From: muratlanne@gmail.com   
      
   "Gerry"  wrote in message news:gn6sck57vn33338v7gi392pbanm1rcnam1@4ax.com...   
      
   > On a 1990 Lumina van, I ended up running numerous grounding straps.   
      
   Power and ground tend to be ignored until one gains troubleshooting   
   experience. On some of the automated test equipment I worked on we had to   
   measure and control their currents, and in sensitive measurement circuits   
   keep the ground current down to a few milliamps. There were separate analog,   
   digital and "High Quality" measurement grounds joined by low value resistors   
   and back-to-back diodes.   
      
   Nothing on the analog side exceeded 10 KHz, measurements were given a few   
   milliseconds to settle, mostly because of capacitive dielectric absorption   
   decay in the custom Teflon-insulated reed relays and coaxial cables made by   
   WL Gore. I built an instrument to measure and display it at the level of   
   picoamps. The project engineer had been a designer at Keithley and I learned   
   a lot about precision automated measurement from him, which helped at Mitre   
   where the engineers were experts in radio but not the computerized voltage   
   measurement that digital radio introduced.   
      
   Microwave digital radio had the additional concern of the impedance of the   
   power circuit which has to be treated as parts of controlled impedance   
   transmission lines, the source and return path for signal currents. I was   
   given a problem board to troubleshoot that another department's Ph.Ds had   
   struggled with for two weeks. A glance at the circuit board inner layer   
   prints showed the problem, they had used low frequency design rules in which   
   the grounds for various circuit functions are separated into lobes and   
   converge in a small area, to break up ground loops.   
      
   What they missed was the large current spike when a high speed and power 74S   
   logic family bus carrying the A/D measurement output from the analog to the   
   digital ground regions changes state. In that nanosecond the ground planes   
   on both sides could have up to 3V momentarily between them, swamping the   
   receiving inputs, before the ground return current could make it to and back   
   from the common junction at about 6" per nanosecond, ~= c/2. The 15 minutes   
   I needed to solve the problem was mostly spend setting up a convincing scope   
   display of the "ground bounce."   
   https://resources.pcb.cadence.com/blog/what-is-ground-bounce   
      
   When I designed a microwave frequency digital radio I controlled   
   interactions in the continuous inner layer power and ground planes by where   
   each circuit was located relative to the power feed, and used multiple   
   values of bypass caps whose self-resonant frequencies didn't overlap.   
   https://resources.pcb.cadence.com/blog/2019-capacitor-self-reson   
   nt-frequency-and-signal-integrity   
   "You can also easily [measure capacitor resonance] with a vector network   
   analyzer."   
   If you have and know how to use one.   
   https://nanovna.com/   
   In college I became totally lost during the lessons on Laplace transforms   
   and the S plane. Later in night school we learned how to use rather than   
   worship complex number math.   
      
   I snatched up a Keithley Micro-Ohmmeter at a flea market. It still worked   
   fine but the 1980's HPIB data output had become obsolete, the company had   
   changed to USB. It can easily show the resistance of an inch of 10 AWG wire,   
   and of course switches, breakers and rusted ground screws. My other problem   
   detector is a thermal imager that reveals heating from IR drops. 5A through   
   10AWG wire shows up clearly, 2A is visible. Bad connections glow like light   
   bulbs.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)