800 - 1000W. However turning 
the linear off and just having 5W from the ANAN 10, although the 
software flags low level, it still reduces the IMD by ~30 dB. Kudos to 
16 bits of dynamic range. A great piece of software ! Thank you HPSDR 
programmers !!

It might be worth including a note on the alternatives here in the Power 
Signal documentation.

73, Dick W0QM, G3FNL

------ Original Message ------
From: "Scott Traurig" <scott.traurig at gmail.com>
To: "Richard Grubb" <richard.n.grubb at comcast.net>
Cc: "hpsdr" <hpsdr at lists.openhpsdr.org>
Sent: 9/5/2018 11:40:40 AM
Subject: Re: [hpsdr] Pure Signal RF Sampler

>You are not missing anything, Dick. You can develop the signal however 
>you wish, as long as you provide a signal whose amplitude is within the 
>dynamic range of the receiver hardware and algorithm.
>
>People have gravitated towards the Xtronic coupler because it provides 
>ruler flat coupling response from 160M to 6M, which takes some of the 
>guesswork out of knowing if you are obtaining the proper feedback 
>signal levels, because the design is bulletproof in terms of power 
>handling, and because the insertion loss is quite small. Oh, and 
>because they don't have to build it and characterize it themselves ;-)
>
>There are certainly a few folks who are using their own coupler/sampler 
>designs, LP-100 sensors, and other, similar solutions. The only 
>solution I don't recommend is the Clean RF sampler. Bump that knob the 
>wrong way and it's possible you could fry your receiver, depending on 
>the power levels you are running and how much additional attenuation 
>you have on the sample port. Maximum safe input power level to any port 
>on Apache hardware is +13dBm based on the component data sheets.
>
>73,
>
>Scott/w-u-2-o
>
>On Wed, Sep 5, 2018 at 1:08 PM Richard Grubb 
><richard.n.grubb at comcast.net> wrote:
>>***** High Performance Software Defined Radio Discussion List *****
>>
>>I am wondering why the documentation on Pure Signal implies the use of 
>>a directional coupler as necessary to obtain the required sample of 
>>the transmitted RF. I can understand why that could be desirable when 
>>the feedback was not adaptive, or where the sampler is remote from the 
>>final amplifier.. However when the predistortion is being generated 
>>adaptively I do not understand why a simple resistive non directional 
>>coupler would not be perfectly satisfactory and much easier to 
>>implement.  The antenna system should be quite linear and the voltage 
>>sensed adjacent to the final amplifier output should surely always be 
>>a valid measurement of the transmitter transfer function. Of course a 
>>directional coupler is always useful to measure VSWR and is usually 
>>already present in modern amplifiers, although often not readily 
>>accessible for use with Pure Signal.
>>
>>I have implemented Pure Signal on an ANAN 10, Juma 1000 PA combination 
>>and used both the published directional coupler design and a simple 
>>resistive pick off at the PA output with equally excellent results.
>>
>>What am I missing?
>>
>>Dick, W0QM, G3FNL
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20180905/7792135a/attachment.html>