[Amps] Pre-Distortion Linearizer

[Roger (K8RI)]

On 5/5/2013 4:19 AM, peter chadwick wrote:
> Manufacturers have gone mad about IMD in receivers but can you use it? e.g. a +40dBm TOIP with a 10dB NF. In SSB, that's an rx noise floor of -130dBm and an SFDR of 113dB. To use that, you need phase noise of -147dBc/Hz averaged over the 3 kHz bandwidth at whatever offset you are worrying about.....even that leads to a 3dB degradation. So to properly use that TOIP, the phase noise should be 10dB lower or -157dBc/Hz.
>
> To quote Dominie Sampson "Prodigious!"
>
> Measurements suggest that 95 to 100dB of instantaneous receiver dynamic range i.e. phase noise or imd limited, is about all the average amateur needs....
>
> Now how far do you need to improve transmitter wideband noise to make a difference to other users? Just consider the wide band noise effects of a 250kW broadcast tx on say 7310 and how much trouble that will make.......
>
> Although the degradation from about -30 or -35dB IM3 wrt tone has been undesirable, I do wonder how much difference it really makes compared with the effects of 'all knobs clockwise' which we will never stop.
>
> For commercial ISB, the usual requirement has been -36dB wrt PEP, and that's based on acceptable crud in the wanted sideband. It's unfortunate that bipolar transistors on 12volts are inherently limited, but that's what the market appears to want at the moment.
>

Now is not the time to expect much from the semiconductor industry 
either.  World wide, I believe the entire industry in in the largest 
pull back ,or contraction from poly Silicon to finished devices in its 
history.

I mention this as the solar industry is the major user of Silicon and 
thus indirectly determines the price price and availability of devices 
for the entire semiconductor industry.

Having been associated with the industry almost since its inception, my 
own educated guess is, including this year, over the next 3 to 5 years 
we will lose between half and 3/4 of the poly producers.  After one of 
the earlier contractions I believe there were only 3 poly producers left 
operating.  Recently I believe there were around 60. However I think 
only 3 or 4 of those are major world players.

This is an industry noted for its "Feast or Famine" cycles, but jut two 
years ago some companies were going through multi Billion dollar 
expansions and now we are looking at losing maybe half or more of 
existing production.  How would you like to be the CFOs who have to 
explain "those expenditures that may never start."?  It's certainly 
cheaper to never start and mothball them rather than run the chemicals 
they use through the pipes.

China has been supporting its solar industry with subsidies that would 
likely make Solyndra look like pocket change.  They couldn't do it 
forever.  That created a market for poly from other countries as well as 
developing their own poly capacity.

This allowed them to sell solar panels cheaper than we could make them 
causing the failure of many panel makers.

To top this I believe (for lack of a current figure) that poly which was 
selling for more than several hundred US dollars a kilo has dropped to 
less than $60 and might go to $20 (my figure, not official and I've been 
ouy-of-the-loop for some time.)

The normal cycle would result in culling the producers because of 
prices, then the prices would again head for the stratosphere due to low 
capacity.  The industry has been doing this for over 50 years.

To point out just how sensitive and precariously balanced this industry 
is: Many years ago, we developed a method of sawing silicon wafers that 
produced 50% less waste. Waste at that time was running 50% so cutting 
it in half was a major change that Virtually over night (as long as it 
takes to set up the saws) we doubled Silicon production with virtually 
no increase in cost.  The industry nearly collapsed from over capacity.

Again this all has an effect on the price and availability of chips 
throughout the industry as well as development of new devices that we 
depend on.

73

Roger (KK8RI)


> 73
>
> Peter G3RZP
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>