[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Cathode Films from Sheldahl
Hi Gerard,
It looks like a good plan and we should talk about it maybe next week.
With regards to the conference call I suggested earlier, I only received
a reply from Dan and you and Dan cannot make it on Monday morning.
I have just started some of this modeling but other projects are taking
priority at the moment in view of the upcoming reviews. I have the
software to perform the analysis, I think. So, send me items 3 and 4
when available.
I think, practically speaking, that we will be limited to 2 um Cu foils
and the resistance of the line won't be an issue. Obviously, this sets
the practical limit for mass reduction in the detector area of the FDC.
Regards,
Fernando
Gerard Visser wrote:
> Hi,
> Just a further remark on this, I was motivated to put together a
> quick SPICE simulation. It confirms that the total resistance of the
> longest cathode strip should optimally be kept less than about 10-20%
> of the characteristic impedance of the cathode strip as a transmission
> line. That value should be in the range 50 - 100 Ohms, surely. So for
> instance the 0.5 um of copper (sheet resistivity 34 mOhm/square, total
> 8.3 Ohms) should be ok. But that's about the limit.
> To analyze this more carefully, the following steps ought to be
> taken.
> 1. Line impedance (& velocity, right now I assume it has no
> dielectric) should be determined from E&M calculation (probably
> numerical calculation). I don't have the relevant software to do this.
> Maybe Fernando?
> 2. The preamp front end, and the interconnect capacitance, and the
> protection diode network, should all be put into the SPICE model. I
> can do these things, but of course it should be a lower priority for
> me than the ADC board, I think.
> 3. Actual chamber induced current pulse shape needs to be put in
> the SPICE model - this is trivial, I'll do it sometime soon.
> 4. Tail cancellation and shaping need to be put in (a simple model
> will suffice) so that we can see how much practical pulse amplitude is
> lost.
> If these four steps are done, we can see the effect of cathode
> strip resistance on the pulse amplitude.
> 5. Actually someone should also look at crosstalk simulations,
> this requires a much more complicated "coupled lines" E&M
> calculation... And SPICE (or other) transient simulation. I doubt
> we'll actually do all that. Perhaps at least some back-of-envelope
> estimates.
> Let's follow up on this further next week when the next hurdle is
> passed.
>
> - Gerard
>
>
> Elke-Caroline Aschenauer wrote:
>> On Tue, 25 Mar 2008, Gerard Visser wrote:
>>
>> Dear Gerard,
>>
>> funny our mails must have crossed in cyber space, but I agree with you
>> 1000% this needs a bit more time and thoughts before we can jump on some
>> new option.
>>
>> cheers elke
>>
>>
>>> Date: Tue, 25 Mar 2008 12:04:02 -0400
>>> From: Gerard Visser <gvisser@indiana.edu>
>>> To: Simon Taylor <staylor@jlab.org>
>>> Cc: Fernando J. Barbosa <barbosa@jlab.org>,
>>> Daniel S. Carman <carman@jlab.org>, Roger Flood <flood@jlab.org>,
>>> Elke-Caroline Aschenauer <elke@jlab.org>,
>>> "halld-tracking-hw@jlab.org" <halld-tracking-hw@jlab.org>
>>> Subject: Re: Cathode Films from Sheldahl
>>>
>>> Hi Simon,
>>> I think you may be overlooking something... -->
>>>
>>> resistivity of gold
>>> 2.2E-8 Ohm*m
>>>
>>> surface resistivity @ 900 Angstroms
>>> 2.2E-8 Ohm*m / 9E-8 m = 0.244 Ohm (per square)
>>>
>>> length of longest strip
>>> about 242 squares
>>>
>>> characteristic impedance
>>> who knows - not simulated - but surely on the order of 100 Ohms at
>>> the
>>> most
>>>
>>> signal loss (at the most optimistic) 1-exp(-242*0.244/100)=45%
>>>
>>> not so good, I think.
>>> and this simple analysis did not mention the noise contribution of this
>>> resistance. probably nonzero.
>>>
>>> I think this all needs to be considered more carefully. A real
>>> calculation of the cathode line impedance and attenuation and the noise
>>> contributed by it, should be done.
>>>
>>> - Gerard
>>>
>>> p.s. Yes I do agree with the concern about handling (and connecting to)
>>> any super-thin layers whether copper or gold.
>>>
>>>
>>> Simon Taylor wrote:
>>>> Hi, Fernando,
>>>>
>>>> As far as thickness goes, 900 Angstroms of Gold should be okay:
>>>> For each gold layer I get 900 x 10^-10 m x 19.3 g/cm^3 / (6.4
>>>> g/cm^2) =
>>>> 2.7 x 10^-5 X0,
>>>> compared to 1.3 x 10^-4 X0 for 2 microns of Copper. My concern for
>>>> such
>>>> a thin layer of conductor is handling: I can well imagine it could be
>>>> quite easy to cause breaks in the strips due to scratches... Also, we
>>>> would probably have to do some selective plating near where the strips
>>>> need to be connected to the outside world.
>>>>
>>>> Simon
>>>>
>>>>
>>>> Fernando J. Barbosa wrote:
>>>>> Sheldahl has a series of gold deposited materials that can be used
>>>>> for
>>>>> the cathodes. These materials are normally specified as thermal
>>>>> barriers in satellites but can be used for the FDC cathodes. Part #
>>>>> 146437 (see below) is a standard product with 900 angstroms of Au on
>>>>> 25 micrometer Polyimide and can be delivered in 48 inches wide rolls.
>>>>> The left column below is for standard products (available) and they
>>>>> range from 8 to 127 micrometer Polyimide.
>>
>> ( `,_' )+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=
>> ) `\ -
>> / '. | +
>> | `, Elke-Caroline Aschenauer =
>> \,_ `-/ -
>> ,&&&&&V Jefferson Lab +
>> ,&&&&&&&&: HALL-D 12C / F381 121-A Atlantic Avenue =
>> ,&&&&&&&&&&; Suite 8 Hampton, VA 23664 -
>> | |&&&&&&&;\ 12000 Jefferson Ave +
>> | | :_) _ Newport News, VA 23606 Tel.: 001-757-224-1216 =
>> | | ;--' | Mail: elke@jlab.org Mobil: 001-757-256-5224 -
>> '--' `-.--. | +
>> \_ | |---' Tel.: 001-757-269-5352 =
>> `-._\__/ Fax.: 001-757-269-6331 -
>>
>> +=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+
>>
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