There are threads on this in the archives.

When I've had an intermittent OD failure it's usually been a failing OD switch or clogged ports. Others here have found it to be the relay but that's never been my problem.
--
240s: 2 drivers and some parts cars

re: "...My car's switch has a mind of its' own. It will be on or off at the startup and will sometime respond or not respond to my pressing of the switch. Sometimes it will engage itself without my pressing the switch at all...."

I had that problem on one of my '93s. The switch would sometimes work, and sometimes not (replacing the switch didn't help). And it sometimes changed over by itself.

Not believing in poltergeists, I tried some things, and found that, when the switch wouldn't work, if I put it in second gear and tried the switch (just to see whether the light went on and off), it did! So after I then lifted the shift cover to see why the shifter position influenced the problem, I discovered that the wires (a pair of very, very thin white wires) had been worn through their insulation, and was intermittantly shorting out.

I spliced in sections of more wire, to lengthen them (caution, as it takes very delicate soldering, and insulating the soldered ends, for these thin wires), and somewhat repositioned them so they wouldn't wear again the metal parts around them. This fixed the problem.

Good luck.

Sounds like the OD relay (white, behind the center dash vents I think) may be on it's way out. OD operation involves 4 components and the wiring that connects them...

1) Fuse 11 (200) or 12 (700/900) supplies voltage to items #2, #3, and #4.

2) The Shifter button/switch turns the OD relay off and on

3) The OD relay controls the OD solenoid (on tranny) and the Upshift ("OD off") Arrow light (separate wires). When the Arrow light is ON, the OD relay and solenoid are OFF.

4) The OD Solenoid allows OD engagement when energized. (this is the "default" OD relay/solenoid state at start-up)

If your shifter switch makes the Arrow light go on and off, then the OD relay is working, although it could still be failing to control the solenoid voltage due to burnt contact points).


--
Bruce Young, '93 940-NA (current), 240s (one V8), 140s, 122s, since '63.