[ARDF] Fun and Frustration

[Gerald Boyd]

Hi Bruce.
 

--I've been following this discussion with some interest.

--My own feeling, and I'm happy to be contradicted, is that power
consumption, 
--and to a lesser degree cost, often overlooked till it's too late, is the 
--killer for battery powered s/w defined radio solutions at this stage.

--We've looked at FPGA (much lower frequency apps than this would be)
solutions 
--over microprocessor solutions for various products at work, and it 
--always comes back to power consumption, esp. if you are running off
batteries. 
--Phillips had a series called "cool running" FPGAs that I think they sold
off to one of the big boys
-- (Altera or Xilinx), but even then you are looking at an order of
magnitude increase 
--over a discrete RF or microcontroller solution.

Actel now has a part that is even lower power then the old cool runner,
However its speed may be to slow for some applications( 30 mhz top).

in the SDR transmitter we are shutting down power to the Xilinx FPGA when
its off cycle to save power. 
The device we are using does have a suspend mode that may be able to reduce
power even more when off cycle ( keep the device powered but in a static
state). 

just have not have enough time to try experimenting with it so went with the
simple kill the power method when off cycle. The desire is to run the entire
thing on a set of AA batteries.

--With a receiver, as distinct from a transmitter, you'd also need an ADC
capable of 300MHz operation 
--(for 2*sampling at 2m), and that takes serious current (and cost). This is
outside the FPGA itself.

Got around that issue of having to use a high speed D/A by using some
special circuits  and one  DCM configured as a 130 MHz LO  and a DBM mixer.
All RF generation for VHF is done between 14 and 18 Mhz ( 144-148) and up
convert to VHF using the minicircuits DBM mixer.  For 80 meters RF
generation is done at direct frequency.

for receiver applications I have some ideas on how to get around having to
use an A/D converter and D/A or codac for audio output. 
Need to experiment with the ideas in hardware (eval board or computer
simulations) to validate.


--Also, don't forget to add the support chips. A RAM based FPGA (probably
the case at the speeds we are talking),
-- you have to add an SPI or IIC serial flash eprom. Then you still need the
microcontroller 

Only requires one low cost chip $ 5 to boot an sram based Spartan. However
the Actel pro asic 3e devices are flash based and don't require any external
support chips other then a clock source. The FPGA is doing a majorityy of
the work including LO generation.

--to do the user interface (pretty much any serious ARDF receiver would have
one now anyway). 
--Yes, you can get VHDL defined micros or special micros embedded in the
FPGA (eg ARM or PowerPC), 
--but again, power consumption is worse than the plain silicon version.

It maybe be possible to over come the power issue with some newer devices. 
I see articles in industry hinting that newer FPGA are taking over processor
based applications. Have been doing some functions in FPGA's that 10 years
ago I would have used a processor.

--Perhaps a solution where you sample at audio frequencies, or zero IF (with
I and Q) could be a useful 
--1/2way SDR solution, but then you still need the associated RF components
(front end, rf filter, IQ 
--linear mixer and PLL VCO) that you'd have had with the 'discrete'
solution, so unless you have a driving 
--need for it (eg. software programmable IF filter bandwidths or special DSP
tone detection/FFT etc), 
--then you almost certainly have a more costly and complicated solution, and
probably having to lug 
--around a bigger battery/LiIon cells as well.

that may be where the new actel low power igloo device can rule?

--Dale's idea of a separate RF and CPU board is perhaps the most currently
helpful idea, but bear in --mind that is exactly what the "Ultra Sniffer"
kit David VK3XJH and I designed years ago was, and it --hardly took the
world by storm. Frankly, there is little incentive for anyone to create such
a thing --for you. There simply aren't the sales in ARDF to make it
worthwhile. It is not really practical to --do this these days using
pin-through components, so we're certainly looking at surface mount. It's
--very hard to do this for a kit, and almost as much work to make the kits
up as to just pick and place the board anyway !  I know...I've been there.
--You are wecome to the circuit and layouts for the old Ultra sniffer RF
board if you want. 

Yes would be interesting to see how it was done.

--It could still form the basis for a solid performing ARDF receiver, and it
can double as a F
--oxOr transmitter too ! The micro board however, by today's standards, is
crap.

--Lastly, I'd be very surprised if an iPhone can directly do a SDR without
an external downconversion. 
--See above comment about needing lots of stuff to do even that.

--That's my 2c. As I said, happy to be contradicted as there could be now
developments of 
--which I'm unaware, as I don't work directly in RF anymore, but for now, I
concur with Dale.

always open to ideas. 

Can an IPOD nano be used as a user interface display receiver control
device? Much lower cost then an IPHONE. you can get them in the $170 price
range ( or lower?).

anyone have interface or command set documents?

I run with one every day attached to may arm. Get very long battery run time
and good audio. 

Lets kick around some ideas before building something. Anyone interested in
an ARTS group on this subject to kick around ideas?

Jerry