About 2 years ago I got a used 817 as a gift. It worked, with some little flaws... On Christmas '22 I decided to change the leftside rotary-encoder (often turned "backwards" or jumped wildely) with an Chinese Part from my QRP-SDX-Project. It worked, but the resolution was bad. So I "recycled" the old encoder and changed it again: perfect - for about 2 months.
Then suddenly some strange muting effects occurred, mainly in the receiving stage. It turned out, that the signal going from mainboard to frontpanel-pot and back was temporarily interrupted. Fixed it with resoldering two pins on the 50-pole Connector (front panel). Hard work with degrading eyes...
Some weeks later (Jan. '23) it got worse. Muting occured in the TX-path, FM-Volume was not anymore settable, other strange effects took place. After reopening for the 15th time, it seemed that the flat cable was broken. Ordered a new one - happy for some days. Then TX-muting reappeared with the new cable...
So I decided to do some measurements. Test-cables from underneath the mainboard to the outside (analog scope HM-205) were soldered, another 20 or 50 times I disassembled the unit and got an idea, that I²C signals going to the mainboard were bad in the state of muting-failure. This was synergy after I got in contact with DC1MC and EA7EE from this board. KA7OEI pages were very helpful to understand what's going on.
To make the long story short: after a complete resoldering of the 50-pole connector on the front-panel and reusing the original flatcable (I broke the mainboard-connector at the 51th disassembling-session and had to replace it additionally) things are o.k. now since about two weeks! No EEPROM-change, no data-loss, no MCU-horror.
0.5 mm 50-pole was a challenge, but a specialist-soldering-guy in my surrounding did a perfect job. With my eyes I cannot do this anymore...
So I have a (momentaryly) perfect working FT-817ND again and want to share this experience with you in DC1MCs thread.
Thanks for your patience and vy 73 DK1TCP