[time-nuts] Using CPLD/FPGA or similar for frequency

lincoln lincoln at ampmonkeys.com
Sat Jun 13 03:44:31 UTC 2015


Hello
 It also depends  on which device primitives you can use.  Xilinx spartan series has an SRL16, 16 bit shift register that can be ganged to form dividers / pre scalers. It only takes up one lut or slice, I forget which.

Link

On Jun 11, 2015, at 4:11 AM, Bob Camp <kb8tq at n1k.org> wrote:

> Hi
> 
> Depending on which chip you are using and how big it is, you can get into the 150 to
> 500 ps range running a carry chain as a TDC.That’s without getting into things like
> hand routing and temperature / voltage issues. 
> 
> How big a chip you need will be a function of how high you can get the internal PLL
> to run while packing a bunch of stuff in the chip. If you can hit 400 MHz, each carry 
> chain will need to handle a bit more than 2.5 ns, but probably less than 5 ns. You 
> can do that with a carry chain a few hundred bits long. 
> 
> There is a bit of handwaving already so this is indeed a guess rather than a design. 
> If you run 320 bit chains and 8 inputs, you will need 2.5K registers for the carry chains. 
> You also will need about 200 registers for the support of each chain, so that adds another
> 1.6K registers. Something in the 5K register range is a possible way to go for 8 inputs.
> 
> Bob
> 
>> On Jun 11, 2015, at 2:04 AM, Gregory Maxwell <gmaxwell at gmail.com> wrote:
>> 
>> On Tue, Jun 9, 2015 at 11:40 AM, Alan Ambrose <alan.ambrose at anagram.net> wrote:
>>> How about a 1pS resolution TIC? :)
>>> 
>>> Or a >12 digit frequency counter? :) :)
>>> 
>>> It's not a proper time-nut project unless there's a nutty element...
>> 
>> Well, how complex? Front end with a fast ADC and make a DSP DMTD device?
>> 
>> In terms of simpler things that (AFAIK) one can't go out and buy:  a
>> TIC with 4 or 8 inputs would be an interesting piece of time nut
>> gear.even if it was 'just' 1ns resolution
>> 
>> Surplus lab TICs are easily had but become quite a pile of equipment
>> when you want to concurrently measure a half dozen oscillators.
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.




More information about the Time-nuts_lists.febo.com mailing list