[hpsdr] [CASMIR] Design v0.3

[Alex]

I believe the answer to the first question is we need to initially cover the same frequency range as MERCURY - 0-30MHz - HF Bands. I think it's a mistake to try to also cover 6m with a unit. Look at the SDR-1000 - 6m is almost an afterthought.
Better for 6m to be dealt with by a VHF/UHF version.

Given that, the HF version would need to have specs at least as good as the following:

Transmitter Frequency Range:    160M-10M Bands 
Antenna impedance:      50 Ohms unbalanced
Frequency stability:     Less than ±0.05ppm (0 to +50°C) after warm up
Frequency resolution:     1Hz (minimum)
Spurious emissions:     More than 60dB (HF bands) 
Carrier suppression:     More than 63dB
Unwanted sideband suppression:     More than 80dB
Output Power: 1 watt


----- Original Message ----
From: Lyle Johnson <kk7p at wavecable.com>
To: Alex <harvilchuck at yahoo.com>
Cc: Christopher T. Day <CTDay at lbl.gov>; hpsdr at hpsdr.org
Sent: Tuesday, June 6, 2006 9:13:17 PM
Subject: Re: [hpsdr] [CASMIR] Design v0.3

> This leads to my query to everyone on LO and I/Q generation. I need some solid arguments not some "I used DDSes in the past and they suck.", but some good technical discussions. So ... here are three directions the design can go:
> 
> (1) Use the AD9958 and generate the LO and the DDS reference clock from the FPGA
> (2) Use a AD9959 and don't let the FPGA generate the LO, just the reference clock for the DDS
> (3) eliminate the DDS and let the FPGA do all the work 

When I was a kid, I remember seeing a comic strip (isn't that where many 
of us got our early education?) that showed a person shooting arrows 
into a fence.

After he shot the arrows, he walked to the fence and began painting 
targets around them so all of his arrows hit bull's eyes.

So, my short list of initial questions (to establish the targets first!) 
are:

1) what is the desired frequency coverage?

2) what is the desired frequency stability?

3) what is the desired phase stability?

4) what is the desired amplitude stability?

5) what is the desired power output range?

6) what is the expected load impedance?

7) what sort of mismatch to that impedance needs to be tolerated?

8) what sort of signal to noise ratio is wanted?

9) what amplitude level of spurs are acceptable?

10) what lock-up time is necessary?

11) will the transmitter be required to be tunable (e.g., adjust the 
center frequency of the output) while transmitting?

12) what information bandwidth is desired to be impressed (modulated) 
onto the transmitter "carrier"?

13) how fast must the transmitter respond to a "transmit on" command and 
a "transmit off" command (e.g., envelope rise and fall times)?

14) what is the budget limit for bill of materials at quantity one?

15) are semiconductor devices limited to packages with leads, or are 
ball grid array and chip scale packages acceptable?

There may be other essential questions, too.

We can then figure out suitable approaches for frequency synthesis, be 
they DDS, PLL or a combination.  And power amplification.  And how to 
create modulation waveforms. And...

Or, it may be that you just want to see what you can do with a given set 
of parts.  A sort of SDR equivalent to the QRP "What sort of radio can 
you make using only 2N2222 transistors for the active devices?"

And, as in most engineering endeavors, there is no correct answer.

Enjoy!

73,

Lyle KK7P







 1149692976.0