Quoted from Pinballflare:Yes, I measured pin 22 of z22 with the power off and it checked out. I think the more telling issue though is that the high score is not retained game to game. If you play and get 95000 the game says high score to date is 95000. The very next game without shutting off the machine you score 30000, it now says high score to date is 30000.
It sounds like the CPU cannot either WRITE to the 5101 or is getting all zeroes back when it reads it. It only bothers to do a write when somethings has changed. So, if you finish a game, it has to pull the previous high score (which might be coming back as 000,000), compare it to your score, and then only save it if its larger in value. Then repeat the process at the end of the next game. Drop a coin, it has to update the bookkeeping value stored. In other words, the NVRAM itself is not being used all the time like on other Manufacturer's systems. The Sys 1 CPU uses its internal memory most of the time and then only does a save or retrieve when necessary (I found this out while working on my Split Data bus 5101 NVRAM module. You might get different results if you try to read the bookkeeping values or enter a new free game threshold, and then exit that with the slam switch (which doesn't remove power from the ICs) as opposed to powering down and back up again.
So, you would have to monitor control lines from the spider chip(s) and then see if they are getting through the various gates to end up at the 5101 pins themselves. A bit harder to probe since the inputs and outputs of the Spider chips are at +5V and -12VDC swings.