The transistion is sometimes LED to NED (Noise emitting Diode) to DeaD. Gary
From: [email protected] [mailto:[email protected]] On Behalf Of Fred Townsend Sent: Thursday, August 17, 2006 10:56 AM To: [email protected] Cc: [email protected] Subject: Re: Noise on a video signal I have designed similar equipment and have successfully tracked down noise problems similar to what you describe. In other words, been there, done that! Before I describe how I tracked down my noise sources (note the plural) let me state I agree with John Woodgate's comments. Using a pad to help match the video to RF impedance sounds like a good idea but the idea of using the spectrum analyzer in the first place is a poor idea. My concern is digital ground noise from your ADC would be coupled to your spectrum analyzer. Spectrum analyzers are not noted for their common mode rejection. Simply you may create a situation where you see additional noise sources that are not part of your original problem. Again, been there, done that! Got the dummy of the week T-shirt. After spinning my wheels with a number of test setups I adopted a brute force approach. The data stream from the ADC was feed to a FPGA with DSP. This allowed the application of many filtering techniques while providing digital interfaces to digital data analyzer tools. Various test programs could be run by merely downloading the appropriate code to the FPGA. This allowed comparing imaging chips (installed a socket for the imagers). This also eliminated any artifacts caused by the DAC required for NSTC viewing (see John Woodgate's comments). A few days of writing test programs enabled us to characterize the entire run of imagers. The data showed the noise pixel addresses (after centroiding) varied from chip to chip and was never seen on the edge of the pixel array. A little more detective work showed the sensor's digital pixel address lines were cross talking into pixels. This information was relayed to the silicon maker (who had previously assured me their chips were perfect). They did a mask revision to move the address lines away from the cells and the noise problems were solved but not until we had burned a lot of money to prove we had bad silicon. The names of the guilty have been omitted to save their sorry asses. Remember the application of a high voltage will convert a LED (Light Emitting Diode) to a DEaD (Dark Emitting Diode). Fred Townsend DC to Light Inc. [email protected] wrote: >Good members of the regulatory Jedi Council :-) > >How about a few thought-provoking questions. > >Let's say that you have a high resolution imaging system. The system has >a focal plane array that detects IR radiation...the signals from this >array are read out as a clocked analog data stream. i.e. each pixel gives >an analog value when it is read. > >This clocked analog stream goes through an A/D converter. > >The sampling/clock rates are all set so that the eventual output of this >imager is a completely digital data stream that is sent out as NSTC video. > >The video is piped out of the imager to a display via coax cable. > >Here's the problem. The display shows what appears to be a "light >snow-like" noise. > >I'm thinking about some analytical and measurement methods that could be >used to track down the noise source. > >What would happen if I fed the video signal directly to a Spectrum >Analyzer??? I'm assuming that I would have to install a "pad" on the >spectrum analyzer input to attenuate the signal and help to impedance >match the 75 Ohm coax with the 50 Ohm spectrum analyzer input. > >It seems to me that we may be able to detect some frequency peaks that >persist even if we change the scene that the camera is looking at...maybe >these peaks would point us toward the noise source in our system? > > >What about an FFT? What would be the feasibility of grabbing a subset of >the data and performing an FFT on it? Since the noise is all over; it >doesn't seem to me that we would have to perform the FFT on all 65,000 >points of pixel data. Maybe we could use a 100x100 subset near the center >of the image? > >Of course, I've also instructed my friend who is working on this to run a >wet finger over the circuits while the video is running to see if he can >crudely locate the noise source by looking for a change in the snow on the >screen. > >Here's another question for those of you who are more familiar with video >than myself. Is there an intuitive method of homing in on the noise >freqencies by looking at how the noise appears? For instance, our noise >is a sort of granular "snow" that drifts on the screen...it doesn't stay >still. > >Thanks in advance for any tidbits that you may turn up. > >Chris Maxwell, M.Sc >Electrical Engineer >Critical Imaging LLC >[email protected] >315-732-1544 x127 >www.criticalimaging.net > >- >---------------------------------------------------------------- >This message is from the IEEE Product Safety Engineering Society >emc-pstc discussion list. Website: http://www.ieee-pses.org/ > >To post a message to the list, send your e-mail to [email protected] > >Instructions: http://listserv.ieee.org/request/user-guide.html > >List rules: http://www.ieee-pses.org/listrules.html > >For help, send mail to the list administrators: > > Scott Douglas [email protected] > Mike Cantwell [email protected] > >For policy questions, send mail to: > > Jim Bacher: [email protected] > David Heald: [email protected] > >All emc-pstc postings are archived and searchable on the web at: > > http://www.ieeecommunities.org/emc-pstc > > > - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. Website: http://www.ieee-pses.org/ To post a message to the list, send your e-mail to [email protected] Instructions: http://listserv.ieee.org/request/user-guide.html List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas [email protected] Mike Cantwell [email protected] For policy questions, send mail to: Jim Bacher: [email protected] David Heald: [email protected] All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. Website: http://www.ieee-pses.org/ To post a message to the list, send your e-mail to [email protected] Instructions: http://listserv.ieee.org/request/user-guide.html List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas [email protected] Mike Cantwell [email protected] For policy questions, send mail to: Jim Bacher: [email protected] David Heald: [email protected] All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________
