Re: question on offset DAC's for the ADC125

["Mark M. Ito" <marki@jlab.org>]

I think it would indeed be nice to have control of the pedestal values 
on a channel-by-channel basis.

Fernando J. Barbosa wrote:
> Hi Gerard,
>
> It is a reasonable cost and I recommend that it should be implemented 
> . I believe that the  impact on power should also be minimal as there 
> are quite a few devices on the market that have multiple channels per 
> chip, low power and low cost.
>
> Thanks,
> Fernando
>
>
>
>
> Gerard Visser wrote:
>> Hi Fernando, Elke,
>>     I have a question concerning the desired specification for offset 
>> DAC's for the ADC125.
>>     As you know the fADC250 has an individual offset control for each 
>> channel.
>>     On the other hand, for the ADC125, the existing specification 
>> GlueX-doc-855-v2 has an offset DAC in groups of 12 channels. [The way 
>> this would be implemented, the 6 such groups of 12 channels will be 
>> the even channels and the odd channels of each of the three input 
>> connectors. That's because the even channels are handled on the main 
>> board, the odd channels on the mezzanine board; we'd have 3 DAC's per 
>> board.]
>>     However, as I am getting this portion of the design settled in 
>> detail, I'm finding that it would not be so difficult to provide an 
>> individual offset DAC for each channel at reasonable cost. I have 
>> identified several candidate parts that could provide this function, 
>> while holding to the other specifications (no detrimental effect on 
>> noise level, +/- 55% offset range).
>>     The added cost of the individual offset scheme versus the 
>> original 6-blocks-of-12 scheme is about $83 per module. So the total 
>> cost increase for the production run of 200 modules is $16,600.
>>     I would recommend that we do this, it will be a more flexible 
>> solution and it could increase the available dynamic range of the ADC 
>> in application by a few percent, in case the GAS-1 (or other input 
>> source) pedestals are not perfectly matched.
>>     A decision should be taken within next few weeks, please. Thanks,
>>
>>     Gerard
begin:vcard
fn:Mark M. Ito
n:Ito;Mark
org:Jefferson Lab
adr:Suite 8;;12000 Jefferson Avenue;Newport News;VA;23601;USA
email;internet:marki@jlab.org
title:Staff Scientist
tel;work:757-269-5295
tel;fax:757-269-6331
tel;pager:757-584-5295
url:http://www.jlab.org/~marki
version:2.1
end:vcard