While I have had this display design working for some time, I have had an insidious, subtle and consistent problem. While the 2 meters I use as testbeds (8840A & 8842A) are fully functional and pass all self tests with both a VFD and the newer blue LED replacement display, I would get test failures with the OLED design. Specifically tests 4, 6, 7, 15 & 22(8842A). These typically point to the A/D which I know is fine!
![Banging Head |O](https://www.eevblog.com/forum/Smileys/default/bangheadonwall.gif.pagespeed.ce.8iLbFYV4Bc.gif)
I at first suspected my level converter design was at fault for possibly overloading the +30V supply. But I found the errors remained after completely removing the level converters (no connection to the mainboard +30V grid and segment signals! They remained after removing the +5V from the OLED controller PCB. The only connection to the mainboard was the pass-through connections to operate the PB switches and the controller PCB ground to the "digital ground" by the power transformer.
Ok, maybe there is something obvious here, but it took some time for me to realize that there has to be a un-expected short between the analog ground and digital ground. When I removed the OLED controller ground from the mainboard the errors disappeared! But where was the short?
I went back over my design and the schematics of the Flukes looking carefully at the logic control and the VFD board wiring. On all of the schematic versions I have of the VFD, pin 16 of J203 on the display has a 7 in the grounding triangle. That is digital ground. But a careful look at the digital main PCA schematic shows J203 pin 16 with a 2 in the triangle! That is analog ground. A quick ohm meter check confirmed this.
What I had done is follow the VFD/PB schematic and on the pass-through wiring for J203 from the main PCA to my OLED PCB to the VFD/PB PCB I connected pin 16 to digital ground everywhere. Voila, analog to digital ground loop! I simply lifted the pin 16 circuit completely, since the VGD PCB ground plane is rather un-necessary now I think, and all tests pass!
The moral of this story is 1) these old hand-written schematics are not always correct, 2) analog and digital grounding issues can be subtle. The impedance and resistance between them may be small, but the currents can play havoc with such a sensitive meter.
kjo - KO3Y