[hpsdr] Status Thor?

[Stijn]

Hello Phil, group,

Let me first express my appreciation for all the time and effort that 
you and of course many others have invested in this wonderful project.
It is a pitty that the project Thor appears to be at a dead end, 
unfortunately I don't have the time or all the skills involved to 
continue this project.
But I still have some questions about the included support for Thor in 
Penelope and PSDR as it is for today, am I correct that Penelope already
provides an envelope signal by using one of the PWM outputs? And is it 
already possible to get a clean phase modulated output from the Penelope
RF port? Not that I am going to play with it in the short future, but 
maybe if I find some time I will start some experiments on low power.
I unfortunately have very basic programming skills and absolutely no 
experience with vhdl and fpga's, so I will be depending on what is already
available.

 From what I find on the net I think that a PWM modulator operating at 
500kHz + is (easy) possible today, this leaves lots of bandwidth below 
the PWM
low-pass filter cut-off. See the IRS2092 as an example. Almost the 
complete PWM modulator in a single ic.
But as I understand it, it is also possible to get the PWM pulses 
directly from Penelope or some of the other cards on the bus. This saves 
a conversion
step and eliminates the distortion that this can create. In that case 
you only need a fast FET driver. It would also be nice to have the 
possibility to time shift
both signals from each other to compensate for filter, cable and other 
delays in the total system.
But I personally think that the modulator is the easy part, making a 
class E pa that covers multiple bands (preferably all including 6M) is a 
completely
different story.

One of the main benefits of HPSDR over an analog transceiver is that we 
don't have to split an analog SSB signal in two different components.

I surely hope that someone stands up to bring this project to a happy end.

Sincerely,

Stijn Nestra PE1RKS

Op 07-08-11 10:05, Phil Harman schreef:
> Hi Stijn,
>
> I've not heard from the last project leader - Steve,  G1YLB - for 
> quite a while so assume the project has stalled for the time being.
>
> There is support for a Class E amplifier in Penny(Lane) and also the 
> KD5TFD version of PowerSDR.
>
> My original low power tests on 160m were very encouraging - I'd really 
> like to see this project progress, just don't have the spare time 
> myself at the moment.
>
> It also looks like the  SGC 'Mini-Lini (TM)'  Class E PA has not made 
> it into a commercial product:
>
> < http://www.sgcworld.com/500watt.html >
>
> perhaps we could acquire or gain access to the IP and run with the 
> project ourselves?  Since, using Penny, we can independently adjust 
> the phase relationship between the envelope and RF drive signals, on a 
> band-by-band basis, we should be able to overcome most of the problems 
> that beset earlier designs.
>
> I gather one of the problems with the SGC design was getting the PWM 
> envelope modulator to drop to 0v during the trough of the modulation.  
> To get round this they ran the PA in Class A for small input levels - 
> an innovative solution.  The other problem is that the bandwidth of 
> the envelope is much higher than the bandwidth of the voice 
> modulation. If you think of the envelope of a two-tone test signal 
> then when the envelope goes to zero there is an almost instantaneous 
> change in direction of the envelope voltage - this needs lots of 
> bandwidth to keep the transition clean.  With higher frequency SMPSU 
> chips now available it may make the design of the envelope LPF some 
> what simpler -  not something I've recent experience with.
>
> Anyone have the time and skills to run with this project?
>
> 73 Phil...VK6APH

 1313009973.0