[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: CDC Electronics Assembly
Hi Yves,
I think I am missing which PCB you are referring to. The connector
assembly is attached to the Plexiglass from the straw side with screws
attached from the outside and pressing the connector on the plenum side
of the Plexiglass (the threaded inserts will be an integral part of the
connector). The screws also have a washer and o-ring seal. The CDC can
then be rotated vertically and the sealant applied around the connectors
and between the connectors and the Plexiglass. It should make a very
nice seal. The conformal seal on the connector assembly is to prevent
leaks through the electrical pins on the connector.
Regards,
Fernando
Yves Van Haarlem wrote:
> Hi Fernando,
>
> Thanks for the drawing. I was actually thinking of sealing the pcb in
> the endplate which would be much easier to seal than the connector.
>
>
> Cheers,
> Yves
>
> /--------------------------------
> / Yves Van Haarlem
> _--~~--_ /
> /~/_| |_\~\ / Carnegie Mellon University
> |____________| / Department of Physics
> |[][][][][][]|:= / Wean Hall room 8404
> __| __ |__ \ Pittsburgh, PA 15213
> | ||. | == | | \ USA
> (| ||__| == | |) \
> | |[] [] == | | \ Tel.: +1 412 268-6949
> | |____________| | \ +1 412 641-9252
> /__\ /__\ \ Fax.: +1 412 681-0648
> ~~ ~~ \-----------------------------------
>
> On Wed, 13 Aug 2008, Fernando J. Barbosa wrote:
>
>> Hi Yves,
>>
>> The shield is at HV (1800 V) and the inner conductors are at the same
>> HV but with the series resistors; the signals are decoupled from the
>> inner conductors with HV caps to the preamps. In the drawing, pins 13
>> and A13 are labeled HVshield. So, the potential between the shield
>> and the inner conductor is about zero.
>>
>> I attaching the new PCB option with the updated drawing. It seems we
>> have email problems.
>>
>> Regards,
>> Fernando
>>
>> Yves Van Haarlem wrote:
>>> Hi Fernando,
>>>
>>> I see, this brings me to the RG-174 cable - so we plan to put the
>>> shield to ground and the signal part to 1800V - the cable is not
>>> being spect to do that, right. Now we leave the shielding to float,
>>> before we put it to 1800~V as well.
>>>
>>> If in the final design we leave the ground of the cables floating
>>> then we do not have a problem with the PCB feedthru. If we plan to
>>> connect also the cable grounds I believe we can make enough real
>>> estate on that feed-through pcb to solder the signal part on one
>>> side and the cable grounds on the other side of that pcb and have
>>> one more solder pad for the grounding to the endplate. Asuming that
>>> the feedthru pcb is as broad as the new EDAC connector is long. The
>>> old EDAC connector is half as long and has 13 solder leads per side,
>>> that times 2 gives us 26 solder leads on both sides of the feed
>>> through pcb.
>>>
>>>
>>> Cheers,
>>> Yves
>>>
>>>
>>>
>>> /--------------------------------
>>> / Yves Van Haarlem
>>> _--~~--_ /
>>> /~/_| |_\~\ / Carnegie Mellon University
>>> |____________| / Department of Physics
>>> |[][][][][][]|:= / Wean Hall room 8404
>>> __| __ |__ \ Pittsburgh, PA 15213
>>> | ||. | == | | \ USA
>>> (| ||__| == | |) \
>>> | |[] [] == | | \ Tel.: +1 412 268-6949
>>> | |____________| | \ +1 412 641-9252
>>> /__\ /__\ \ Fax.: +1 412 681-0648
>>> ~~ ~~ \-----------------------------------
>>>
>>> On Wed, 13 Aug 2008, Fernando J. Barbosa wrote:
>>>
>>>> Hi Yves,
>>>>
>>>>
>>>> Yves Van Haarlem wrote:
>>>>> Hi Fernando,
>>>>>
>>>>> Thanks a lot for the drawings. Some questions:
>>>>>
>>>>> What is meant by the Conformal Seal? Does it refer to make the
>>>>> connector gas tight with glue? The problem with those new
>>>>> connectors is that the space between solder leads and connector
>>>>> are larger compared to the previous (smaller) one which makes that
>>>>> much more difficult. We even do not know yet if it is possible
>>>>> with the smaller connectors.
>>>> There are adhesives with different viscosities for this purpose but
>>>> we need to try them. I will test with the connectors you are
>>>> sending me. The choice of the adhesive is very important in terms
>>>> of its chemistry so that we don't end up with polymerization
>>>> (aging) inside the straws.
>>>>>
>>>>> Going back to the PCB-feedthrough design. The argument was that
>>>>> when you solder the wires (coming from the straws) to it, they
>>>>> would make a 90 degree angle with that PCB. Could we not make
>>>>> solder pads on that PCB (like you have for SM components) so that
>>>>> the wires can be soldered at 0 degrees. Those PCBs with wires can
>>>>> be prepared before detector assembly and we can use the new
>>>>> connector on the other side of that PCB.
>>>> Yes, this is a possibility. The issue with this approach is that we
>>>> need to solder all over the shield (300 degrees?), not just at the
>>>> shield/PCB interface. If we don't, the wires on the shield will be
>>>> all over the place. However, we could just solder at the shield/PCB
>>>> interface and then use a conformal coat of adhesive. We need to
>>>> check this. Maybe two small PCBs, one for each row on a connector?
