Jim

I second Jack's concern.  I'd pretty much written off the possiblilty of 
doing DSP feedback at anything but very low frequency.  I'd calculate that 
an absolute maximum for the DSP delay (assuming 37Hz gain crossover) would 
be about 13ms.  7ms or a bit more would probably be quite tolerable.  Is 
there any way to do that?

Brett

At 01:48 PM 4/13/2006 -0400, you wrote:
>Jim,
>Have you done any analysis to see if the data rate would be
>sufficient to close your loop at high frequencies?  One thing I have
>wondered about is whether the delay produced by these high-resolution
>A/Ds is longer than is implied by the sample rate, i.e. is there a
>pipeline of data coming through and your output is a few samples
>behind the real world.  This could cause stability problems.
>
>Actually, it might be worth someone calling up Analog Devices,
>a friendly applications engineer might have some useful information.
>
>I can only speak for myself, but I'm interested in hearing developments
>in this area.
>
>Jack



__________________________________________________________

Public Seismic Network Mailing List (PSN-L)