dsp/fft. Help with simple design

Hi
I hope someone in this group can help me. I have a basic understanding
of electronics and uC, but my knowledge is - to say the least - a bit
rusty.
A friend of mine asked me to help build a circuit, so now I ask here:-)

Wanted: I need to produce a 100kHz sine wave with an amplitude in the
range [2 5] V. I'm not sure about the resolution, but I guess (!) it
should be 10 or 12 bits. This is used as input to a substance via two
electrodes, and I need to measure on this substance using two other
electrodes and do a fft on the measurements.
The current design uses a uC (atmel something) and a AD5933 which then
produces the result for use in the uC. But now we are looking for a
simpler way in which we hopefully can use just one chip. Needless to
say it should be small and low power!

Can anyone give a hint to what chip to use for this or other
guidelines? Any help is greatly appreciated.
Best,
John Doe

tilnews@gmail.com wrote:
> Hi
> I hope someone in this group can help me. I have a basic understanding
> of electronics and uC, but my knowledge is - to say the least - a bit
> rusty.
> A friend of mine asked me to help build a circuit, so now I ask here:-)
>
> Wanted: I need to produce a 100kHz sine wave with an amplitude in the
> range [2 5] V. I'm not sure about the resolution, but I guess (!) it
> should be 10 or 12 bits. This is used as input to a substance via two
> electrodes, and I need to measure on this substance using two other
> electrodes and do a fft on the measurements.
> The current design uses a uC (atmel something) and a AD5933 which then
> produces the result for use in the uC. But now we are looking for a
> simpler way in which we hopefully can use just one chip. Needless to
> say it should be small and low power!

You won't find any uC (internal A2D) fast enough to match the AD5933.
What's wrong with the 2 chips design?

>
> Can anyone give a hint to what chip to use for this or other
> guidelines? Any help is greatly appreciated.
> Best,
> John Doe

tilnews@gmail.com wrote:
> Hi
> I hope someone in this group can help me. I have a basic understanding
> of electronics and uC, but my knowledge is - to say the least - a bit
> rusty.
> A friend of mine asked me to help build a circuit, so now I ask here:-)
> 
> Wanted: I need to produce a 100kHz sine wave with an amplitude in the
> range [2 5] V. I'm not sure about the resolution, but I guess (!) it
> should be 10 or 12 bits. This is used as input to a substance via two
> electrodes, and I need to measure on this substance using two other
> electrodes and do a fft on the measurements.
> The current design uses a uC (atmel something) and a AD5933 which then
> produces the result for use in the uC. But now we are looking for a
> simpler way in which we hopefully can use just one chip. Needless to
> say it should be small and low power!
> 
> Can anyone give a hint to what chip to use for this or other
> guidelines? Any help is greatly appreciated.

I'll bet you can do with just the AD5399 and a i00Hz crystal oscillator.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

Jerry Avins wrote:
> tilnews@gmail.com wrote:
> > Hi
> > I hope someone in this group can help me. I have a basic understanding
> > of electronics and uC, but my knowledge is - to say the least - a bit
> > rusty.
> > A friend of mine asked me to help build a circuit, so now I ask here:-)
> >
> > Wanted: I need to produce a 100kHz sine wave with an amplitude in the
> > range [2 5] V. I'm not sure about the resolution, but I guess (!) it
> > should be 10 or 12 bits. This is used as input to a substance via two
> > electrodes, and I need to measure on this substance using two other
> > electrodes and do a fft on the measurements.
> > The current design uses a uC (atmel something) and a AD5933 which then
> > produces the result for use in the uC. But now we are looking for a
> > simpler way in which we hopefully can use just one chip. Needless to
> > say it should be small and low power!
> >
> > Can anyone give a hint to what chip to use for this or other
> > guidelines? Any help is greatly appreciated.
>
> I'll bet you can do with just the AD5399 and a i00Hz crystal oscillator.

Yes, a three chips solution:

AD5933, AD5399 and XC3S1000 should work.

