Serious mechanical errors in CCD webcams.Posted By Larry Stange |
In the process of converting a Vesta webcam for astrophotos a slight bend in the CCD assy board was noted being caused by the microphone pressing against metal housing. Much more serious flaws were then observed after the removal of the microphone, regarding the angle of the CCD and its assy board to the focal plane. The board was not flush to the *square* cast housing surrounding the CCD because the square housing was not flat NOR level!. Without a doubt the threaded lens would have been way off the perpendicular axis of CCD. The CCD error in angle was obvious AFTER it was discovered. Some of the circuit heat could be dissipated from a flat contact of the square housing to the copper ground plane on both sides of the CCD. In short, nothing was true in the metal castings although fairly close.
I sectioned off the square metal casting surrounding the CCD and trued the bottom flat & removed the casting bulges on the top to a height accuracy of .003 inch. The CCD now seats flush to the square casting and the boards copper ground planes now make contact with the casting.
Because my CCD true'ing work had an error of at least .003 inch, I changed my housing design to allow for small corrections of the CCD alignment in the focal plane.
This was accomplished by using 3 nutcerts in a triangular fashion on the housing box(see photo.). The draw tube is tightly seated in a thick aluminium plate which has a rubber sandwich between itself and the housing box. Three adjusting screws compress the rubber adjusting for any tilt which may still be left after the CCD work thus insuring alignment.
To summarize: Everything was done to promote full contact of ALL metal parts for heatsinking. Everything was done to insure the CCD face is truley perpendicular to the focal plane. There is more than enough adjustment range in compressing the semi-hard rubber to handle any future problems.
The box is not finished by a long shot because before any electronic mod's are done I want to examine what I think may be the REAL heat problem in webcams....i.e., after a 1 hour test of running the CCD in semi-light conditions with the box open, ALL of the heat was coming from the PROCESSORS, not the CCD. This means I may cut minature heat sinks from a computer processor down to postage stamp size and epoxy them on the two processors in the CCD...IF I can find a heat conductive epoxy and add a quiet fan etc.
I know I'm on the right track because now for the first time I can obtain sharp focus on a white window frame about 1/8 mile away in BOTH the vertical and horizontal axis. Before I could focus on the vertical OR the horizontal, but could not get them both together. Photo above illustrates some of the points I referred too. -LarryS
P.S. Plumbing tube is chrome plated brass 1.252 in dia. It is the EXACT fit to my Celestron camera adapter without ANY play in the seating! Everything inside tube is flat blacked including inside the lttle square housing. :-)
I sectioned off the square metal casting surrounding the CCD and trued the bottom flat & removed the casting bulges on the top to a height accuracy of .003 inch. The CCD now seats flush to the square casting and the boards copper ground planes now make contact with the casting.
Because my CCD true'ing work had an error of at least .003 inch, I changed my housing design to allow for small corrections of the CCD alignment in the focal plane.
This was accomplished by using 3 nutcerts in a triangular fashion on the housing box(see photo.). The draw tube is tightly seated in a thick aluminium plate which has a rubber sandwich between itself and the housing box. Three adjusting screws compress the rubber adjusting for any tilt which may still be left after the CCD work thus insuring alignment.
To summarize: Everything was done to promote full contact of ALL metal parts for heatsinking. Everything was done to insure the CCD face is truley perpendicular to the focal plane. There is more than enough adjustment range in compressing the semi-hard rubber to handle any future problems.
The box is not finished by a long shot because before any electronic mod's are done I want to examine what I think may be the REAL heat problem in webcams....i.e., after a 1 hour test of running the CCD in semi-light conditions with the box open, ALL of the heat was coming from the PROCESSORS, not the CCD. This means I may cut minature heat sinks from a computer processor down to postage stamp size and epoxy them on the two processors in the CCD...IF I can find a heat conductive epoxy and add a quiet fan etc.
I know I'm on the right track because now for the first time I can obtain sharp focus on a white window frame about 1/8 mile away in BOTH the vertical and horizontal axis. Before I could focus on the vertical OR the horizontal, but could not get them both together. Photo above illustrates some of the points I referred too. -LarryS
P.S. Plumbing tube is chrome plated brass 1.252 in dia. It is the EXACT fit to my Celestron camera adapter without ANY play in the seating! Everything inside tube is flat blacked including inside the lttle square housing. :-)