>>>>
>>>> Regards,
>>>> Fernando
>>>>
>>>>>
>>>>> Cheers,
>>>>> Yves
>>>>>
>>>>>
>>>>>
>>>>> /--------------------------------
>>>>> / Yves Van Haarlem
>>>>> _--~~--_ /
>>>>> /~/_| |_\~\ / Carnegie Mellon University
>>>>> |____________| / Department of Physics
>>>>> |[][][][][][]|:= / Wean Hall room 8404
>>>>> __| __ |__ \ Pittsburgh, PA 15213
>>>>> | ||. | == | | \ USA
>>>>> (| ||__| == | |) \
>>>>> | |[] [] == | | \ Tel.: +1 412 268-6949
>>>>> | |____________| | \ +1 412 641-9252
>>>>> /__\ /__\ \ Fax.: +1 412 681-0648
>>>>> ~~ ~~ \-----------------------------------
>>>>>
>>>>> On Tue, 12 Aug 2008, Fernando J. Barbosa wrote:
>>>>>
>>>>>> Hello everyone,
>>>>>>
>>>>>> Following discussions with Yves and Slava, I have attached two
>>>>>> pdf drawings describing the assembly process. Note that these
>>>>>> drawings are not to scale.
>>>>>>
>>>>>> Please provide feedback. Board design will proceed with the new
>>>>>> EDAC connector.
>>>>>>
>>>>>> Regards,
>>>>>> Fernando
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Fernando J. Barbosa wrote:
>>>>>>> Hello everyone,
>>>>>>>
>>>>>>> I have attached a pdf describing the edge card connector from
>>>>>>> EDAC. This is similar to the one Gerard suggested but in 0.156"
>>>>>>> contact spacing. The connector part number is 307-030-500-202,
>>>>>>> which is a 30 contacts on dual row with a HV rating of 1800 V AC
>>>>>>> rms, Thermoplastic Polyester insulator and $3.05. I searched
>>>>>>> for similar connectors and Sullins also has a similar one rated
>>>>>>> at 1800 V DC but with a different insulator (PPS Ryton).
>>>>>>> However, the cost for the Sullins part is five times higher at
>>>>>>> ~$15. The EDAC connector is stocked at Newark.
>>>>>>>
>>>>>>> Now for the assembly details which are based on Slava's
>>>>>>> drawings. Slava, I would appreciate if you could email your
>>>>>>> drawings to those on this distribution list.
>>>>>>>
>>>>>>> The assembly attempts to minimize the cable length from the
>>>>>>> straw to the connector:
>>>>>>>
>>>>>>> 1. 24 RG-174 cables are prepared as follows:
>>>>>>> a. Strip both ends so that there is the center conductor for
>>>>>>> soldering, insulation of center conductor, braid and outer
>>>>>>> insulation. Each is about .125 long so there are steps
>>>>>>> (different diameters).
>>>>>>> b. On the straw side, prepare as is now planned with solder ball
>>>>>>> and conductive rubber. Maybe a heat shrink piece could be
>>>>>>> applied to secure the conductive rubber cylinder and cover the
>>>>>>> shield in one step.
>>>>>>> c. On the connector side, slide each of the 24 cables into a
>>>>>>> small copper plate (perhaps a bit less than .125 thick) with 24
>>>>>>> holes and solder the shield braids at the exit of the holes. The
>>>>>>> shields will all be connected together to the shield HV and to 2
>>>>>>> connector pins. Solder each of the 24 center conductors to the
>>>>>>> connector.
>>>>>>>
>>>>>>> 2. The connector has 24 pins connected as in 1.c above, the next
>>>>>>> 2 adjacent pins connect to the copper plate to the shield HV, as
>>>>>>> in 1.c above. The next two pins are left unconnected and
>>>>>>> insulated or cut. the final 2 pins are soldered to a cable to
>>>>>>> attach to the endplate for grounding (ground on card is
>>>>>>> connected to endplate in this way.
>>>>>>>
>>>>>>> 3. The back side (soldered connections) is sealed with the
>>>>>>> proper viscosity sealer and with low out-gassing characteristics
>>>>>>> (no silicone here). The finished part forms the connector assembly.
>>>>>>>
>>>>>>> 4. Connect all the cables from all the cable assemblies to the
>>>>>>> straws and endplate (grounds).
>>>>>>>
>>>>>>> 5. Attach dummy boards with straps to the edge card connectors
>>>>>>> on each of the cable assemblies.
>>>>>>>
>>>>>>> 6. Position the gas plenum Plexiglass cover and feed all the
>>>>>>> straps through their respective holes on the Plexiglass cover.
>>>>>>> Slide this cover while pulling or straightening the straps
>>>>>>> until the cover touches the connectors. Attach the connectors
>>>>>>> with screws, washer and neoprene seal washer.
>>>>>>>
>>>>>>> 7. Rotate CDC so that connector side is at top.
>>>>>>>
>>>>>>> 8. Apply sealer (see 3 above) around each of the
>>>>>>> connectors/plexiglass holes.
>>>>>>>
>>>>>>> In this way, we only need a short distance between the straws
>>>>>>> and the gas plenum cover. This is not very elaborate but I can
>>>>>>> provide a drawing if necessary to better explain this technique.
>>>>>>> Please note that if one of the wires in the straws brakes and
>>>>>>> shorts tripping the HV, all the 24 channels in the group will be
>>>>>>> off. This is required because the shields are also at HV and the
>>>>>>> RG-174 is not rated for this level of HV. So, this is inherently
>>>>>>> safe.
>>>>>>>
>>>>>>> There will also be an aluminum plate or aluminized mylar cover
>>>>>>> over the whole gas plenum and grounded for shielding.
>>>>>>>
>>>>>>> Please provide feedback so we can finish the design of the HVB
>>>>>>> PCB. The HVB PCB will have the HV distribution network, the
>>>>>>> connections to the preamp card connector, grounding, etc. The
>>>>>>> preamp card, as in the FDC, is removable from the HVB.
>>>>>>>
>>>>>>> Regards,
>>>>>>> Fernando
>>>>>>
>>>>
>>
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