Re: Timing Info

["Fernando J. Barbosa" <barbosa@jlab.org>]

Hi Matt,

My understanding from reading the note is that the peak sample, which is 
critical to do the fitting of the leading edge of a pulse, is obtained 
prior to "degrading" the data for 8 or 10-bit and 250 MSPS - a priori 
knowledge of the peak is required. The remainder of the technique is 
very nice, though.

For random pulses, we agree that the sampling of the actual peak may be 
off by as much as 4 ns. I contend that it will be hard to get the timing 
resolution you need. There is also the question of how to implement any 
algorithm on-board. LUTs are compact (=fast). One of the fADC250's 
on-board FPGAs includes a Power PC core which we don't use but it may be 
slow for any kind of computationally intensive algorithm.

To me, a proof of principle would require realistically employing any 
timing technique that satisfies the timing specifications, i.e., in a 
single channel demonstration. Maybe this is just a matter of semantics.

 I really appreciate that you will be working on this issue.

Regards,
Fernando




Matthew Shepherd wrote:
>
> Hi Fernando,
>
> I should first say that I only read the note myself and did not do the 
> study, but I think I can address your point:
>
> On Jan 25, 2008, at 1:57 PM, Fernando J. Barbosa wrote:
>
>> This is a very interesting note but it does not serve as "proof of 
>> principle" that this technique will work with a real fADC, in my 
>> opinion. I think, in this context, there is a flaw in the application 
>> of the technique. The technique relies primarily on having a sample 
>> (peak) that represents the actual peak value of the input signal and 
>> then using two samples to determine the crossing point. Obtaining a 
>> sample at the actual peak of a signal is not guaranteed by a sampling 
>> ADC. Please note that no ADC was employed at the time this note was 
>> written.
>
> I don't think the technique requires having a sample at the peak, but 
> rather just a peak sample.  Every fADC digitized buffer of a 
> reasonably-shaped PMT pulse will have a peak.  Sure, it won't happen 
> right in time with the actual analog peak necessarily, but there will 
> be a peak.  True the study does not use a real fADC, but it does use 
> real phototube pulses.  I trust relatively well a computer's ability 
> to sample at 4 ns and digitize to n-bits.  A fit to the high 
> resolution digitized pulse is used to determine the true 50% crossing 
> point, this is compared with what  is obtained after the sample is 
> digitized and processed by the algorithm.
>
>> As we discussed before, I suggest you perform the tests with 8-bit 
>> ADC from Paul and compare the results to the note. This is easier 
>> than getting the fADC, CODA....
>
> Of course we plan to do this, but I still don't see any obvious flaw 
> in this note as a proof of principle.  Thanks for the other reference 
> -- I'll take a look at it.
>
> Cheers,
>
> -Matt
>
begin:vcard
fn:Fernando J. Barbosa
n:Barbosa;Fernando J.
org:Jefferson Lab
adr:Suite #10, 12B3;;12000 Jefferson Ave.;Newport News;VA;23606;USA
tel;work:757-269-7433
version:2.1
end:vcard