>
> Jerry
> --
> Engineering is the art of making what you want from things you can get.
> =AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=
=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=
=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF=AF

linnix wrote:

> You won't find any uC (internal A2D) fast enough to match the AD5933.
> What's wrong with the 2 chips design?
>

The speed is not the most important issue, overall size is. It doesn't
need to be "real time" fft. It is okay, if we just send a sine for a
few ms and then use a few seconds processing the result.

tilnews@gmail.com writes:
> [...]
> Can anyone give a hint to what chip to use for this or other
> guidelines? Any help is greatly appreciated.

Hi John,

How about something like TI's MSP430F155? It has on-board 12-bit ADC and DAC,
is very low-power, and enough MIPS to do the simple DSP tasks you ask for.
-- 
%  Randy Yates                  % "Remember the good old 1980's, when 
%% Fuquay-Varina, NC            %  things were so uncomplicated?"
%%% 919-577-9882                % 'Ticket To The Moon' 
%%%% <yates@ieee.org>           % *Time*, Electric Light Orchestra
http://home.earthlink.net/~yatescr

Randy Yates wrote:
> tilnews@gmail.com writes:
> > [...]
> > Can anyone give a hint to what chip to use for this or other
> > guidelines? Any help is greatly appreciated.
>
> Hi John,
>
> How about something like TI's MSP430F155? It has on-board 12-bit ADC and DAC,
> is very low-power, and enough MIPS to do the simple DSP tasks you ask for.

That looks very interesting. I'll give that one a check.
Thanks

tilnews@gmail.com wrote:
> Randy Yates wrote:
> > tilnews@gmail.com writes:
> > > [...]
> > > Can anyone give a hint to what chip to use for this or other
> > > guidelines? Any help is greatly appreciated.
> >
> > Hi John,
> >
> > How about something like TI's MSP430F155? It has on-board 12-bit ADC and DAC,
> > is very low-power, and enough MIPS to do the simple DSP tasks you ask for.
>
> That looks very interesting. I'll give that one a check.
> Thanks

Be aware that it can probably do a few hundred KSPS at best, not MSPS.

"linnix" <me@linnix.info-for.us> writes:

> tilnews@gmail.com wrote:
>> Randy Yates wrote:
>> > tilnews@gmail.com writes:
>> > > [...]
>> > > Can anyone give a hint to what chip to use for this or other
>> > > guidelines? Any help is greatly appreciated.
>> >
>> > Hi John,
>> >
>> > How about something like TI's MSP430F155? It has on-board 12-bit ADC and DAC,
>> > is very low-power, and enough MIPS to do the simple DSP tasks you ask for.
>>
>> That looks very interesting. I'll give that one a check.
>> Thanks
>
> Be aware that it can probably do a few hundred KSPS at best, not MSPS.

It's an 8 MHz part - you can probably get 1 MSPS out of it,
if you're just outputting to a DAC and collecting input from a ADC.
Note that it has DMA, so you won't waste cycles transferring data.
-- 
%  Randy Yates                  % "How's life on earth? 
%% Fuquay-Varina, NC            %  ... What is it worth?" 
%%% 919-577-9882                % 'Mission (A World Record)', 
%%%% <yates@ieee.org>           % *A New World Record*, ELO
http://home.earthlink.net/~yatescr

linnix wrote:

> 
> tilnews@gmail.com wrote:
>> Randy Yates wrote:
>> > tilnews@gmail.com writes:
>> > > [...]
>> > > Can anyone give a hint to what chip to use for this or other
>> > > guidelines? Any help is greatly appreciated.
>> >
>> > Hi John,
>> >
>> > How about something like TI's MSP430F155? It has on-board 12-bit ADC
>> > and DAC, is very low-power, and enough MIPS to do the simple DSP tasks
>> > you ask for.
>>
>> That looks very interesting. I'll give that one a check.
>> Thanks
> 
> Be aware that it can probably do a few hundred KSPS at best, not MSPS.

I went and checked out the data sheet, do not count on more than about 250 k
samples / second.  Try a real DSP of the TMS320c5000 series.

-- 
 JosephKK
 Gegen dummheit kampfen die Gotter Selbst, vergebens.  
  --Schiller