MODIFICATION
59 -- SILICON DECTOR AND PROCESSING OF SILICON DETECTOR WAFERS WITH ALUMINUM METALLISATION AND PASSIVATION
- Notice Date
- 2/13/2002
- Notice Type
- Modification
- Contracting Office
- NASA/George C. Marshall Space Flight Center, Procurement Office, Marshall Space Flight Center, AL 35812
- ZIP Code
- 35812
- Solicitation Number
- 8-1-1-SS-D5366
- Response Due
- 2/22/2002
- Point of Contact
- Kenneth R. Smith, Contract Specialist, Phone (256) 544-0327, Fax (256) 544-9081, Email ken.smith@msfc.nasa.gov - Valerie N. Holmes, Contracting Officer, Phone (256)544-0314, Fax (256)544-9354, Email valerie.holmes@msfc.nasa.gov
- E-Mail Address
-
Email your questions to Kenneth R. Smith
(ken.smith@msfc.nasa.gov)
- Description
- THIS NOTICE CONSTITUTES AMENDMENT NO. [3] TO THE COMBINED SYNOPSIS/RFQ FOR [TITLE OF PROCUREMENT: Silicon Detector for the Prototype for ACCESS Silicon Matrix ]. Companies shall acknowledge all amendment(s) in their quote. This notice serves as the official amendment to subject synopsis/RFQ and a written amendment will not be issued. The purpose of this amendment is to [Provide an overview of questions and answers received from prospective vendors. Diagrams are attached for viewing].--Question (1): We have been looking into how to route the signals from the central pads in the square array of pads on these detectors to an edge where they can be connnected to pins. We began with kapton foils carrying the traces that were placed on the surface of the detector and bonded to each pad. This will work, but now we are considering to put these traces on the detector itself. The idea is to put a thick glass passivation layer on top of the detector and run the metal traces on top of it. The glass would have penetrations to the aluminum pad surfaces where the traces would connect and these traces would run to wire-bonding pads on the edges of the detector. Have you done something like this? Our signal dynamic range is large so we need a relatively thick glass layer and thin traces to avoid cross coupling the signals into the other pads crossed by the traces, but it seems possible to us. [Details for the expansion of the question. Will it be possible to change the design by adding connecting traces from middle pads to two rows of pads from bonding on two sides of detector.[Refer to attached drawing as Det_traces.pdf]: The suggested changing of the existing detector production process is:-After the thermal passivation is done to run additional metallization process for plating of traces and pads for bonding;-Final polyimide passivation should have now windows for 36 contact pads and windows for two rows of pads for bonding. If this would be possible then the questions are: 1. How thick could be the thermal passivation (1...4...10 um?) 2. How much could the thickness of traces be increased 9from standard 0.7um)? 3. What is the thickness of polyimide passivation? 4. Could the plating of traces be done above polymide after the polymide passivation? [Answer: The tracing of metal lines is possible: 1. In first metal layer: We can do it in the first metal layer: We have two options for this. - Place p+ under the metal lines. Then will the metal lines be sensitive to radiation. - Do not place p+ under the metal lines. This will add capacitance at the order of 5pF/10mikron linewidth /cm lengthof the metal line. 2. We can add a second metal layer with polyimide (thickness 5mikron) ad dielectric. We can then add a 2mikron polyimide as top passivation. This will add capacitance at the order of 0.6pF/10mikron linewidth / cm length of the metal line. this will add the price considerably as this requires 2 extra mask layers. General Comments: The metal resistance is approxiamtely 22 Ohm/cm/ 10 microns linewidth (for 1.2 micron metal thickness). Max metal thickness with standard polyimide passivation is approximately 1.8 mikron. Thicker metallisation is possible if we use thicker polyimide passivation (5 micron), but thicker metal and thicker passivation will add some cost. There will be space for metal lines with widths over 100 microns in this design, but this will be a trade off with the capacitance. The element capacitance for 1X1 cm pads should be around 34pF.] [Reply to the answer: we prefer method 2. We will have ceramic plates made with 33 pins along opposite edges. One pin on each edge will be connected to a gold plating on the ceramic plate. This will be the blocking contact connection. The detectors must be attached to the ceramic plate by gold paste (or glue) and the wire bonds made between the 32 pad connections on opposite edges of the detector with the matching pins on the ceramic plate. These wirebonds will need to be stabilized with a silicone compound.[Refer to drawings Cer_Substr.pdf and Contact Window.pdf attached]. The drawing is in a 36-pad version. Now it will be best if we provide the plates and you mount the detectors and do the wire bonds before shipping.-- [Question 2 from vendor: Just to clear something up, will the ceramic plate be continously without a hole? That is the whole detetctor backside glued. Will the pattern be thick film technology?]-- Answer to Question 2. The requirement is for a rectangular ceramic plate with 33 small holes along opposite edges. In these holes there will be pins. One pin on each side will connect to a gold layer fired on the surface of the plate. In the example we made the gold layer had the shape of an "eye" with the corners of the "eye" coming to the middle pin on each side for contact. The detector had gold on its blocking contact side. Gold paste will be used to fix the middle of the blocking contact side of the detector to the middle of the gold "eye" on the ceramic plate. We want to use a glue spot that is large enough to provide mechanical strength so that the detector will not come loose during shock and vibration, but not too large because differential thermal expansion between the ceramic plate and the silicon could break the detector. -- Question 3. -- Will the pattern be thick film technology? In the example we tested, we just painted on gold paste and fired it. This is to be a gold surface that is strongly fixed to the ceramic surface.-- Answer OK. -- Question 4. Are there any special mechanical tolerances of the assembly? Do you want Al or Au bonding?-- Answer -- The bonding pads along the edges of the detector can be either gold or aluminum, but seemingly aluminum will be less expensive since it seems to avoid another masking step.]-- Four quotes are to be forwarded on the following: (1.) 16 detectors for use with the seperate kapton foil for routing the signals. (2.) The cost per detector for producing up to four additional detectors of this type. (3.) 16 detectors with the signals routed on top of the detectors (4.) The cost per detector for producing up to four additional detectors of this type. The following diagrams are attached: The due date for receipt of offers [is] extended to [February 22, 2002]. Companies shall provide the information stated in the synopsis/RFQ posted on the NASA Acquisition Internet Service (NAIS) on November 7, 2001. Documents related to this procurement are available over the Internet and are in Microsoft Office 97 format and reside on the World Wide Web (WWW) server which may be accessed using a WWW browser application. The Internet site, or URL, for the NASA/MSFC Business Opportunities home page is http://nais.msfc.nasa.gov/cgi-bin/EPS/bizops.cgi?gr=CL&pin=62
- Web Link
-
Click here for the latest information about this notice
(http://nais.msfc.nasa.gov/cgi-bin/EPS/bizops.cgi?gr=D&pin=62#99193)
- Record
- SN00026407-W 20020215/020213213307 (fbodaily.com)
- Source
-
FedBizOpps.gov Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |