Resistor (SMD): Value unintelligible / not decodable

[Eheran]

I have some washing machine motor driver boards (Fisher & Paykel, 421389NAP and others like those) here. They all have rather obscure issues. I just found out that resistor values are all over the place, for example 200 kOhm resistors that are extremely consistently at 160 kOhm over >10 samples (100 % of tested) over 2 years of production. Same with other resistors (80 Ohm instead of 330, 80 kOhm instead of 100 kOhm). I can read the markings and will thus replace them, hoping that it will fix the obscure issues.

However, one resistor is unintelligible across all of the boards. I have 2 pictures of 2 different boards attached, one from 2005 and one from 2007. Take a look and have a guess. When measured out of circuit, one was reading something with a 2 at the front, I think it was ~2 kOhm. The 4 and 3 are very clearly readable. But the 3. digit is impossible to make out. An 8, which seems like the most likely number, makes zero sense, since that would be 4.3 GOhm. Has anyone every seen something like this? What could that be? Is it an R, so 43 Ohm? Other resistors on the boards use 0 instead of R to denote such values (at least above 10 Ohm), so that would seem to be inconsistent.

Also, any ideas how they can drift to such a degree? Or is that from production? Seeing such terrible silk screen makes it seem like a super low quality. However, all the caps on there are nichicon... so... jeah.

[langwadt]

https://www.hobby-hour.com/electronics/eia96-smd-resistors.php ?

43B   = 2.74kΩ

[Swainster]

And 158K decodes to 20D - not impossible to read as 204 if the character is not well formed and you are expecting a number instead of a letter. Not sure if mistaking 88X for 331 is quite as plausible though, and no idea how an EIA code for 80k would look like 104 

One thing to note is that it's not uncommon to see EIA codes and conventional codes mixed on the same board.

[PwrElectronics]

If these were drifting higher in value instead of lower, I would suspect sulfur corrosion.

At my company, we've had a lot of field issues in the past with that problem and now restrict supply to only certain brands that are more resistant.  The corrosion eats away at the end terminals so the resistor tends to slowly go open eventually.

[tautech]

https://www.hobby-hour.com/electronics/eia96-smd-resistors.php ?

43B   = 2.74kΩ

Yep, and that's the thing, not all know there are at least 2 common marking methods for SMD resistors.
Often it depends on what components can be sourced for production or if special values are required.

[Eheran]

https://www.hobby-hour.com/electronics/eia96-smd-resistors.php ?

43B   = 2.74kΩ

Wow. Thank you! I have never heard of the EIA system. Seems to be correct then (or at least close). This could also explain why there are 4 number markings, to avoid such ambiguity.

The other resistors are marked clear as day, "2003" or "1003" or "331" (clearly not 88X or something). However, I did not test those out of circuit. So it might all just be a fluke due to the expected 4-something-Ohm (no EIA knowledge) compared to the 2-something-Ohm (actual value it should have) that I checked out of circuit. In either case, I will order 100 replacement resistors each, does not cost much, and replace everything in a test board (and also measure out of circuit then).

Edit: The other resistors are indeed all okay, no drift.