A psychedelic light-show gets quirky and Glen has to figure out why.
Back in the 1970s, color organs were popular as a means for young partiers to bring psychedelic light shows (and possibly other psychedelic substances) into their home rumpus rooms. These were generally made up of four strings of miniature Christmas lights, each string the same color (red, green, blue, yellow) arranged in a pattern behind a screen and housed in a thin, square box. Four audio bandpass filters (bass, lower mid range, upper mid-range, and treble) controlled SCRs, which caused each light string to pulse to the amplitude and frequency rhythms of music from a stereo system.
A typical color organ can be seen here:
The Heath company made a color organ in the form of a build-it-yourself kit. One day a young guy came into the Heath retail store where I worked complaining that his color organ was unstable. He said that he had to keep re-adjusting the sensitivity threshold of the light strings every time he moved the unit between his home and a friend's home. This guy, apparently, liked to party.
Try as I might I could not reproduce the problem at my bench, once the trimpots were set so the lights just turned off with no input signal the settings stayed put. No intermittent parts or solder joints were causing the problem. I had to tell him that it looked like no fault with the unit and that it might be due to changes in the AC line voltage at the different locations, and that he would have to manually adjust the trimpots each time. At the time I had no means of varying the AC line voltage to test my theory.
About six months later, Heath released a new isolated variac AC power supply product, and I acquired one for my bench. About this time, the same young man called up and said we really needed to figure out his problem -- each time he had to pull the rear cover off the unit to mess with the trimpots and he was getting tired of it. (I suspected he was more likely having visual problems and was finding it difficult to get the screwdriver into the trimpot slots while seeing double.) But I told him to bring it in again, and since I now had the variac to test my varying line voltage hypothesis I was willing to give it another shot. Another nice effect of the variac was the isolation from the AC powerline that allowed me to connect scope probe grounds to the hot parts of the circuit.
When I got the unit onto the bench I saw that my hypothesis about sensitivity to variations in AC powerline was indeed correct. When aligned at 110 VAC the light strings turned on when the voltage was increased to 120 VAC. When re-aligned at 120 VAC the lights turned on at 110 VAC.
The exact details are a bit hazy in my mind; this was a few decades ago and a few of my own parties in between. But I recall that it had something to do with a charged filter capacitor at the output of the fullwave bridge rectifier, and that when the light strings were set to not turn on with no input signal, a residual charge remained that changed the SCR gate thresholds. When the line voltage changed, this charge caused the SCRs to trigger on every 2nd AC half cycle and light when they were not supposed to.
The fix was simple -- install a 10 watt resistor of a few kilohms across the bridge rectifier output to ensure the capacitor fully discharged each half cycle. Removing or reducing the capacitor caused another problem, but I can't recall what it was. Heath engineering approved the change and published it as a modification to be done when a customer complained about this specific problem.