Quoted from yonkiman:I'm an electrical engineer. Purely out of heat and efficiency reasons (and because I could) I replaced the LM323 linear 5V regulator in my HS2 with a switching regulator. Ideally we'd all do that to save power and reduce heat and (if done correctly) eliminate this reset problem. But it's not a simple fix, it requires removing the driver board, desoldering/soldering, etc. So it's not for everyone.
Most resets are connector based. Few are bridges/caps. Many of the unsolvable ones I have found, where 5v has not improved from all the usual fixes, or the 5v is good but still resets, have been attributed to the 5v regulator. In my opinion, this is the single most overlooked failing part in WPC's. All of the issues a couple years back where I had troubles like the above, that didn't fall under the usual reset suspects, I swapped the regulator, and this eventually fixed it.
I read eventually they are something like a 10% +/- in either direction for a true 5v. This means you could have a new one, brand new, new part of NOS, and it could potentially be one on the low end, and not capable of providing enough 5v for a WPC to run. I question that Williams must have known this and tested these prior to installing, as I have not read of a lot of resets for NIB b/w machines.
Early on when repairing my TZ, I didn't know this. Swapped a LM323 in, still low 5, assumed it was something else. It was only when I found out about these switching regulators, and how they output exactly 5v, did I readdress some of these oddball scenarios, which has fixed each issue.
At this point of the game, when a reset falls into the category of not a connector drop and the test point 5v is low, I immediately swap the 5v regulator, takes no time, and it has done the trick each time.
I of course use the Kahr board in the interim, but not long term. As I have said from day one, great tool, but I personally do not feel comfortable with it for a permanent fix, or should I better say, I don't feel comfortable knowing there is an unresolved issue. I had raised the point that the board may be a good replacement as a diverter from the overused 5v, but that didn't seem to go over well with many. I still wonder about that.