I used these a lot when they first came out and never had a problem. If someone DID have a problem, the solution would be to bypass the blocking diode.

FM1608 is proven to work just fine and like I said, if there WERE a problem, the solution would be to simply bypass the blocking diode.

I used these for probably two years before the FM16W08 came out and had no problems. I used them in adapters for 5101, 6116, and as a sub for 6264 and gave a bunch away to people at shows (I think Tim Arnold and Clay have them for sure) and sold a bunch to people. No complaints.

It was a nice way to start out making NVRAM adapters actually because the DIP parts were easy to work with. Then I switched to using FM1608 in SOIC in my adapters (again, before FM16W08 came out) because they discontinued the DIP parts. There are probably ~100 machines out there in the world with FM1608 in them in the form of these adapters. Then ultimately I switched to FM16W08 because of the wider voltage range assuming that one day there would probably come a situation where it was warranted. Then they started only making the W part, and the rest is history.

I still have ~200 FM1608. I wouldn't hesitate to use them to replace any RAM in pinball because I've already been doing it for years with no problems.

Definitely do the test, but I can confirm that this will work. This is exactly how I test my adapters (6116 and 6264).

What's the specific issue that can be caused using these? I hate to throw gas on the fire but I'm genuinely curious because I've never had any issues using these and there shouldn't really be any based on the specs for the RAM outlined in the data sheet.

Which boards have you seen show these problems? If it's a specific system it would be good information to share.

I took a look at the data sheets for FM1608 and FM16W08 because I was curious about the "write enable voltage". I didn't see specific conditions noted for the /WE pin, but the inputs for both chips are listed as 2V minimum as "input high voltage". The FM1608 specifies 2V. The FM16W08 specifies 0.7 × VDD so the minimum would be 1.89V, but that would be at 2.7V supply. At the voltages present in a pinball machine the minimum voltage for a high condition is actually HIGHER (4.5V x 0.7 = 3.15) for the FM16W08 according to the data sheet, but that's not necessarily the case. Either way, it's basically a wash. They're rated the same.

The obvious difference between them is min/max supply voltage. The FM1608 is advertised to operate down to 4.5V and the FM16W08 to 2.7V. Makes sense for their part to be able to work in 3.3V systems so the W part makes sense in that context. Other than that they're basically the same in terms of DC operating conditions.

Data sheets for 6116 and 6264 show basically the same operating conditions (4.5V-5.5V supply voltage, 2V minimum for high input). So in theory, any machine that has problems with FM1608 would also be vulnerable to the same problems with 6116 or 6264 (the RAM already in the machine).

I would expect to find a power supply problem in any game that has the RAM flaking out because of supply voltage. The FM16W08 may be rated down to 2.7V but most of the other chips on the MPU are rated 4.5V-5.5V (including the CPU). I've seen games happily running along below 4.5V, and maybe dropping in a different RAM would cause a problem in those cases, but the real problem is that the power supply is old and tired.

I built the adapters Scott is referring to in those threads. I know that there were no issues with those specific adapters because I tested them myself (in a system 11 MPU with blocking diode installed).

I'm curious if Scott posted those suggestions based on experience or simply as a matter of best practice. I remember sending him some RAMs for free years ago and a couple other boards that we were kind of working on together but I don't remember him specifically mentioning any problems. I would like to pick up on some of that stuff, actually. I wonder what he's up to these days. I do remember telling virtually everyone that I gave those original RAMs to that they should bypass the blocking diode because the voltage drop might put the RAM out of spec. When I cover my ass I like to use both hands. I would probably do the same thing if I were selling adapters with FM1608 today because a discussion like this is pretty much unavoidable otherwise with so many people jumping into the NVRAM adapter game. But like I said, I don't recall a single person reporting a problem and I never had any. I've never claimed it's not a possible issue. Just shared my personal experience without injecting bias or conjecture.

It's a weird irony that I'm kind of indirectly providing information for both sides of this argument.

I acknowledged the potential for an issue and provided a solution in my first reply. That should preclude me from team sweep-up-no-clue. As far as it being a "known issue" that I should be aware of because of a history with my adapters, I don't because it's not.